Consider the following commutative diagram of schemes:  $\require{AMScd}$ \begin{CD} T @>{}>> X\\ @VVV @VVV\\ T' @>{}>> Y \end{CD}

where  $T\hookrightarrow T'$ is a Henselian extension and $X\to Y$ is a smooth morphism. Can one always find a morphism $T'\to X$ such that the following diagram is commutative:

$$\begin{array}{ccc} T & \rightarrow & X \\ \downarrow & \nearrow & \downarrow \\ T' & \rightarrow & Y \end{array} $$

If $T\hookrightarrow T'$ is a first order thickening , this is just the claim that smooth implies formally smooth. It follows that the claim is true if we replace $T'$ by its completion along T.  What about the general case?

We are mainly interested in the case when $T$ is a spectrum of a field $F$. In this case, $T'$ is a spectrum of Henselian ring with residue field $F$.

In fact, we did not find the notion of Henselian extensions in the literature (though it looks rather natural to have). we will be grateful for a reference for this notion too.


In the local case this is a well known property of henselian rings. It is standard when $X \to Y$ is étale; in the general case it follows from the fact that any smooth maps is Zariski-locally a composite $X \to \mathbb{A}^n_Y \to Y$, where $X \to \mathbb{A}^n_Y$ is étale.

About the general case, for the lifting property to hold you need $T$ to be affine. I don't know what a "henselian extension" is; if it is of the type $\operatorname{Spec}(A/I) \subseteq \operatorname{Spec}A$, where $(A,I)$ is a henselian pair, I would suspect that the answer is still positive, but I don't know a reference.

  • $\begingroup$ Is it still true (in the affine case) if we replace smooth by formally smooth? $\endgroup$ – Yai0Phah Nov 6 '19 at 11:02

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.