I consider a Lie group $G$, with a group element $g$ parametrised in some manner with parameter $\theta_i$, $i=1,\cdots, \dim G$. Suppose that $K\subset G$. I want to compute the variation of an group element under the left action of $K$ on $G$.

I'm a little confused to how I can do this: Something like $\delta g =g\delta\theta_iT^i$, with $T^i$ the generators of $G$? but restricted to the subgroup $K$?

  • 2
    $\begingroup$ As written, this is a pure math question. $\endgroup$ Jun 15, 2015 at 15:33

1 Answer 1


Let's say $T^i$ ($i = 1, \dots, \dim G$) are the generators of $G$ chosen in such a way that $T^\mu$ ($\mu = 1, \dots, \dim K$) are the generators of $K \subset G$.

Let $k \in K$ and $g \in G$, then the left action of $k$ on $g$ is just $g' = k g$. Parametrize $k = \exp( \delta \theta_\mu T^\mu )$ then $g' = (1 + \delta \theta_\mu T^\mu) g$, so I would say $\delta g = \delta \theta_\mu T^\mu g$.

  • $\begingroup$ Thanks, but shouldn't it be $g'=g(1+\delta \theta_\mu T^\mu)$ in first order approximation? $\endgroup$
    – user41746
    Jun 15, 2015 at 14:55
  • $\begingroup$ Sorry, I forgot some brackets. Have edited the post now. Why do you say left action and then always put $k = 1 + \delta \theta_\mu T^\mu$ on the right? $\endgroup$
    – Noiralef
    Jun 15, 2015 at 15:18

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.