[summary: An abelian group is a group where the operation is commutative. That is, an abelian group G is a pair (X,∙) where X is a set and ∙ is a binary operation obeying the four group axioms plus an axiom of commutativity:
- Closure: For all x,y in X, x∙y is defined and in X. We abbreviate x∙y as xy.
- Associativity: x(yz)=(xy)z for all x,y,z in X.
- Identity: There is an element e such that for all x in X, xe=ex=x.
- Inverses: For each x in X is an element x−1 in X such that xx−1=x−1x=e.
- Commutativity: For all x,y in X, xy=yx.
Abelian groups are very "well-behaved" groups that are often easier to deal with than their non-commuting counterparts.]
An abelian group is a group G=(X,∙) where ∙ is commutative. In other words, the group operation satisfies the five axioms:
- Closure: For all x,y in X, x∙y is defined and in X. We abbreviate x∙y as xy.
- Associativity: x(yz)=(xy)z for all x,y,z in X.
- Identity: There is an element e such that for all x in X, xe=ex=x.
- Inverses: For each x in X is an element x−1 in X such that xx−1=x−1x=e.
- Commutativity: For all x,y in X, xy=yx.
The first four are the standard group axioms; the fifth is what distinguishes abelian groups from groups.
Commutativity gives us license to re-arrange chains of elements in formulas about commutative groups. For example, if in a commutative group with elements {1,a,a−1,b,b−1,c,c−1,d}, we have the claim aba−1db−1=d−1, we can shuffle the elements to get aa−1bb−1d=d−1 and reduce this to the claim d=d−1. This would be invalid for a nonabelian group, because aba−1 doesn't necessarily equal aa−1b in general.
Abelian groups are very well-behaved groups, and they are often much easier to deal with than their non-commutative counterparts. For example, every Subgroup of an abelian group is normal, and all finitely generated abelian groups are a [group_theory_direct_product direct product] of cyclic groups (the [structure_theorem_for_finitely_generated_abelian_groups structure theorem for finitely generated abelian groups]).
Comments
Qiaochu Yuan
I strongly recommend keeping to the standard term "abelian group," even though "commutative group" would be more systematic and sensible. The term "abelian group" is universal - I don't know a single mathematician, book, or paper that uses the term "commutative group" - and people comparing what they read here to what they read anywhere else are just going to be confused, and/or are going to confuse third parties when they ask questions.