kernel of a homomorphism between algebraic systems
Let be a homomorphism between two algebraic systems and (with as the operator set). Each element corresponds to a subset in . Then forms a partition of . The kernel of is defined to be
It is easy to see that . Since it is a subset of , it is relation on . Furthermore, it is an equivalence relation on :11In general, is a partition of a set iff is an equivalence relation on .
- 1.
is reflexive
: for any , , so that
- 2.
is symmetric
: if , then , so that
- 3.
is transitive
: if , then , so .
We write to denote .
In fact, is a congruence relation: for any -ary operator symbol , suppose and are two sets of elements in with . Then
so . For this reason, is also called the congruence induced by .
Example. If are groups and is a group homomorphism. Then the kernel of , using the definition above is just the union of the square of the cosets of
the traditional definition of the kernel of a group homomorphism (where is the identity of ).
Remark. The above can be generalized. See the analog (http://planetmath.org/KernelOfAHomomorphismIsACongruence) in model theory.