4
$\begingroup$

I really couldn't figure out the answer to the following question: Let $X$ be a scheme of finite type over a field $k$ and let $K$ be an extension field of $k$. Let $X_K := K \times_k X$ be the base extension and let $p: X_K \rightarrow X$ be the projection. Is it true that $p$ maps closed points to closed points?

This should be true if $K$ is of finite type over $k$ (so not necessarily finite!). Proof: $X$ is Jacobson, the projection $p$ is of finite type and now it follows for example from the Stacks Project Lemma 27.17.8 (see http://stacks.math.columbia.edu/tag/01TB) that $p$ maps closed points to closed points. But I would like to have the same for any $K$. Is this true?

$\endgroup$
8
$\begingroup$

No, it is not ! For instance, if $k=\bar{\mathbb{Q}}$, $K=\mathbb{C}$ and $X$ has dimension $\geq 1$, there is exactly one closed point of $X_K$ above each closed point of $X$. But there is only countably many closed points of $X$, whereas $X_K$ has uncountably many closed points.

In general, a point of $X$ will be the image of a closed point of $X_K$ if and only if its residue field may be embedded in a finite extension of $K$ (in a way that respects $k$).

In particular, your statement is not true in general, even when $K$ is of finite type over $k$ : for instance, if $K$ is the residue field of a generic point of $X$. The problem with your argument is that $K$ is of finite type over $k$ as a field, but not as an algebra.

| cite | improve this answer | |
$\endgroup$

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.