0
$\begingroup$

Here is the question:

Let F be a field, and R a commutative ring with identity. If φ:F[x]→R is a surjective ring homomorphism such that ker φ = (m(x)) for some m(x) ∈ F [x], prove that φ induces an isomorphism F[x]/(m(x)) →∼ R.

So based on my understanding, to prove isomorphism I must show the transformation is injective and surjective. We already know that it is surjective so I just need to show that it is also one-to-one.

Since the kernal is m(x), we know the kernal is non-trivial. So I'm guessing for that reason φ cannot be injective, but somehow F[x]/(m(x)) →∼ R is injective but I am stuck with that.

I am thinking that I need to show that the kernal of F[x]/m(x)) is 0. But the question says it is m(x). But m(x) cannot be 0 since it is the denominator.

$\endgroup$
  • $\begingroup$ Remember the notation $F[x]/(m(x))$ is for the space of polynomials $F[x]$ modulo the ideal $(m(x))$. This not a fraction in the usual sense (which would make no sense here), but a space in which $m(x)$ and all its multiples are zero. $\endgroup$ – Joel Cohen Nov 11 '15 at 23:57
  • $\begingroup$ I never understand why people ask students such conflated questions. The real question is: Let $\phi : A \to B$ be a surjective ring homomorphism show that $\phi$ induces an isomorphism $A/\text{Ker}(\phi)\to B$. Or an even better question: Let $\phi : A \to B$ be a surjective ring homomorphism. Show that for any ring homorphism $\psi : A\to C$ such that $\psi(\text{Ker}(\phi))=\{0\}$ there exists a (unique) homorphism $\bar \psi : B \to C$ such that $\bar\psi \circ \phi = \psi$. $\endgroup$ – Nex Nov 25 '15 at 8:33
1
$\begingroup$

You're conflating two separate things here:

  • There is the map $\varphi$, from $F[x]$ to $R$. $\varphi$ is surjective but not injective. Certainly, $\varphi$ is not an isomorphism.

  • Meanwhile, we can consider the ring $F[x]/(m(x))$, where $(m(x))$ is the kernel of $\varphi$. The question says that $\varphi$ induces a map from $F[x]/(m(x))$ to $R$; this induced map is different from $\varphi$, although closely related to it. It is this map which you want to show is an isomorphism. So, the first step is . . .

What is the induced map?

(If your abstract algebra class looks anything like the one I took, you're probably much more comfortable with groups right now. If so, think about the analogous question for groups: I have a group $G$, and a surjective homomorphism $\psi$ from $G$ to some group $H$. This induces a map $\hat{\psi}: G/ker(\psi)\rightarrow H$ - do you remember how?)

Once you've identified the induced map, you'll need to show that it is:

  • a (well-defined) homomorphism,

  • surjective, and

  • injective.

$\endgroup$

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.