{
  "id": "isomorphism-theorems",
  "title": "Isomorphism Theorems",
  "lang": "en",
  "source_lang": "en",
  "summary": "The isomorphism theorems describe how homomorphisms, kernels, substructures, and quotients determine equivalent algebraic structures in groups, rings, modules, and vector spaces.",
  "category": "mathematics",
  "subcategory": "algebra",
  "type": "concept",
  "aliases": [
    "Noether isomorphism theorems",
    "homomorphism theorems"
  ],
  "generated_at": "2026-11-05",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/isomorphism-theorems/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/isomorphism-theorems/",
      "json": "https://aiwiki.page/zh/isomorphism-theorems.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "The isomorphism theorems are fundamental results in abstract algebra that describe the relationships between structure-preserving maps, their kernels and images, and quotient structures. Their central assertion is that a homomorphism becomes an isomorphism onto its image after its kernel has been factored out. Related theorems explain how substructures interact with quotients and how successive quotients can be combined. The results have parallel forms for groups, rings, modules, and vector spaces."
    },
    {
      "heading": "Overview and terminology",
      "text": "The theorems are commonly organized as follows:\n\n- First isomorphism theorem: the quotient of a structure by the kernel of a homomorphism is isomorphic to the homomorphism’s image.\n- Second isomorphism theorem: a quotient involving a substructure and another substructure can be identified with a quotient by their intersection.\n- Third isomorphism theorem: taking a quotient of a quotient is equivalent to taking a suitable quotient of the original structure.\n- Correspondence theorem: substructures of a quotient correspond to substructures of the original structure containing the object being factored out.\n\nNumbering varies between textbooks. The correspondence theorem is sometimes called the fourth isomorphism theorem, and the second and third theorems are occasionally numbered differently.\n\nThese results concern explicit, naturally determined isomorphisms, not merely equality of sizes. Each theorem supplies a map that explains why two differently constructed objects carry the same algebraic structure."
    },
    {
      "heading": "Group-theoretic formulation",
      "text": "In group theory, the relevant substructures are subgroups, and quotients are formed using normal subgroups. Write \\(H\\leq G\\) when \\(H\\) is a subgroup of \\(G\\), and \\(N\\trianglelefteq G\\) when \\(N\\) is a normal subgroup.\n\nFirst isomorphism theorem\n\nLet \\(f:G\\to H\\) be a group homomorphism. Its kernel and image are\n\n\\[\\ker f=\\{g\\in G:f(g)=e_H\\}, \\qquad \\operatorname{im}f=\\{f(g):g\\in G\\}.\\]\n\nThe kernel is normal in \\(G\\), the image is a subgroup of \\(H\\), and\n\n\\[G/\\ker f\\cong\\operatorname{im}f.\\]\n\nThe isomorphism is\n\n\\[\\overline f:G/\\ker f\\longrightarrow\\operatorname{im}f, \\qquad g\\ker f\\longmapsto f(g).\\]\n\nThus every group homomorphism factors as\n\n\\[G\\xrightarrow{q}G/\\ker f \\xrightarrow{\\overline f}\\operatorname{im}f \\xrightarrow{\\iota}H,\\]\n\nwhere \\(q\\) is the quotient projection, \\(\\overline f\\) is an isomorphism, and \\(\\iota\\) is inclusion.\n\nIn particular:\n\n- If \\(f\\) is a surjective function, then \\(G/\\ker f\\cong H\\).\n- The homomorphism \\(f\\) is an injective function exactly when \\(\\ker f=\\{e_G\\}\\).\n\nProof. Two elements have the same image precisely when\n\n\\[f(g)=f(g') \\quad\\Longleftrightarrow\\quad g^{-1}g'\\in\\ker f.\\]\n\nThis is exactly the condition that they belong to the same coset of \\(\\ker f\\). Consequently, \\(\\overline f\\) is well-defined and injective. It is surjective onto \\(\\operatorname{im}f\\) by definition, and it preserves multiplication because \\(f\\) does.