The kernel is defined as the set that maps all elements to zero.

$$\text{ker}\ \alpha = \alpha^{-1}(\mathbf{0})=\{ \mathbf{u} \in U \ : \alpha \mathbf{u}=\mathbf{0}\}$$

Suppose that $\alpha$ is homomorphism between two rings, then we have not one binary operation but two; with two identities: One with respect to addition, the other one with respect to multiplication. But we could also define another kernel in an analogous way (for the other identity):

$$\text{ker}^{'}\ \alpha = \alpha^{-1}(\mathbf{1})=\{ \mathbf{u} \in U \ : \alpha \mathbf{u}=\mathbf{1}\}$$

Do any useful properties arise from defining the fibre over the multiplicative identity?


Sure, you can forget the addition of your ring and get a monoid $(R, \times, 1)$, something like a group but where elements aren't necessarily invertible.

A morphism of rings $f : R \to S$ is in particular a morphism of monoids (i.e. $f(ab) = f(a) f(b)$), and what you're considering is the kernel of that morphism of monoids. So it has all the properties that kernels of monoid morphisms have.

I don't really think you can say anything more specific than that, it's a notion that plays pretty poorly with the addition of $R$ ($1+1$ isn't $1$ very often, for example), so the ring structure is not very relevant, I believe. Compare that with the usual kernel, which is the kernel of the group morphism $(R,+,0) \to (S,+,0)$; since $0 \times x = 0$ for all $x$, the multiplication does play interestingly with the kernel.


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.