1
$\begingroup$

Let $K$ be a number field (possibly of infinite degree over $\mathbb{Q}$) and $E$ an elliptic curve without complex multiplication.

Let $L:= K(E_{5^{\infty}7^{\infty}11^{\infty}...})$ be the field obtained by adjoining all torsion points of order prime to $6$ to $K$ and let $G:=\operatorname{Gal}(L/K)$.

Let $p\geq 5$ be a prime. Is the group $$(pE(L))^{G}/pE(K)$$ zero for almost all primes $p$?

I am particularly interested in the cases

(1) $K=\mathbb{Q}[\sqrt{-5},\sqrt{-7},\sqrt{-11},...]$

(2) $K=\mathbb{Q}$

EDIT: One must of course assume $\operatorname{rank}_{K}(E)\geq 1$.

$\endgroup$
1
  • $\begingroup$ That looks like a stab in the dark to me. I can't see a reason why this quotient is finite for a single p, let alone trivial. The rank of E(L) is infinite, the quotient is a subgroup of $H^1(G, E[p])$ which is infinite dimensional, because G has many cyclic quotients of order p. $\endgroup$ Commented Aug 7, 2016 at 18:01

0

You must log in to answer this question.

Browse other questions tagged .