### 2-Monoid of Observables on String-G

#### Posted by Urs Schreiber

The baby version of the Freed-Hopkins-Teleman result, as explained by Simon Willerton, suggests that we should be thinking of the modular tensor category

that govers $G$-Chern Simons theory and $G$ Wess-Zumino-Witten theory rather in terms of the representation category

of a central extension of the action groupoid

of the adjoint action of $G$ on itself.

This monoidal category should arise as the *2-monoid of observables* # that acts on the 2-space of states over a point as we consider the 3-particle propagating on a target space that resembles $B G$.

In turn, this 2-monoid of observables should arise # as the endomorphisms of the trivial transport on target space

Here I would like to show that when we model target space as

where $\mathrm{String}_G$ is the strict string 2-group #, coming from the crossed module # $\hat \Omega_k G \to P G$, then sections on configuration space of the boundary of the 3-particle form a module category for

As discussed elsewhere (currently at the end of these notes), this should imply that states over a point are a module for

This proof is a slight variation of that of prop. 4 in the notes on sections of 2-reps. The difference is that there I considered just the 2-particle (“string”), whereas now I am looking at the boundary of the 3-particle (“membrane”).

This may sound like almost the same thing, but the extended configuration spaces differ in both situations.

For the closed 2-particle propagating on $\Sigma(G_2)$, which we model by

the configuration space is the sub-2-category

whose morphisms are restricted to describe only rotations of the string, but no true propagation.

On the other hand, what is propagation for the string is, in this sense, just reparameterization of the membrane. Hence for $\mathrm{par}$ considered as the boundary of the 3-particle, we take all of

Or so I propose. I am trying to check this proposed formalism against examples, that’s what this here is about.

So I think I checked the following two statements:

**proposition:**
*For
$1 : \Sigma(G_2) \to \mathrm{Bim}$
the trivial 2-rep of the strict 2-group $G_2$, and for
$1_* : \mathrm{conf} \to [\mathrm{par},\mathrm{Bim}]$
its transgression to configuration space, we get
$\mathrm{End}(1_*) \simeq \Lambda \mathrm{Rep}(\Lambda G_2)$
as an equivalence of monoidal categories.
*

Here $\Lambda G_2$ is the **loop 1-groupoid** of the 2-group $G_2$. This is a slight generalization of the loop groupoid $\Lambda G$ of an ordinary group as used by Simon Willerton. $\Lambda G_2$ is obtained from $\mathrm{conf} = [\Sigma(\mathbb{Z}),\Sigma(G_2)]$ by identifying isomorphic 1-morphisms.

To make use of that, we need to figure out how the loop groupoid of the String 2-group is like.

**proposition**
*The loop groupoid $\Lambda \mathrm{String}_G$ of the String 2-group is a central extension of the loop groupoid of $G$ itself *

I didn’t try to determine exactly which central extension it is. But there is only one plausible candidate: it should be the central extension obtained from the 3-cocycle on $G$ at level $k$ - that which governs $\hat \Omega G$ and hence $\mathrm{String}_G$ - transgressed from $G$ to $\Lambda G$, as described in Simon Willerton’s paper.

You can find the details here:

$\;\;\;$ 2-Monoid of Observables of $\mathrm{String}_G$.

This is in a sense just an addendum to

$\;\;\;$ sections of 2-reps.

I’d consider this result to be encouraging. In as far as it is meaningful, the next thing one would like to do is to pass from the configuration space of just the boundary of the 3-particle to the entire 3-particle. This should go along with passing from target space a (suspended) 2-group to a full 3-group - as discussed here and here.

## Rep(Lambda String_G) and K-theory

I am beginning to suspect that the representation ring of the groupoid that I called

is indeed the twisted equivariant K-theory of $G$.

I have included a discussion of that at the end of my above notes.

It amounts to observing that the groupoid $\Lambda \mathrm{String}_G$ is

generatedfrom $\mathrm{String}_G$ (regarded as just a groupoid) together with the action groupoid of $G$ acting on $P G$.But notice that, as a groupoid, $\mathrm{String}_G$ is nothing but the canonical bundle gerbe # on $G$, and that a groupoid representation of $\mathrm{String}_G$ (i.e. suitable functor $\mathrm{String}_G \to \mathrm{Vect}$) is nothing but a gerbe module #, otherwise known as a twisted bundle.

So from the fact that $\Lambda\mathrm{String}_G$ is generated from $\mathrm{String}_G$ and the action groupoid of the adjoint action, it follows that a representation of $\Lambda \mathrm{String}_G$ is in particular a gerbe module - and one equipped with the additional structure that makes it also a representation of the action groupoid. But means nothing but that it must be an equivariant gerbe module.

If follows that

is a category of equivariant twisted bundles on $G$.

I still need to check some literature, but this should say that the representation ring coming from $\mathrm{Rep}(\Lambda \mathrm{String}_G)$ is nothing but the twisted equivariant K-theory of $G$.

If there is anyone reading this who could confirm this, or else point out where I went wrong, I’d greatly appreciate it.