Group Theory/The symmetric group
Definition (symmetric group):
Let be a set. Then the symmetric group of is defined to be
- ;
that is, it is the set of all bijective functions from to itself with composition as operation.
Definition (permutation):
A permutation is, by definition, an element of .
Proposition (symmetric group essentially depends only on the cardinality of the underlying set):
Let be sets of the same cardinality. Then there exists a group isomorphism
- .
Proof: Suppose that is a bijective function. Then the group isomorphism is given by
- ;
indeed, an inverse is given by
- .
Definition (finite symmetric group):
Let . Then the symmetric group of order , denoted , is defined to be
- .
Theorem (Cayley's theorem):
Let be a finite group, and set . Then there exists a subgroup of which is isomorphic to .
Proof: acts transitively on itself by left multiplication in the category of sets. This means that we have a group homomorphism . Moreover, this morphism is injective; indeed, only the identity element of induces the identity element in . Hence, the claim follows from the first Noether isomorphism theorem.
Definition (matrix representation of a permutation):
The representation of in the category of vector spaces over a field given by
is called the matrix representation of the permutations contained within .
Definition (sign):
Let be a permutation. Then the sign of , written , is defined to be , where there exist transpositions such that .
The following proposition shows that this notion is well-defined:
Proposition (equivalent characterisations of the sign of a permutation):
Definition (alternating group):
Let . Then the alternating group, a subgroup of , is defined to be
- .
Proposition (alternating group is maximal and normal in the symmetric group):
Let . Then , and further is a maximal subgroup of .
In particular, is a maximal normal subgroup in (ie. maximal among the normal subgroups).
Proof: Note first that is normal as the kernel of a group homomorphism. We then have that is a group homomorphism from to , and by the first Noether isomorphism theorem, . In particular, there are only two cosets of . Suppose that there existed a subgroup . Then by the degree formula, we would have , so that either or . In both cases, one of the inclusions is not strict, a contradiction.
Proposition (conjugation in the symmetric group is re-labeling):
Let be a cycle, and let be any element. Then
- .
Proposition (in degrees 5 or larger all three-cycles are conjugate in the alternating group):
Let , and let and be any two three-cycles in . Then there exists such that .
Proof: Since being in the same conjugacy class is an equivalence relation, assume .
Theorem (in degrees 5 or larger the alternating group is simple):
Let . Then is a simple group.
Proposition (in degrees 5 or larger neither the alternating nor the symmetric group are solvable):
Let . Then and are both not solvable.
Proof: Since is a maximal normal subgroup of ,