Skip to main content
Post Closed as "Duplicate" by Qmechanic
added 37 characters in body; edited tags
Source Link
Qmechanic
  • 213.1k
  • 48
  • 590
  • 2.3k

In 'Chern Simons Gauge Theory as a String Theory'Chern Simons Gauge Theory as a String Theory, Witten derives the general coordinate variation of a Wilson loop, i.e., equation 3.11. My question is, how does one derive this? I only managed to obtain a similar result by doing a direct variation and integration by parts, the result I obtained had \begin{equation} F_{IJ}=\partial_I A_J-\partial_J A_I. \end{equation} This is clearly incorrect, since he specifically says that $F=dA+A\wedge A$. I believe I am supposed to vary the path somehow, or that the path ordering somehow gives rise to the term $A\wedge A$, which is just the commutator $[A_I,A_J]$. However, I cannot see how this happens.

In 'Chern Simons Gauge Theory as a String Theory', Witten derives the general coordinate variation of a Wilson loop, i.e., equation 3.11. My question is, how does one derive this? I only managed to obtain a similar result by doing a direct variation and integration by parts, the result I obtained had \begin{equation} F_{IJ}=\partial_I A_J-\partial_J A_I. \end{equation} This is clearly incorrect, since he specifically says that $F=dA+A\wedge A$. I believe I am supposed to vary the path somehow, or that the path ordering somehow gives rise to the term $A\wedge A$, which is just the commutator $[A_I,A_J]$. However, I cannot see how this happens.

In Chern Simons Gauge Theory as a String Theory, Witten derives the general coordinate variation of a Wilson loop, i.e., equation 3.11. My question is, how does one derive this? I only managed to obtain a similar result by doing a direct variation and integration by parts, the result I obtained had \begin{equation} F_{IJ}=\partial_I A_J-\partial_J A_I. \end{equation} This is clearly incorrect, since he specifically says that $F=dA+A\wedge A$. I believe I am supposed to vary the path somehow, or that the path ordering somehow gives rise to the term $A\wedge A$, which is just the commutator $[A_I,A_J]$. However, I cannot see how this happens.

Post Undeleted by Mtheorist
Post Deleted by Mtheorist
Source Link
Mtheorist
  • 1.2k
  • 5
  • 17

Coordinate variation of a Wilson loop

In 'Chern Simons Gauge Theory as a String Theory', Witten derives the general coordinate variation of a Wilson loop, i.e., equation 3.11. My question is, how does one derive this? I only managed to obtain a similar result by doing a direct variation and integration by parts, the result I obtained had \begin{equation} F_{IJ}=\partial_I A_J-\partial_J A_I. \end{equation} This is clearly incorrect, since he specifically says that $F=dA+A\wedge A$. I believe I am supposed to vary the path somehow, or that the path ordering somehow gives rise to the term $A\wedge A$, which is just the commutator $[A_I,A_J]$. However, I cannot see how this happens.