3
$\begingroup$

Let

$ E $ - Elliptic curve defined over $ {\mathbb{Q}} $.

$G_{\mathbb{Q}}$ - The absolute Galois group, $\text{Gal}(\bar{\mathbb{Q}}/\mathbb{Q}) $ of $\mathbb{Q}$.

$ E[3] $ - $3$-torsion points of $ E $.

Suppose $ \rho$ denotes the $ G_{\mathbb{Q}} $-representation associated to $ E[3]$. If $ E $ has a $3$-torsion point over $ \mathbb{Q} $, how to prove $$ \rho \sim \left( {\begin{array}{cc} 1 & \eta \\ 0 & \chi \\ \end{array} } \right) $$ where $ \chi $ is the mod 3 cyclotomic character $?$

Improve this question
$\endgroup$

1 Answer 1

Reset to default
9
$\begingroup$

Assuming that $E$ has a rational $3$-torsion point, that point must be fixed, and hence your representation takes the form $$ \rho \sim \left( \begin{matrix} 1 & \eta_{i} \\ 0 & \chi \end{matrix}\right)$$ as you say. It's a general fact that the determinant of the $G_{\mathbb{Q}}$ representation attached to an elliptic curve is the cyclotomic character. This follows because the Weil pairing is bilinear, alternating, and Galois-invariant. (See Proposition III.8.1 in Silverman's Arithmetic of Elliptic Curves.)

Improve this answer
$\endgroup$
3
  • $\begingroup$ If $E$ has no rational $3$-torsion point, will $\rho$ still have the same representation $?$ $\endgroup$
    – Robert
    Commented May 22, 2014 at 17:14
  • $\begingroup$ @Robert No, not necessarily. $\endgroup$
    – Olivier
    Commented May 22, 2014 at 17:47
  • 2
    $\begingroup$ @Robert No, never. $\endgroup$
    – Joël
    Commented May 22, 2014 at 19:28

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .