If you want to compute crystalline cohomology of a smooth proper variety $X$ over a perfect field $k$ of characteristic $p$, the first thing you might want to try is to lift $X$ to the Witt ring $W_k$ of $k$. If that succeeds, compute de Rham cohomology of the lift over $W_k$ instead, which in general will be much easier to do. Neglecting torsion, this de Rham cohomology is the same as the crystalline cohomology of $X$.

I would like to have an example at hand where this approach fails: Can you give an example for

A smooth proper variety $X$ over the finite field with $p$ elements, such that there is no smooth proper scheme of finite type over $\mathbb Z_p$ whose special fibre is $X$.

The reason why such examples have to exist is metamathematical: If there werent any, the pain one undergoes constructing crystalline cohomology would be unnecessary.

  • 2
    $\begingroup$ Even if such lifting always existed, it need not be functorial, which is a big deal for a cohomology theory. $\endgroup$ Commented May 20, 2010 at 11:14
  • $\begingroup$ I think very similar questions have come up on MO before. You might try searching for them, especially under the tag characteristic-p. $\endgroup$ Commented May 20, 2010 at 11:30
  • $\begingroup$ Here is a link: mathoverflow.net/questions/423/… $\endgroup$
    – Ravi Vakil
    Commented Mar 13, 2013 at 6:15

3 Answers 3


This paper of Serre gives an example (I've justed pasted I. Barsotti's math-sci review). (The paper can be found in Serre's "Collected Works vol. II 1960-1971)

Serre, Jean-Pierre Exemples de variétés projectives en caractéristique $p$ non relevables en caractéristique zéro. (French) Proc. Nat. Acad. Sci. U.S.A. 47 1961 108--109.

An example of a non-singular projective variety $X_0$, over an algebraically closed field $k$ of characteristic $p$, which is not the image, $\text{mod}\,p$, of any variety $X$ over a complete local ring of characteristic 0 with $k$ as residue field. The variety $X_0$ is obtained by selecting, in a 5-dimensional projective space $S$, and for $p>5$, a non-singular variety $Y_0$ which has no fixed point for an abelian finite subgroup $G$ with at least 5 generators of period $p$, of the group $\Pi(k)$ of projective transformations of $S$, but which is transformed into itself by $G$; then $X_0=Y_0/G$. The reason for the impossibility is that $\Pi(K)$, for a $K$ of characteristic 0, does not contain a subgroup isomorphic to $G$. {Misprint: on the last line on p. 108 one should read $s(\sigma)=\exp(h(\sigma)N)$.}

  • 1
    $\begingroup$ The paper is available here. $\endgroup$
    – Watson
    Commented Dec 1, 2018 at 15:01

A general method for for constructing schemes with "arbitrarily bad" deformation spaces (including the non-existence of liftings from char. p to char. 0) is in the following paper:

R. Vakil "Murphy's Law in algebraic geometry: Badly-behaved deformation spaces", Invent. Math. 164 (2006), 569--590.


In general the obstruction to lift a scheme $X$ in characteristic zero is in $H^2(X,T_X)$. For examples of $3$-folds in positive characteristic that connot be lifted in characteristic zero you may look at Theorem 22.4 in Hartshorne's "Deformation Theory".


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.