1
$\begingroup$

In Riemann geometry one can formally solve the parallel transport equation

$$ \dot{v}^\mu + \Gamma^\mu_{\rho\sigma} \, u^\rho \, v^\sigma = 0 $$

of a vector $v$ along a curve with unit tangent vector $u^\mu = \dot{x}^\mu$ using the path-ordered exponential

$$ P^\mu_\nu(s,0) = \left( \text{P exp} - \int_0^s ds \, \Gamma \, u \right)^\mu_\nu $$

$$ v^\mu(s) = P^\mu_\nu(s,0) \, v^\nu(0) $$

Suppose we have

$$ \langle v(s), w(s) \rangle = \langle P(s,0) \, v(0), P(s,0) \, w(0) \rangle $$

with

$$ \langle v, w \rangle = g_{\mu\nu} \, v^\mu \, w^\nu $$

Question: can one show that the path-ordered exponentials cancel?

$$ \langle v(s), w(s) \rangle = \langle v(0), w(0) \rangle $$

$\endgroup$
4
  • $\begingroup$ Can you give some reference to this path ordered exp? In any case the integral in the formula for P should be an exponent $\endgroup$
    – magma
    Commented Sep 2, 2020 at 2:57
  • $\begingroup$ Please refer to Carroll‘s lecture notes, parallel propagator, eq. (3.45) ned.ipac.caltech.edu/level5/March01/Carroll3/Carroll3.html $\endgroup$
    – TomS
    Commented Sep 2, 2020 at 5:10
  • $\begingroup$ The answer is rather simple in SU(N) gauge theories b/c one sees explicitly that the path ordered exponential $P$ is a SU(N) matrix, and that reversing the path order sends each factor to its own inverse; one can use the algebraic properties of the su(n) generators. For the connection coefficients $\Gamma$ I do not see this algebraic property, so the inversion stays on the formal level. $\endgroup$
    – TomS
    Commented Sep 2, 2020 at 5:19
  • $\begingroup$ Ok, this is my guess: reversing the path order sends the tangent vector $u \to -u$, that means each factor $(1 + \Gamma u) ds \to (1 - \Gamma u) ds$ which should proof that $P$ is an SO(3,1) matrix, and that for any path $C$ we have $P^{-1}[C] = P[-C]$ as expected. Hope this is sufficient. $\endgroup$
    – TomS
    Commented Sep 2, 2020 at 6:04

1 Answer 1

1
$\begingroup$

The answer ist „yes“.

https://www.astro.caltech.edu/~george/ay21/readings/carroll-gr-textbook.pdf

Lecture Notes on General Relativity

Sean M. Carroll, Institute for Theoretical Physics

Chapter 3. Curvature

  • the parallel propagator
$\endgroup$

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.