5
$\begingroup$

Before I begin: I'm new to category theory.

I'm trying to show that if $k$ is a kernel of some morphism of some category, then it is monic.

Here is my reasoning so far:

Suppose the kernel of some morphism $f: A \longrightarrow B$ is $k: K \longrightarrow B$. I want to show that for any object $C$ and for any morphisms $g,h: C \longrightarrow K$, $kg=kh$ implies $g=h$.

By the definition of the kernel, $fk = 0_{KB}$ where $0_{KB}$ is the unique zero morphism $K \longrightarrow B$. Next, note that $kg, kh : C \longrightarrow A$ are morphisms such that $fkg = 0_{CB} = fkh$ by uniqueness of $0_{CB}$. So $0_{KB} \ g = 0_{CB} = 0_{KB} \ h$. Here is where I am stuck: intuitively it seems obvious that $g=h$ after this. Why?

I also would like to use the existence of unique morphisms $u,v: C \longrightarrow K$ such that $ku = kg$ and $kv = kh$, respectively; however, I don't know where to apply them.

Thanks for any help.

$\endgroup$
  • $\begingroup$ This statement is a special case of more general abstract categorical statement: every equalizer is monic. The kernel of $f$ is the equalizer of $f$ and $0_{AB}$. $\endgroup$ – Oskar Jan 21 '14 at 16:06
  • $\begingroup$ i have a question, in your question, you define $k : K \to B$. i don't understand, by my decision, kernel is $k : K \to A$. $\endgroup$ – user207839 Jan 14 '15 at 14:10
4
$\begingroup$

The universal property says that whenever $l:C\to A$ is a morphism such that $fl=0$, then there is a unique morphism $l':C\to K$ such that $kl'=l$.
Now $kg=kh$ is a morphism $C\to A$ such that $f(kg)=0$, so there is a unique morphism $l':C\to K$ such that $kl'=kg=kh$. We conclude that $l'=g=h$.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.