# What are the different definitions of torsion-free connections in GR? [closed]

What are the different definitions of torsion-free connections in GR?

I am summarizng torsion-less condition in various context, if you know anything about it please let me know.

1. One typical definition of torsion is \begin{align} T(X,Y) = \nabla_X Y - \nabla_Y X - [X,Y] \end{align} with this one can say torsion \begin{align} T_{MN}{}^K = 2 \Gamma_{[MN]}{}^K = \Gamma_{MN}{}^K - \Gamma_{NM}{}^K \end{align} and demand that $T_{MN}{}^K=0$, we have symmetrized $\Gamma$ for first two indices (In above notation).

I can re-write this as \begin{align} [X,Y](\partial) = [X,Y](\nabla) \end{align} where $[X,Y]$ are usual Lie bracket.

2. In Physics, focusing on general relativity, they consider torsion-less connection from the beginning. In some textbook, states via geodesic equation \begin{align} \Gamma_{\mu\nu}{}^\lambda = \frac{\partial x^\lambda}{\partial \xi^{\alpha}} \frac{\partial^2 \xi^\alpha}{\partial x^\mu \partial x^\nu} \end{align} or other textbook states in terms of orthonormal basis \begin{align} \nabla_{e_i} e_j = \nabla_{e_j} e_i = \Gamma_{ij}{}^k e_k \end{align} Above two connections imply torsion-less in its definition. (Of course, one can distinguish $\nabla_{e_i} e_j$ and $\nabla_{e_j} e_i$. $i.e$, One can define torsion as a subtraction between them. )

3. Or some textbook as Carroll, defines torsion as a commutation relation from covariant derivatives $i.e$, \begin{align} [\nabla_A, \nabla_B] V_K = - R_{MNK}{}^L V_L - T_{MN}{}^K \nabla_L V_K \end{align} apparently, with some computation we see that \begin{align} T_{MN}{}^K = 2 \Gamma_{[MN]}{}^K \end{align} which agrees with first statement.

4. Or one can deduce this from \begin{align} L_X(\partial) = L_X(\nabla) \end{align} where $L_X$ is a Lie-deriative along $X$ direction. Applying for $V_K$ we have $\Gamma_{MN}{}^K = \Gamma_{NM}{}^K$

## closed as unclear what you're asking by John Rennie, knzhou, ACuriousMind♦, user36790, ja72Jul 27 '16 at 18:44

Please clarify your specific problem or add additional details to highlight exactly what you need. As it's currently written, it’s hard to tell exactly what you're asking. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

• Would it be a good idea to include a specific question in your post? It looks like , to me anyway, you want to discuss the pros and cons of the various definitions. – user108787 Jul 27 '16 at 6:54
• If you want to turn your findings into a resource, the most common way to do it is to post a question (like, "what are the different definitions of torsion-free connections in GR?"), answer your own question (using the stuff you posted above), and accept that answer. – knzhou Jul 27 '16 at 8:26
• Otherwise, it's unclear what we should do with this question... – knzhou Jul 27 '16 at 8:26
• @knzhou, Thanks! That was my intention!. I changed my question – phy_math Jul 27 '16 at 13:10
• ...and the question here is? – ACuriousMind Jul 27 '16 at 13:37