Given a semisimple Lie algebra $\frak{g}$ over $\mathbb{C}$, and a finite dimensional irreducible representation $V$, with dual representation $V^*$, we know that the decomposition of $V \otimes V^*$ must contain a copy of the trivial module. Is it true that it will only ever contain a single copy of the trivial representation?

Does this generalise to the setting of suitably qualified semisimple rigid monodial categories? A specific guess:

For a semisimple rigid braided monoidal category $\mathcal{C}$, with $V$ a simple object in $\mathcal{C}$, and $V^*$ its dual, the decomposition of $V \otimes V^*$ into simple objects contains one and only one copy of the unit object of $\mathcal{C}$.

  • 5
    $\begingroup$ Identifying $V\otimes V^{*}\simeq \operatorname{Hom}(V,V)$, this follows from the Schur's lemma. $\endgroup$ – Victor Protsak Jun 21 '18 at 14:16

$\newcommand{\C}{\mathcal{C}}$ To expand on Victor's comment: you don't actually need $\C$ to be semi-simple (let's assume it's abelian to be safe). By definition the multiplicity of $1_C$ in $V\otimes V^*$ is the dimension of $$Hom_\C(1_\C,V\otimes V^*)\cong Hom_\C(V,V).$$

Now Schur Lemma is valid at this level of generality, so if $V$ is simple this space is 1-dimensional.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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