2
$\begingroup$

Suppose $\mathbb{K}\subseteq\mathbb{L}$ algebraic and $\mathbb{L}\subseteq\mathbb{F}$, and take $S\subseteq\mathbb{F}$ a set of transcendental elements over $\mathbb{L}$: then is $\mathbb{K}(S)\subseteq\mathbb{L}(S)$ algebraic?

I'm not sure if this line of reasoning is correct: the following holds $$\mathbb{L}(S)=\mathbb{K}(S)(\mathbb{L})\label{*}\tag{*}$$ Because the smallest subfield of $\mathbb{F}$ containing $\mathbb{K}$, $\mathbb{L}$ and $S$ is exactly the one containing $\mathbb{L}$ and $S$.

The field $\mathbb{L}(S)$ is being generated over $\mathbb{K}(S)$ by elements which are algebraic over $\mathbb{K}$ and hence over $\mathbb{K}(S)$ (for it contains $\mathbb{K}$), so the extension is in fact algebraic. Is this sufficient, or are there further considerations to be made?

$\endgroup$
  • 2
    $\begingroup$ I find that sufficient. If you want, you can say any element of $L(S)$ is a a rational function in the elements of $S$ with coefficients from $L$; if we simply adjoin the coefficients to $K(S)$ (which will be an algebraic extension) we get an extension field containing that arbitrary element of $L(S)$. $\endgroup$ – arctic tern Dec 1 '16 at 15:47
1
$\begingroup$

You need to show that an element of $L(S)$ is algebraic over $K(S)$.

Since quotients of algebraic elements are algebraic, it is not restrictive to assume the element you pick in $L(S)$ actually belongs to $L[S]$. But then you just need to show that a monomial in $L[S]$ is algebraic over $K(S)$, because sums of algebraic elements are algebraic. Now use the fact that products of algebraic elements are algebraic.

$\endgroup$

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.