0
$\begingroup$

I believe I've read this before, but I can't find it despite hours of searching on Google.

I've know the common definition of isogeny in elliptic curves, as $\phi:E_1 \rightarrow E_2$ a nonconstant morphism of curves satisfying $\phi(0)= 0$, however somewhere else I read a really lovely intuitive explanation that basically stated that if a curve $E_1$ is isogenic with curve $E_2$, then, not only will points, such as $p_1$, $q_1$ map to $p_2$, $q_2$ but moreover, the mapping of the point $E_1(p_1+q_1)$ to $E_2$, will result in the same point on $E_2$ as when the points $E_1(p_1)$ & $E_1(q_1)$ are first mapped to $E_2$ and then added on the $E_2$ curve.

I hope this makes sense, am I confusing this or does anyone have a resource that shows this detail?

$\endgroup$
6
  • 1
    $\begingroup$ Actually, preserving the group operation is what "morphism" means; that $\phi(p_1+q_1) = \phi(p_1) + \phi(q_1)$. Just a random mapping that isn't constrained to preserve the group properties wouldn't be very interesting... $\endgroup$ – poncho Jun 17 '20 at 20:39
  • $\begingroup$ Ok so my take is correct? Do you have any link to docs that show this characteristic? I can’t find it anywhere. Thanks @poncho $\endgroup$ – Woodstock Jun 17 '20 at 20:47
  • $\begingroup$ Other than the Wikipedia article on "morphism", well, no, I don't have one available. It's one of those ideas that's so "in the water" that it's usually just assumed everyone knows (and so there's little point in spelling it out) $\endgroup$ – poncho Jun 17 '20 at 21:12
  • 2
    $\begingroup$ See theorem 4.8 on page 71 of The Arithmetic of Elliptic Curves by Silverman. It proves that an isogeny between elliptic curves is a group homorphism. $\endgroup$ – user69015 Jun 17 '20 at 21:42
  • $\begingroup$ Thanks all, appreciate it. $\endgroup$ – Woodstock Jun 17 '20 at 21:43
3
$\begingroup$

Given two elliptic curves $E_1$ and $E_2$, an isogeny from $E_1$ to $E_2$ is an algebraic map $\phi$ (it is basically defined from rational functions) that sends the neutral element from $E_1$ to the neutral element of $E_2$.

As stated in Theorem 4.8 in page 71 of The Arithmetic of Elliptic Curves by J.H. Silverman:

Let $$ \phi : E_1 \longrightarrow E_2 $$ be an isogeny. Then $$ \phi(P+Q) = \phi(P) + \phi(Q)\quad \text{for all } P, Q \in E_1.$$

In other words, the isogeny is a group homomorphism (the group law of the curve $E_1$ is preserved after the transfer to the curve $E_2$).

The proof involves the divisor class groups of the curves, using the fact that the isogeny induces a group homomorphism between them.

$\endgroup$
1
  • 1
    $\begingroup$ thank you! marked as answered $\endgroup$ – Woodstock Jun 22 '20 at 11:39

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.