logo

Cayley's Theorem Proof 📂Abstract Algebra

Cayley's Theorem Proof

Theorem 1

Every group is isomorphic to some subgroup of the symmetric group.

Explanation

This concise yet significant theorem conveys the message that by studying symmetric groups, one can grasp all groups.

Proof

Although the proof may seem tedious at first glance, the techniques involved are quite interesting, so it is recommended to try and follow it at least once.


Part 1. f:GGf : G \to G' is injective then Gf(G)G \simeq f (G)

For group GG and GG', let’s show that if a homomorphism f:GGf : G \to G' is injective, then Gf(G)G \simeq f (G).

Definition of a group: A group is a binary operation structure that satisfies the following properties.

  • (i): The operation is associative.
  • (ii): There is an identity element for all elements.
  • (iii): There is an inverse for all elements.

ff, when viewed as a function f:Gf(G)f : G \to f(G) with the image of GG under ff, f(G)Gf(G) \subset G', as its codomain, is naturally surjective. Assuming that ff is injective, showing that f(G)f(G) is a group leads to Gf(G)G \simeq f(G).

If we let x,yGx,y \in G and x,yGx', y ' \in G', then f(xy)=f(x)f(y)=xy f(xy) = f(x) f(y) = x ' y’ so f(G)f(G) is closed under the operation of GG'. Also, since f(G)Gf(G) \subset G' and GG' is a group, it satisfies the associative law.

If we designate ee as the identity of GG, and ee' as the identity of GG', ef(e)=f(e)=f(ee)=f(e)f(e) e' f(e) = f(e) = f(ee) = f(e) f(e) then f(G)f(G) has the identity element ee'.

Lastly, e=f(e)=f(xx1)=f(x)f(x1)=xf(x1) e' = f(e) = f(x x^{-1}) = f(x) f(x^{-1}) = x ' f(x^{-1}) this means every xf(G)x' \in f(G) has an inverse f(x1)f(x^{-1}), so f(G)f(G) is a group.


Part 2. ϕ:GSG\exists \phi : G \hookrightarrow S_{G}

Now, proving the existence of a monomorphism (a homomorphism that is also injective) ϕ:GSG\phi : G \to S_{G} concludes the proof.

Defining λx:GG\lambda_{x} : G \to G for xGx \in G as λx(g):=xg\lambda_{x} (g) := xg, λx(a)=λx(b)    xa=xb    a=b \lambda_{x} (a) = \lambda_{x} (b) \implies xa = xb \implies a = b shows that λx\lambda_{x} is injective. Also, for every cGc \in G, λx(x1c)=xx1c=c \lambda_{x} (x^{-1} c ) = x x^{-1} c = c thus λx\lambda_{x} is surjective, and therefore λx:GG\lambda_{x} : G \to G is a permutation of GG.

So, now it is possible to define ϕ:GSG\phi : G \to S_{G} for xGx \in G as ϕ(x)=λx\phi (x) = \lambda_{x}. If we define ϕ(x)=ϕ(y)\phi (x) = \phi (y), then λx=λy\lambda_{x} = \lambda_{y} and λx(e)=λy(e)\lambda_{x} (e) = \lambda_{y} (e), therefore xe=ye    x=y x e = y e \implies x = y meaning, ϕ\phi is injective. Meanwhile, since ϕ(xy)=λxy\phi (xy) = \lambda_{xy}, thus λxy(g)=(xy)g\lambda_{xy}(g) = (x y) g, and since λx,λy\lambda_{x} , \lambda_{y} is a permutation, (λxλy)(g)=λx(λy(g))=λx(yg)=x(yg) ( \lambda_{x} \lambda_{y} ) (g) = \lambda_{x} (\lambda_{y} (g)) = \lambda_{x} (yg) = x (yg) To summarize, ϕ(xy)=λxy=λxλy=ϕ(x)ϕ(y)\phi (xy) = \lambda_{xy} = \lambda_{x} \lambda_{y} = \phi (x) \phi (y) thus, ϕ\phi is a homomorphism.


Therefore, GG is isomorphic to some subgroup of SGS_{G}.


  1. Fraleigh. (2003). A first course in abstract algebra(7th Edition): p82. ↩︎