3
$\begingroup$

Let $X$ be a projective variety.

Consider $Sym^2X$, the quotient of $X \times X$ by the involution $(x, x') \mapsto (x', x)$.

What is the relation between the (mixed) Hodge numbers of $Sym^2 X$ and the ones of $X$? Is there a simple formula?

$\endgroup$

1 Answer 1

4
$\begingroup$

For a $V$-manifold the Hodge structure in each degree is pure (see Peters-Steenbrink, § 2.5 in Mixed Hodge structures, Springer). Thus $H^{p,q}(Sym^2X)$ is just the part of $H^{p,q}(X\times X)$ invariant under the involution. It is easy to compute the dimension using the Künneth formula. For instance, if I didn't make a mistake, you find $h^{2,0}(Sym^2X)= h^{2,0}(X)+\frac{1}{2} h^{1,0}(X)(h^{1,0}(X)-1)$, $h^{1,1}(Sym^2X)= h^{1,1}(X)+ h^{1,0}(X)^2 $.

$\endgroup$
2
  • 3
    $\begingroup$ I think you're assuming $X$ smooth (but maybe "pokerhodge" is doing that too without saying it). What you're saying is true also in the context of mixed Hodge theory, though. If $X$ is any complex algebraic variety and $G$ is a finite group then the rational MHS on $H^\bullet(X/G)$ coincides with the MHS on the $G$-invariants on $H^\bullet(X)$. $\endgroup$ Oct 31, 2013 at 9:18
  • $\begingroup$ No, I wasn't asuming $X$ smooth! Thanks to abx and Dan $\endgroup$
    – pokerhodge
    Oct 31, 2013 at 9:22

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.