5
$\begingroup$

Below is an excerpt from Berthelot's book on crystalline cohomology. I don't understand the last sentence, namely why it follows that $\sigma\circ \varepsilon$ is an isomorphism. For what it's worth, $P^1$ is the sheaf of principal parts and $E$ is an $\mathcal O _X$-module. I can elaborate on what $\sigma,\varepsilon,\tau$ are, but perhaps I'm just missing some basic algebra...

We have an endomorphism of a module which becomes the identity modulo a square zero ideal. Why, in this case, is it an isomorphism?

enter image description here

$\endgroup$

1 Answer 1

10
$\begingroup$

Suppose you have a commutative ring $R$, a square zero ideal $I\subset R$, a $R$-module $M$ and an endomorphism $u$ of $M$ which is the identity modulo $IM$. Then $v:= 1_M-u$ maps $M$ to $IM$, hence $v^2=0$. Therefore $u=1_M-v$ is invertible — its inverse is $1_M+v$.

$\endgroup$
2
  • $\begingroup$ Dear @abx, thanks for the lovely method! If I'm not mistaken, the calculation in the endomorphism ring should just be $(1-u)^2=0$ which gives $1=u(2-u)$, and not $u(1+u)=1$. Am I missing something? $\endgroup$
    – Arrow
    May 6, 2021 at 20:30
  • 1
    $\begingroup$ Sorry, this is a typo — I meant that the inverse is $1+v=2-u$. I edit. $\endgroup$
    – abx
    May 7, 2021 at 4:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.