In the book of Srednicki in equation (35.29) a rather peculiar algebraic operation is carried out on spinors that I am not able to understand. It's

$$ [\psi_\dot{a}^{\dagger} \overline{\sigma}^{\mu\dot{a}c}\chi_c]^{\dagger} = \chi^{\dagger}_\dot{c} (\overline{\sigma}^{\mu a \dot{c}})^{\ast}\psi_a$$

Actually, I would expect

$$ [\psi_\dot{a}^{\dagger} \overline{\sigma}^{\mu\dot{a}c}\chi_c]^{\dagger} =\chi^{\dagger}_\dot{c} (\overline{\sigma}^{\mu \dot{a} c})^{\dagger}\psi_a= \chi^{\dagger}_\dot{c} (\overline{\sigma}^{\mu c\dot{a}})^{\ast}\psi_a$$

as $x^\dagger \equiv (x^{T})^{\ast}$ and "transpose" usually exchanges indices. Furthermore, Srednicki mentions that the hermiticity of $\overline{\sigma}$ is only used in a following step of (35.29) so apparently it has not be used here. If somebody could give me some insight I would be grateful.


You are thinking about the operation $\dagger$ incorrectly.

For $\psi$, $\chi$ being a fermionic fields taking values in some representation (I will gloss over the subtleties corresponding to dotted and undotted indices), define a map $$ \dagger: \psi \mapsto \psi^\dagger $$ with the (non-exhaustive list of) properties $$ \begin{align} (\alpha \psi)^\dagger &= \alpha^\ast \psi^\dagger,\\ (\psi^\dagger)^\dagger &= \psi,\\ (\psi^\dagger\chi)^\dagger &= \chi^\dagger\psi. \end{align} $$ Now consider a linear combination $\chi'{}^i = \bar{\sigma}^{ij} \chi_j$, then clearly by the first property $$ (\chi'{}^i)^\dagger = (\bar{\sigma}^{ij})^\ast \chi^\dagger_j . $$ With this you will understand what is going on: $$ [\psi_i^\dagger \bar{\sigma}^{ij} \chi_j]^\dagger = [\psi_i^\dagger \chi'{}^i]^\dagger = (\chi'{}^i)^\dagger \psi_i = (\bar{\sigma}^{ij})^\ast \chi^\dagger_j \psi_i = \chi^\dagger_j (\bar{\sigma}^\dagger)^{ji} \psi_i , $$ where now confusingly the symbol $\dagger$ is also used to denote the composition of complex and conjugation of the coefficient matrix $\bar{\sigma}$.

An algebra with such a map is known as a C*-algebra.

  • $\begingroup$ I still don't get it. The statement in Dirac's notation seems to refer on the second equality, but according as it is written it seems to be that $[\overline{\sigma}^{\mu\dot{a}c}]^{\dagger} =(\overline{\sigma}^{\mu\dot{c}a})^{\dagger}$ and I actually don't see why. $\endgroup$ – Frederic Thomas Sep 1 '18 at 15:46
  • $\begingroup$ I thought about this some more and have updated the answer accordingly. This should clear up your confusion. $\endgroup$ – yalda Sep 2 '18 at 7:39
  • $\begingroup$ There was a second comment from you I can't find anymore. In that comment there was a proposal wanted to check out, but haven't had the time to do it before it disappeared. $\endgroup$ – Frederic Thomas Sep 5 '18 at 12:56
  • $\begingroup$ In my previous comment I suggested you consider the following: Let $a,b \in \mathbb{R}^n$ and $M$ be an $n\times n$ real matrix. Then you can check easily that $a^T M b = b^T M^T a$. I wanted to call to your attention that it does not make sense to ask what $(a^T M b)^T$ is, since it is a scalar. You can go along these arguments to understand the formula, but I suggest that you try to understand my updated answer. $\endgroup$ – yalda Sep 7 '18 at 5:22

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.