3
$\begingroup$

Question. Is it true that a finite extension $K:F$ is simple iff the purely inseparable closure is simple over $F$?

I think have an argument to support the above.

First we show the following:

Lemma. Let $K:F$ be a finite extension and $S$ and $I$ be the separable and purely inseparable closures. Then $K=SI$.

Proof. We note that $K$ is separable over $I$. This is because if $K$ is not separable over $I$, then the purely inseparable degree of $K$ over $I$ is greater than $1$. So there would exist an element $\alpha\in K\setminus I$ which is purely inseparable over $I$. But then $I(\alpha)$ would be purely inseparable over $F$, giving $I(\alpha)=I$, that is $\alpha\in I$, a contradiction. Similarly we have $K$ purely inseparable over $S$. Therefore $K$ is both separable and purely inseparable over $SI$, meaning $K=SI$.

The following is a well known theorem:

A finite extension is simple iff there are only finitely many intermediate fields.

Now we show that if $K:F$ is a finite simple extension, then so is $I:F$, where $I$ is the purely inseparable closure. Since there are only finitely many intermediate fields between $K$ and $F$, therefore there are only many intermediate fields between $I$ and $F$. Therefore $I:F$ is simple.

Conversely, assume that $I:F$ is simple. We show that there are only finitely many intermediate fields between $K$ and $F$. Let $M$ be an intermediate field between $K$ and $F$. Let $M_I$ be the purely inseparable closure of $M$ over $F$ and $M_S$ be the separable closure of $M$ over $F$. Then by the lemma above we have $M=M_IM_S$. Also, $M_I=M\cap I$ and $M_S=M\cap S$ are subfields of $I$ and $S$ which contain $F$, and knowing that $S:F$ is simple (since finite separable extensions are simple), there are only finitely many composites $M_IM_S$. Therefore there are only finitely many intermediate field between $K$ and $F$ and thus $K:F$ is simple.

EDIT: In the proof of the lemma I write that if the purely inseparable degree of $K:I$ is more than $1$, there exists $\alpha\in K$ which is purely inseparable over $I$. I am, on a careful revision, not able to justify this statement.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.