4
$\begingroup$

In [1,2] the authors proved the Hard Lefschetz theorem in intersection cohomology:

Let $Z$ be a complex projective variety of pure complex dimension $d$, with $\xi\in H^2(Z,\mathbb{Q})$ the first Chern class of an ample line bundle over $Z$. Then there are isomorphisms given by cup product $$ \xi^k\cdot\_\colon IH^{d-k}(Z,\mathbb{Q})\to IH^{d+k}(Z,\mathbb{Q}) $$ for any integer $k>0$.

Question: Does a Hard Lefschetz theorem in intersection cohomology hold for projective varieties of pure dimension over algebraically closed field of characteristic $0$? If it is possible: does one need the formalism of derived categories or something like this?


[1] A. A. Beilinson, J. N. Bernstein, P. Deligne (O. Gabber) - Faisceaux pervers, Astérisque 100 (1982), Soc. Math. Fr.

[2] M. A. A. De Cataldo, L. Migliorini - The Hodge theory of algebraic maps, Ann. Sci. École Norm. Sup., Serie 4, 38 (2005) 693-750.

$\endgroup$

1 Answer 1

7
$\begingroup$

For an algebraically closed field, you need to use the étale topology as the analytic topology is not available. You then must use $\mathbb Q_\ell$-coefficients rather than $\mathbb Q$-coefficients. As far as I know, you must use the formalism of the derived category - I don't know how to define intersection homology in the étale setting without perverse sheaves which are defined via the derived category.

Indeed, the method to deduce the $\mathbb C$ case in section 6 of BBD(G) is to first deduce the $\mathbb C$ case in the étale topology with $\mathbb Q_\ell$-coefficients and then from that deduce the statement in the analytic topology. The first step works without modification in an arbitrary algebraically closed field of characteristic zero.

$\endgroup$

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.