My question concerns the $U(1)$ Chern-Simons theory with the action $$S = \frac{k}{2\pi}\int A\wedge \mathrm{d}A.$$

In my lecture, it is stated that:

  1. A large gauge transformation involves taking $A\rightarrow A+\eta$, with $\eta$ closed but not necessarily exact. Furthermore $\eta$ has integral periods.

  2. The extra term generated by the gauge transformation $\int \eta \wedge \mathrm{d}A$ is equal to $(2\pi)^2 n$ with $n$ an integer, thus $e^{iS}$ is invariant provided $k$ is an integer.

My questions are:

  1. "$\eta$ has integral periods" -- does this mean $\int \gamma^*\eta$ is an integer for any loop $\gamma$?

  2. How do I see that $\int \eta\wedge \mathrm{d}A$ is of the form $(2\pi)^2n$, where $n$ is some integer?


1 Answer 1

  1. Yes.

  2. This rely on the fact that $\eta$ (which may be a multiple-valued closed 1-form) and $F=\mathrm{d}A$ (which may be a non-exact closed 2-form) are both assumed to have integral periods. Then the wedge product $\eta\wedge F$ also has integral periods, cf. e.g. this Math.SE post.


  1. G. Moore, 2019 TASI Chern-Simons lecture notes; subsection 2.2.1.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.