Let $\mathcal O$ be a complete DVR with fraction field $K$, maximal ideal $\mathfrak p$ and residue field $\widetilde K=\mathcal O/\mathfrak p$. Now consider a subring $A\subset \mathcal O$ with the following properties:

  1. $A$ is a local ring with maximal ideal $\mathfrak m$ and residue field $L=A/\mathfrak m$
  2. $\mathcal O$ is the integral closure of $A$ in $K$.

Then, is it true or not the following claim?

$\widetilde K|L$ is a finite field extension


This is not true. Consider the following counterexample: denote by $\overline{\Bbb{Q}}$ the algebraic closure of $\Bbb{Q}$.

Call $\mathcal{O} = \overline{\Bbb{Q}} [[X]]$ (it is well known that this is a complete DVR), $\mathfrak{p}$ its maximal ideal, and $A= \Bbb{Q} + \mathfrak{p}$.

Then $A$ is a local subring of $\mathcal{O}$ with maximal ideal $\mathfrak{p}$, and since $\mathcal{O}$ is an integral extension of $A$, it is its integral closure.

But now, $[\mathcal{O} / \mathfrak{p} : A / \mathfrak{p} ] = [\overline{\Bbb{Q}} : \Bbb{Q}] = \infty$.

  • $\begingroup$ Is there some condition we can impose on $A$ to obtain a positive result? $\endgroup$ – ginevracoal Mar 22 '16 at 22:06
  • 1
    $\begingroup$ I don't think so. This problem sounds like asking: "suppose we have an algebraic extension of fields: under which assumptions it is a finite extension?". This question has no meaningful answers in my opinion. $\endgroup$ – Crostul Mar 22 '16 at 22:11

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.