So I'm reading Susskind's "The Theoretical Minimum" and on pages 128 and 129, he has the following equations:

First he starts with the Lagrangian for the system:

$$L=\frac{1}{2}(\dot{q_1}^2+\dot{q_2}^2) -V(q_1-q_2)$$

Then he shows the following two derivations:

$$\dot{p_1} = -V'(q_1-q_2)$$ $$\dot{p_2} = V'(q_1-q_2)$$

Where $V'$ is stated to be the derivative of $V$.

The derivation itself is left as an exercise to the reader and I can't figure it out.

In the previous chapter, he had defined the momentum as:

$$p_i=\frac{\partial L}{\partial \dot{q_i}}$$ on page 124.

But since $V$ does not depend upon either $\dot{q_1}$ or $\dot{q_2}$, equations 2 and 3 don't seem to follow from equation 1. Could someone show me what this derivation looks like? Or maybe $p$ means something else here.

  • 2
    $\begingroup$ Are you familiar with the Hamiltonian formalism? You might want to look that up ... $\endgroup$
    – Sanya
    Nov 25 '16 at 0:10
  • 2
    $\begingroup$ The equation you're missing is the Euler-Lagrange equation $\dot{p}_i = \partial L / \partial q_i$. $\endgroup$
    – knzhou
    Nov 25 '16 at 0:11
  • $\begingroup$ Does he give classical mechanics lectures on YouTube, as well as the QM and others that I know are up there? $\endgroup$
    – user108787
    Nov 25 '16 at 0:33
  • $\begingroup$ Sanya, that is the next chapter. knzhou, thank you. I had not recognized it in that form. I can accept your answer if you post it. CountTo10 yes he does. $\endgroup$
    – azani
    Nov 25 '16 at 0:51

I know knzhou already offered the proper answer in the form of a comment, but allow me to expand on it by showing a series of canonical relationships, not all of which are mentioned I believe in Susskind's book.

Given the action $S=\int L~dt$, defined in terms of the Lagrangian $L(q,\dot{q})$, we have the following relationships:

\begin{align} L&=\frac{dS}{dt},\\ p&=\frac{\partial S}{\partial q}=\frac{\partial L}{\partial\dot{q}},\\ H&=p\dot{q}-L=-\frac{\partial S}{\partial t},\\ \dot{p}&=\frac{\partial L}{\partial q}=-\frac{\partial H}{\partial q},\\ \frac{dH}{dt}&=\frac{\partial H}{\partial t}=-\frac{\partial L}{\partial t},\\ \dot{q}&=\frac{\partial H}{\partial p}. \end{align}

These can all be derived from the properties of the Lagrangian, the Euler-Lagrange equation, or the definition of the Legendre transformation that leads to the Hamiltonian.


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.