3
$\begingroup$

For each of the following rings find a list of representatives of all simple $R$-modules:

1) $R=\mathbb C[x]$

2) $R=\mathbb R[x]$

What I've tried was:

I know that $M$ is a simple $R$-module if and only if $M \cong R/I$ where $I$ is a maximal ideal, so, for each case, first I've tried to find all maximal left ideals.

1) Let $I$ be a maximal left ideal of $\mathbb C[x]$, then $I=<x-\alpha>, \alpha \in \mathbb C$. So every simple module $M$ is isomorphic to a quotient of the form $\mathbb C[x]/<x-\alpha>, \alpha \in \mathbb C$. My question is: How many quotients are up to isomorphisms?, I suspect that $C[x]/<x-\alpha>$ is isomorphic to $\mathbb C$ for any $\alpha$ but I couldn't prove this.

With the same idea, in 2) I've concluded that if $I$ is a maximal ideal in $\mathbb R[x]$, then $I=<x-\lambda>, \lambda \in \mathbb R$ or $I=<(x-z)(x-\overline{z})>, z=a+bi, b \neq 0$. So if $M$ is a simple module, then $M \cong \mathbb R[x] /<x-\lambda>$ or $M \cong \mathbb R[x] /<(x-z)(x-\overline{z})>$. I have the same problem that in 1), I suspect that $\mathbb R[x]/<x-\lambda>$ is isomorphic to $\mathbb R$ but I don't know how to prove this, as for the quotients of the form $\mathbb R[x]/<(x-z)(x-\overline{z})>$ I don't know how many are there up to isomophism.

Any help with these points would be greatly appreciated.

$\endgroup$
  • 1
    $\begingroup$ To write ideals using angle brackets use \langle and \rangle. $\endgroup$ – Mariano Suárez-Álvarez Nov 3 '14 at 7:40
1
$\begingroup$

You're almost done, everything's fine so far.

My question is: How many quotients are up to isomorphisms?, I suspect that ${\mathbb C}[x]/(x−\alpha)$ is isomorphic to ${\mathbb C}$ for any $\alpha$ but I couldn't prove this.

As complex vector spaces, all the ${\mathbb C}[x]$-modules ${\mathbb C}[x]/(x-\alpha)$ are $1$-dimensional (spanned by the cosets of constant polynomials), hence isomorphic to ${\mathbb C}$; in particular, they are mutually isomorphic. However, as ${\mathbb C}[x]$-modules - and this is what you consider here - they are pairwise non-isomorphic. To prove this, think on how you can recover $\alpha$ from the action of $x$ on ${\mathbb C}[x]/(x-\alpha)$.

The real case is similar. As ${\mathbb R}$-vector spaces, all the modules ${\mathbb R}[x]/(x-\alpha)$ are $1$-dimensional, hence isomorphic to ${\mathbb R}$, but you want to consider them as ${\mathbb R}[x]$-modules. The other family you exhibited consists of $2$-dimensional modules. To show that for $\lambda\neq\mu\in{\mathbb R}$ the ${\mathbb R}[x]$-modules ${\mathbb R}[x]/(x-\lambda)$ and ${\mathbb R}[x]/(x-\mu)$ are not isomorphic, proceed as in the complex case. For $\lambda\notin {\mathbb R}$, again look at the action of $x$ on ${\mathbb R}[x]/(x-\lambda)(x-\overline{\lambda})$ and try to recover $\{\lambda,\overline{\lambda}\}$ from it.

$\endgroup$
  • $\begingroup$ Thanks for your answer. Let me see if I could partially follow you: consider $M=\mathbb C[x]/ \langle x-\alpha \rangle$ and $N=\mathbb C[x]/ \langle x-\lambda \rangle$, $\alpha \neq \lambda$. In $M$, $x.z=(x-\alpha+\alpha).z=(x-\alpha).z+\alpha.z=\alpha z$. Similarly, in $N$ we have $x.z=\lambda z$. This means that the ring acts differently in each module if $\lambda \neq \alpha$. So, my question may be silly but how can I assure that this implies these two modules can't be isomorphic? $\endgroup$ – user16924 Nov 3 '14 at 11:59

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.