\n\nThe theorem says that the fibers of a homomorphism are precisely the cosets of its kernel. Passing to the quotient group identifies exactly those elements that the original map cannot distinguish.\n\nSecond isomorphism theorem\n\nLet \\(H\\leq G\\) and \\(N\\trianglelefteq G\\). Define\n\n\\[HN=\\{hn:h\\in H,\\ n\\in N\\}.\\]\n\nThen \\(HN\\) is a subgroup of \\(G\\), \\(N\\trianglelefteq HN\\), \\(H\\cap N\\trianglelefteq H\\), and\n\n\\[H/(H\\cap N)\\cong HN/N.\\]\n\nThe isomorphism is\n\n\\[h(H\\cap N)\\longmapsto hN.\\]\n\nTo prove it, restrict the quotient projection \\(G\\to G/N\\) to \\(H\\). This restricted homomorphism has kernel \\(H\\cap N\\). Its image is \\(HN/N\\), since every coset \\(hnN\\) equals \\(hN\\). The first theorem therefore gives the result.\n\nA common more general formulation assumes only that \\(H\\) normalizes \\(N\\), meaning \\(hNh^{-1}=N\\) for every \\(h\\in H\\). The same conclusion follows. Without suitable normality or normalization hypotheses, the product \\(HN\\) need not be a subgroup, and the displayed quotient groups need not exist.\n\nThird isomorphism theorem\n\nSuppose\n\n\\[N\\subseteq K\\subseteq G, \\qquad N\\trianglelefteq G, \\qquad K\\trianglelefteq G.\\]\n\nThen \\(K/N\\trianglelefteq G/N\\), and\n\n\\[(G/N)/(K/N)\\cong G/K.\\]\n\nThe map inducing this isomorphism is\n\n\\[G/N\\longrightarrow G/K, \\qquad gN\\longmapsto gK.\\]\n\nIt is well-defined because \\(N\\subseteq K\\), surjective, and has kernel \\(K/N\\). Applying the first theorem proves the claim.\n\nThe theorem is sometimes called the quotient-of-a-quotient theorem. It states that first identifying elements modulo \\(N\\), then modulo the remaining part of \\(K\\), has the same effect as identifying elements modulo \\(K\\) immediately.\n\nThe condition \\(K\\trianglelefteq G\\) cannot simply be replaced by \\(K\\trianglelefteq N\\) or by normality in an unrelated intermediate subgroup: the quotient \\(G/K\\) itself requires normality in \\(G\\).\n\nCorrespondence theorem\n\nLet \\(N\\trianglelefteq G\\), and let \\(\\pi:G\\to G/N\\) be the quotient projection. There is an inclusion-preserving bijection\n\n\\[\\{H\\leq G:N\\subseteq H\\} \\longleftrightarrow \\{L\\leq G/N\\}\\]\n\ngiven by\n\n\\[H\\longmapsto H/N, \\qquad L\\longmapsto\\pi^{-1}(L).\\]\n\nUnder this correspondence,\n\n\\[H\\trianglelefteq G \\quad\\Longleftrightarrow\\quad H/N\\trianglelefteq G/N.\\]\n\nMore generally, for \\(N\\subseteq H\\subseteq K\\subseteq G\\),\n\n\\[H\\trianglelefteq K \\quad\\Longleftrightarrow\\quad H/N\\trianglelefteq K/N.\\]\n\nThe correspondence preserves intersections and generated subgroups. It therefore identifies the subgroup lattice of \\(G/N\\) with the interval of the subgroup lattice of \\(G\\) consisting of subgroups containing \\(N\\)."
    },
    {
      "heading": "Ring-theoretic formulation",
      "text": "For a ring, normal subgroups are replaced by two-sided ideals. If \\(f:R\\to S\\) is a ring homomorphism, then\n\n\\[R/\\ker f\\cong\\operatorname{im}f, \\qquad r+\\ker f\\longmapsto f(r).\\]\n\nWhen rings and homomorphisms are required to preserve a multiplicative identity, the theorem is understood within that convention.\n\nFor a subring \\(A\\subseteq R\\) and a two-sided ideal \\(I\\subseteq R\\),\n\n\\[A/(A\\cap I)\\cong(A+I)/I,\\]\n\nwhere\n\n\\[A+I=\\{a+i:a\\in A,\\ i\\in I\\}.\\]\n\nHere \\(A+I\\) is a subring and \\(I\\) is an ideal in it. For ideals \\(I\\subseteq J\\) of \\(R\\),\n\n\\[(R/I)/(J/I)\\cong R/J.\\]\n\nThe correspondence theorem identifies ideals of \\(R/I\\) with ideals of \\(R\\) containing \\(I\\). Under this correspondence, prime ideals correspond to prime ideals and maximal ideals to maximal ideals.\n\nFor noncommutative rings, specifying two-sided ideals is essential when forming quotient rings. A one-sided ideal generally does not support a well-defined quotient multiplication."
