5
$\begingroup$

Are automorphic quotient for quaternion algebras always compact (safe the totally split case)?

Is there any good reference for proof of this fact, or easy arguments to say do?

$\endgroup$
8
$\begingroup$

Weil's "Adeles and algebraic groups" proves this for general division algebras. It is labelled "Fujisaki's Lemma", and is a natural division-algebra generalization of the analogue for number fields themselves, namely, that $\mathbb J^1/k^\times$ is compact, where $\mathbb J^1$ is the collection of ideles of idele-norm $1$. For number fields, this compactness implies finiteness of (generalized) class number, and implies the (generalized) units theorem.

$\endgroup$
  • 3
    $\begingroup$ For those with a physical book rather than a computer-searchable .pdf, this is Lemma 3.1.1. $\endgroup$ – nfdc23 Jan 15 '18 at 23:28
  • $\begingroup$ Paul, I thought Fujisaki's lemma is true for any central simple algebra (not just division algebras), including in this case $B=\mathrm{M}_2(F)$. So the OP need not exclude the ``totally split'' case. Right? But the orbifolds obtained from quotients by discrete subgroups are only compact when $B$ is a division algebra (by the theorem of Hey). So there are two layers here, I think. $\endgroup$ – John Voight Jan 17 '18 at 0:21
  • $\begingroup$ @JohnVoight, perhaps there is ambiguity in what "Fujisaki's lemma" refers to. For me, it only refers to the division algebra case. I don't off-hand know what more sophisticated analogous assertion would be true in the non-compact quotient case. In particular, although I may just be out of the loop, the impression I have is that there is a measure-theoretic argument that gives compactness in the division algebra case, and would not have an obvious extension otherwise. Perhaps this is just a naming convention... I'm curious: what further "F's lemma" exists in the wild? $\endgroup$ – paul garrett Jan 17 '18 at 0:28
  • $\begingroup$ @paulgarrett, you're right, in Fujisaki's article "On the Zeta-Function of the Simple Algebra over the Field of Rational Numbers", the statement is Theorem 8.3, pg. 599, and at the beginning of section 8, he supposes that $B$ is a division algebra! I must be wrong about the matrix algebra case. $\endgroup$ – John Voight Jan 18 '18 at 3:17
5
$\begingroup$

What does ``automorphic quotient'' mean?

Does my book (http://quatalg.org), Main Theorem 38.4.3 (a theorem of Hey) answer your question? A quaternion algebra $B$ over a field $F$ is either isomorphic to $\mathrm{M}_2(F)$ (is this your ``totally split case''?) or is a division algebra over $F$, and in the latter case, Hey's theorem applies.

$\endgroup$
  • 1
    $\begingroup$ Usually, "automorphic quotient" means adelic points mod rational points, though in this case I suppose the OP means also mod center. $\endgroup$ – Kimball Jan 16 '18 at 4:55

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.