7
$\begingroup$

In his book Projective Varieties and Modular Forms, M. Eichler uses the notation $L(\lambda, M)$ for the Hilbert function of a finite graded $R=k[x_0, \dots, x_n]$-module $M$. So, $L(\lambda, M) = \dim_k M_\lambda$. By the classical theorem of Hilbert, $L(\lambda, M)$ coincides with a numerical polynomial for large $\lambda$.

On p.27 of his book Eichler writes an equation which amounts to

$$L(\lambda + \mu, M) = L(\lambda, M \otimes \mathcal O(\mu))$$

Of course, this is absolutely self-evident because tensoring with $\mathcal O(\mu)$ only shifts the degree by $\mu$.

This is very much analogous to the equation

$$L(s + \mu, V) = L(s, V\otimes \mathbf Q_{\ell}(\mu)),$$ where $V$ is a geometric representation over $\mathbf Q$.

On p.43, Eichler states in the following form a generalization of Riemann-Roch and Serre dualiy:

Theorem of Duality. Let $n>1$. For a finite, reflexive, and quasifree $R$-module $M$ the following equations hold $(i=1, \dots, n-1)$: $$L(\lambda, \text{Ext}_R^i(M, R)) = L(-\lambda -2n -2, \text{Ext}_R^{n-i}(M_*, R)).$$

This looks suspiciously like a functional equation relating the $L$-functions of $H^i_{et}(X, \mathbf Q_\ell)$ and $H^{2n-i}_{et}(X, \mathbf Q_\ell)$, for a variety of dimension $n$ over $\mathbf Q$.

What is going on here? Where does this similarity comes from? Eichler says nothing of it, but the suggestive notation is probably deliberate...

$\endgroup$
2
  • 1
    $\begingroup$ Isn't this one of the main analogies that motivated the $\mathbb F_1$ program? $\endgroup$ Jul 21, 2013 at 2:55
  • 1
    $\begingroup$ Dear @GjergjiZaimi, would you mind elaborating? $\endgroup$ Aug 6, 2013 at 19:38

0

Your Answer

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

Browse other questions tagged or ask your own question.