I read the book of Kazuya Kato——"Number Theory: Iwasawa theory and modular forms".
In the proof of Mordell's proof of Ramanujan's conjecture,

he constructed "Mordell operator"

I've known his proof, but still don't known how the motivation of his construction. The appearance of "Mordell operator" is a little strange, can anybody explain the motivation? Besides, he said the "Mordell operator" can be generalized to "Hecke operator", and it is much more complicated...... So, I just begin with the "simpler" case.
Any help will be appreciated.:)
Thanks for @David Loeffler's answer, and I read serre's book further. He used lattice to define Hecke operator,
and then derive how it looks like. Besides, he calculates the example of Eisenstein series and $\Delta$-function. For $\delta$ the corresponding eigenvalue is $\tau(n)$.
The definition of Hecke operator is much more simple and great. However, I still wonder the motivation of the definition and what's the nature of it?
Thanks for @David Loeffler again~~ :)
