1
$\begingroup$

Let $\phi:R \rightarrow S$ be a ring homomorphism and let $J$ be an ideal in $S$, then it is quite easy to prove that $\phi^{-1}(J)$ is an ideal in $R$. But can someone help me with proving the opposite? i.e. if $\phi^{-1}(J)$ is an ideal in $R$ then $J$ is an ideal in $S$. (of course, it could be that the statement I am making is false in which case, can someone explain why that is so?)

$\endgroup$
  • 3
    $\begingroup$ It's false, e.g., when $\phi$ is an inclusion map and $J$ is the union of an ideal of $R$ with a random subset of $S \setminus R$. $\endgroup$ – darij grinberg Apr 30 '15 at 14:56
3
$\begingroup$

The statement you want to prove is wrong, think for example of $\phi \colon \mathbf Z \to \mathbf Q$ the inklusion and $J = 2\mathbf Z$. Then $\phi^{-1}(J) = 2\mathbf Z \subseteq \mathbf Z$ is an ideal, but $J = 2\mathbf Z\subseteq \mathbf Q$ is not, as $\mathbf Q$ is a field and $2\mathbf Z \ne 0,\mathbf Q$.

$\endgroup$
2
$\begingroup$

The converse is not true.

Let $\phi:\mathbb{Z}\rightarrow \mathbb{Q}$ be the inclusion homomorphism. Then $\phi^{-1}(\mathbb{Z})= \mathbb{Z}$ is an ideal in $\mathbb{Z}$ but $\mathbb{Z}$ is not an ideal in $\mathbb{Q}$.

Ideals are kernels of ring homomorphisms. Viewing them like this might help understand why this only works one way: the pre-image along map $g$ of a kernel of map $f$ is the kernel of the composition $fg$ but knowing that the pre-image of a subset $X$ is the kernel of some homomorphism $g$ does not help to get a homomorphism $f$ that has $X$ as kernel.

$\endgroup$

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.