The shift of a scalar field $\Phi$: $$ \Phi \rightarrow \Phi'=\Phi - i \epsilon $$ is generated by $$ G = -i \frac{d}{d\Phi},$$ because $$ \mathrm{e}^{-i \epsilon \frac{d}{d\Phi} } \Phi = (1-i\epsilon \frac{d}{d\Phi} + \ldots ) \Phi \approx \Phi - i \epsilon .$$

What is the analogous generator for shifts of a Weyl spinor field $\Psi$ $$ \Psi \rightarrow \Psi'=\Psi - i \xi \quad ? $$ In index notation, we have $$ \Psi_\alpha \rightarrow \Psi_\alpha'=\Psi_\alpha - i \xi_\alpha .$$ Naively, one gets $$ G_\beta = -i \frac{d}{d\Psi_\beta},$$ $$ \mathrm{e}^{-i \xi_\beta \frac{d}{d\Psi_\beta} } \Psi_\alpha = (1-i\xi_\beta \frac{d}{d\Psi_\beta} + \ldots ) \Psi_\alpha \approx \Psi_\alpha - i \xi_\alpha .$$

where I used $\frac{d}{d\Psi_\beta} \Psi_\alpha = \delta_{\alpha \beta} $.

I'm really unsure about the spinor indices here. Shouldn't there be a spinor metric $i \sigma_2$ somewhere, because we are multiplying spinors? Does $\frac{d}{d\Psi_\beta} \Psi_\alpha = \delta_{\alpha \beta} $ make sense at all?

  • 1
    $\begingroup$ $\frac{\partial}{\partial\Psi_\beta}\Psi_\alpha=\delta^\beta_\alpha$ is a well-defined result of the basic algebra of spinors (but note the position of indices). $\endgroup$ Jan 4, 2017 at 11:10

1 Answer 1


As in my comment above, $$ \frac{\partial}{\partial\Psi_\beta}\Psi_\alpha=\delta^\beta_\alpha $$ is one of the basic results of the algebra of spinors (but notice that the index $\beta$ is an upper index).

The generator $$ G_\beta = -i \frac{d}{d\Psi_\beta}, $$ is used, for example, to define supercovariant derivatives (e.g., see Superspace, where you can find $D=\frac{\partial}{\partial\theta}+\cdots$).


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.