12
$\begingroup$

Let $\Delta$ be the unique normalized cusp form of weight 12 and level $1$. Then Jacobi's wel-known formula states: $$\Delta(z) = q \prod_{n=1}(1-q^n)^{24},$$ where $q=e^{2 i \pi z}$.

For a graduate class I am teaching on modular forms, I am looking for a conceptual proof of this formula, probably using the elliptic interpretation of $\Delta(z)$ as the discriminant of the Weierstrass equation $\mathbb C / \langle 1, z \rangle$. I believe that such a proof exists because Serre, in "cours d'arithmétique", mentions it, with a reference to a prewar paper in German by Hurwitz (Gesamm. Abh, III, no 62), that I was not able to find in my library. Serre himself gives a proof based on the quasi-modularity of the false Eisenstein series $E_2$, and I known another proof from Apostol's textbook, attributed to Siegel, and that uses a clever computation of residues for the $\eta$ function. Both are interesting, but never can be qualified as conceptual by modern standards.

Do you know a proof of Jacobi's formula, or a reference (if possible modern and accessible) for such a proof ?

$\endgroup$
5
  • 1
    $\begingroup$ It would be nice to mention the Hurwitz reference here (which journal and which year ?). $\endgroup$ Oct 1, 2012 at 16:00
  • $\begingroup$ You're right. Done. $\endgroup$
    – Joël
    Oct 1, 2012 at 16:18
  • 1
    $\begingroup$ It's not "conceptual" in the sense you desire, but Weil's "Sur une formule classique" (Collected works vol III pp 198-200) uses the primordial version of "converse theorems" (in a self-contained fashion) to give a substantially believable proof. $\endgroup$ Oct 1, 2012 at 16:51
  • $\begingroup$ Dear Paul, thanks for your answer. This is doubtlessly a conceptual proof, which for me is satisfying. But it is not something I can put in my course since I haven't time to discuss the converse theorems. $\endgroup$
    – Joël
    Oct 2, 2012 at 13:59
  • 2
    $\begingroup$ You may be aware of this already, but an anonymous blogger gave a nice argument using Hecke eigenvalues of the $E_2$ series: galoisrepresentations.wordpress.com/2012/10/26/… $\endgroup$
    – S. Carnahan
    Nov 6, 2012 at 9:14

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.