MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

If C is the symmetric monoidal category I know that the morphisms of Repr(C) are the monoidal natural transformation between functors from C to Vect.

I've read that these transformations are linear at least in the case of C=nCob but I can't figure out why and if it is trye on a general level.

share|cite|improve this question

This is generally (always?) false. Suppose you are working over a field $\mathbb K$. Let $G$ be a finite group, and $C$ the category of $G$-sets. Then the category of symmetric monoidal functors $G \to \mathrm{Vect}_{\mathbb K}$ and monoidal natural transformations is a groupoid, canonically equivalent to the groupoid of $G$-torsors over $\mathbb K$. Put another way, let $\mathrm{Gal}(\mathbb K)$ denote the absolute separable Galois group over $\mathbb K$; then the category of symmetric monoidal functors $C \to \mathrm{Vect}_{\mathbb K}$ is equivalent to the groupoid whose objects are homomorphisms $\mathrm{Gal}(\mathbb K)\to G$; an isomorphism between $x,y: \mathrm{Gal}(\mathbb K)\to G$ is an element $g\in G$ such that $x = \mathrm{ad}_g \circ y$, where $\mathrm{ad}_g$ is the inner automorphism of $G$ given by conjugation by $g$. In particular, when $\mathbb K = \mathbb C$, this groupoid is equivalent to the groupoid with one object and $G$ morphisms. For related examples, see Deligne's excellent papers on Tannakian categories.

Indeed, it is not linear even when $C = n\mathrm{Cob}$. Consider the case $n=1$; then a symmetric monoidal functor is the same data as a finite-dimensional vector space. But what are the monoidal natural transformations? You can work out that they are just the isomorphisms. When $n=2$, you get commutative Frobenius algebras; by no stretch of the imagination do these have a linear space of morphisms, but in fact you don't even get all the homomorphisms, just the isomorphisms. In general, Lurie's theorem says that for the fully-extended $n$-dimensional cobordism category, functors from it are the same as "$n$-dualizable" objects in your target category, and symmetric monoidal natural transformations are the same as isomorphisms (or, rather, equivalences in whatever higher-categorical sense is most natural).

That said, it is not always true that the category of symmetric monoidal functors between two symmetric monoidal categories is a groupoid. If $C$ is the category of $M$-sets for some finite monoid and $\mathbb K = \mathbb C$, then the category of symmetric monoidal functors $C \to \mathrm{Vect}$ is naturally equivalent to the category with one object and morphisms $M$.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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