    },
    {
      "heading": "Modules and vector spaces",
      "text": "Let \\(M\\) and \\(P\\) be modules over the same ring \\(R\\), and let \\(f:M\\to P\\) be \\(R\\)-linear. Then\n\n\\[M/\\ker f\\cong\\operatorname{im}f.\\]\n\nFor submodules \\(A,B\\subseteq M\\), the second theorem takes the symmetric form\n\n\\[A/(A\\cap B)\\cong(A+B)/B.\\]\n\nFor submodules \\(A\\subseteq B\\subseteq M\\),\n\n\\[(M/A)/(B/A)\\cong M/B.\\]\n\nThere is also an inclusion-preserving correspondence between submodules of \\(M/A\\) and submodules of \\(M\\) containing \\(A\\).\n\nUnlike arbitrary group subgroups, submodules require no additional normality condition: their underlying additive groups are abelian, and quotient module operations are well-defined for every submodule.\n\nA vector space is a module over a field, so all these statements apply to linear maps and linear subspaces. In particular, for \\(T:V\\to W\\),\n\n\\[V/\\ker T\\cong\\operatorname{im}T.\\]\n\nIf \\(V\\) is finite-dimensional, the dimension formula for a quotient vector space gives\n\n\\[\\dim V-\\dim\\ker T=\\dim\\operatorname{im}T,\\]\n\nwhich yields the rank–nullity theorem.\n\nSimilarly, applying the second theorem to finite-dimensional subspaces \\(U,W\\subseteq V\\) gives\n\n\\[\\dim(U+W) = \\dim U+\\dim W-\\dim(U\\cap W).\\]"
    },
    {
      "heading": "Examples",
      "text": "Reduction modulo an integer\n\nFor a positive integer \\(n\\), the reduction map\n\n\\[\\mathbb Z\\longrightarrow\\mathbb Z/n\\mathbb Z, \\qquad a\\longmapsto a+n\\mathbb Z\\]\n\nis a surjective ring homomorphism with kernel \\(n\\mathbb Z\\). The first theorem identifies the quotient of the integers by this kernel with the ring used in modular arithmetic.\n\nA less immediate example of the second theorem uses the additive subgroups \\(m\\mathbb Z,n\\mathbb Z\\subseteq\\mathbb Z\\), for positive \\(m,n\\). They satisfy\n\n\\[m\\mathbb Z+n\\mathbb Z=\\gcd(m,n)\\mathbb Z,\\]\n\nand\n\n\\[m\\mathbb Z\\cap n\\mathbb Z = \\operatorname{lcm}(m,n)\\mathbb Z.\\]\n\nHence\n\n\\[m\\mathbb Z/\\operatorname{lcm}(m,n)\\mathbb Z \\cong \\gcd(m,n)\\mathbb Z/n\\mathbb Z.\\]\n\nPolynomial evaluation\n\nLet \\(F\\) be a field and \\(a\\in F\\). Evaluating a polynomial at \\(a\\) defines a surjective ring homomorphism\n\n\\[\\operatorname{ev}_a:F[x]\\to F, \\qquad p(x)\\longmapsto p(a).\\]\n\nIts kernel is the ideal generated by \\(x-a\\). Therefore,\n\n\\[F[x]/(x-a)\\cong F.\\]\n\nThe isomorphism sends the residue class of a polynomial to its value at \\(a\\). This illustrates how imposing an algebraic relation through a quotient can reproduce a familiar structure.\n\nSuccessive integer quotients\n\nIn the additive group \\(\\mathbb Z\\), take \\(N=12\\mathbb Z\\) and \\(K=4\\mathbb Z\\). Since \\(N\\subseteq K\\), the third theorem gives\n\n\\[(\\mathbb Z/12\\mathbb Z)/(4\\mathbb Z/12\\mathbb Z) \\cong \\mathbb Z/4\\mathbb Z.\\]\n\nThe subgroup being factored out on the left consists of the residue classes of \\(0,4,\\) and \\(8\\)."
    },
    {
      "heading": "Universal-property interpretation",
      "text": "The first theorem is closely related to the universal property of a quotient. If \\(N\\trianglelefteq G\\), then every homomorphism \\(f:G\\to H\\) satisfying \\(N\\subseteq\\ker f\\) factors uniquely through \\(G/N\\):\n\n\\[f=\\widetilde f\\circ q,\\]\n\nwhere \\(q:G\\to G/N\\) is the quotient projection.\n\nWhen \\(N=\\ker f\\), the induced map \\(\\widetilde f:G/N\\to H\\) is injective and its image is \\(\\operatorname{im}f\\). Restricting its codomain to that image gives the first isomorphism theorem.\n\nThe second and third theorems follow by applying this construction to suitable restricted or induced homomorphisms. This explains both their shared proof pattern and why their isomorphisms are canonical: the maps are determined by the given homomorphisms, inclusions, and quotient projections, without arbitrary choices."
    },
    {
      "heading": "Scope and limitations",
      "text": "The algebraic hypotheses are part of the statements, not merely technical conveniences. Quotient groups require normal subgroups; quotient rings require two-sided ideals; quotient modules require submodules. An arbitrary subset cannot generally be factored out while retaining the relevant operations.\n\nAdditional structure also requires additional care. In topology, a continuous bijection need not be a homeomorphism. Accordingly, an algebraic isomorphism obtained from the first theorem for a continuous homomorphism of topological groups need not preserve topology in both directions.\n\nFor example, the identity map from the additive group \\(\\mathbb R\\) with the discrete topology to \\(\\mathbb R\\) with its usual topology is continuous and bijective. Its kernel is trivial, so the first theorem gives an algebraic isomorphism, but the inverse map is not continuous. A topological version requires the induced bijection to have a continuous inverse; this holds, for example, when the original homomorphism is open onto its image."
    }
  ],
  "links": [
    "abstract-algebra",
    "homomorphism",
    "isomorphism",
    "group-theory",
    "normal-subgroup",
    "surjective-function",
    "injective-function",
    "coset",
    "ring-mathematics",
    "ideal-ring-theory",
    "module-mathematics",
    "vector-space",
    "linear-map",
    "quotient-vector-space",
    "rank-nullity-theorem",
    "integer",
    "modular-arithmetic",
    "field-mathematics",
    "polynomial",
    "universal-property",
    "topology",
    "homeomorphism"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}