I want to define a tensor that has the first two symmetries of Riemann tensor or maybe the last two. The symmetries of Riemann tensor are: 1) $ R_{\alpha\beta\gamma\lambda}=R_{\gamma\lambda\alpha\beta}$ 2)$R_{\alpha\beta\gamma\lambda}=-R_{\alpha\beta\lambda\gamma}$ and $R_{\alpha\beta\gamma\lambda}=-R_{\alpha\beta\lambda\gamma}$ 3)$R_{\alpha \beta \gamma \lambda} +R_{\alpha\lambda\beta\gamma} + R_{\alpha\gamma\lambda \beta }=0$ Could you help me to learn to define a tensor with mixed symmetries?


1 Answer 1


The skew symmetries and the interchange symmetry can be implemented easily by specifying them as symmetry generators in SymmetrizedArray:

   {{2,1,3,4},-1},(* Skew symmetry in first two slots *)
   {{1,2,4,3},-1},(* Skew symmetry in last two slots *)
   {{3,4,1,2},1}(* Interchange symmetry between first and last pair of slots*)



This seems to work fine: one can check the symmetries by looking at the components and the number of components is also correct: if I recall my General Relativity lectures correctly the Riemann tensor has 21 independent components in four dimensions (20 when including the first (algebraic) Bianchi identity $R_{\alpha \beta \gamma \lambda} +R_{\alpha\lambda\beta\gamma} + R_{\alpha\gamma\lambda \beta }=0$).

The first (algebraic) Bianchi identity can not be implemented easily in terms of symmetry generators but one could manually select a number of components ($\frac{1}{24} d (d-1) (d-2) (d-3)$ where $d$ is the dimension of the tensor (slots)) to be expressed in terms of two other components. Depending on your dimension $d$ this might make sense or not. In $d<3$ the identity gives no additional symmetry, in $d=4$ just one, and so it only looks profitable for larger $d$.


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.