1
$\begingroup$

Let $A$ and $B$ be two abelian varieties over any algebraically closed field. Let $A[p^n]$ and $B[p^n]$ denotes the set of $p$-power torsion points of $A$ and $B$. Assume that $A[p^n]$ and $B[p^n]$ have infinitely many points in common, that is, $A[p^{\infty}] \cap B[p^{\infty}]$ is infinite.

Do there exist some formal group laws $F_A$ and $F_B$ associated with $A$ and $B$, respectively, such that $F_A[p^n]$ and $F_B[p^n]$ also have infinitely many points in common?


I am treating formal group and formal group law as two distinct object, though in many places people simple use formal group to mean formal group law.

For abelian variety $A$ has a unique formal group $\hat{A}$ associated with its structure, but, upon choice of coordinates, one can have an isomorphic class of formal group laws $F_A$ of the underlying formal group $\hat{A}$. Similarly, the abelian variety $B$ has a unique formal group $\hat{B}$ associated with its structure, but, upon choice of coordinates, one can have an isomorphic class of formal group laws $F_B$.

The formal groups $\hat{A}, \hat{B}$ capture the local behaviour of the abelian varieties $A,B$ at the identity element and therefore the formal group laws of $F_A$ and $ F_B$ also capture the local behaviour of $A$ and $B$, respectively, near the identity.

Now the $p$-torsion points $A[p^n]$ of $A$ are the solutions of the multiplication-by-$p$ map $[p]_A(x)=0$. These points $A[p^n]$ must also appears in the formal group $\hat{A}$ and therefore solutions of the multplication-by-$p$ map $[p]_{F_A}(x)$ of some suitable formal group law (or formal power series) $F_A$ of $\hat{A}$. The upshot is that the torsion points of an abelian variety transfers (upon suitable choice of coordinates) to some formal group law associated to it.

Thus I think, if $A$ and $B$ shares infinitely many $p$-torsion points, then $F_A$ and $F_B$ will also share infinitely many points.

Elliptic curve and its formal group law can be an example, I think.

$\endgroup$
5
  • 5
    $\begingroup$ For two abstract abelian varieties, what does it mean for them to share torsion points? This seems to inherently rely on a choice of coordinates. $\endgroup$ Commented Jul 24 at 16:52
  • 2
    $\begingroup$ @AchimKrause Maybe the residue field extensions of the torsion points are isomorphic as extensions of the ground field . . . $\endgroup$ Commented Jul 24 at 17:03
  • $\begingroup$ @AchimKrause, I mean the groups of p-power torsion points A[p^n] and B[p^n] have common elements. The points are not abstract, they exists. Otherwise, how can one consider field extensions by those points. Perhaps, I should replace $\mathbb{Q}$ by any algebraically closed field of characteristics 0 or prime $p$ $\endgroup$
    – Learner
    Commented Jul 25 at 1:07
  • 2
    $\begingroup$ If you compare points, you need to fix equations and say these two models of the two varieties intersect in infinitely many points in the ambient projective space. Once these models are fixed there is a choice of the formal group fixed as well and their associated groups can be intersected. Maybe that is what you mean. Certainly the way it is written right now it is very confusing. $\endgroup$ Commented Jul 25 at 12:23
  • $\begingroup$ @ChrisWuthrich, thanks for explaining my question better than me. Indeed, that is something I want to say. Maybe someone can put a sketch/hints of a proper answer, if necessary, by assuming condition $\endgroup$
    – Learner
    Commented Jul 25 at 13:05

0

You must log in to answer this question.