ScalingStacks

6.1. Topological Field Theory from perfect stacks

Let 2​Cob2{\rm Cob} denote the ∞\infty-category whose objects are compact oriented 11-manifolds, and whose morphisms 2​Cob​(C1,C2)2{\rm Cob}(C_{1},C_{2}) consist of the classifying spaces of oriented topological surfaces with fixed incoming and outgoing components C1,C2C_{1},C_{2}. This ∞\infty-category has a symmetric monoidal structure given by disjoint union of 1-manifolds. We will abuse notation by denoting the object consisting of the disjoint union of mm copies of S1S^{1} by mm.

0NYQ

Definition 6.1. A 2-dimensional topological field theory valued in a symmetric monoidal ∞\infty-category 𝒞\mathcal{C} is a symmetric monoidal functor F:2​Cob→𝒞F:2{\rm Cob}\to\mathcal{C}.

Fix a base commutative derived ring kk.

0NYR

Definition 6.2. Let Cat∞ex{\rm Cat}_{\infty}^{\rm ex} denote the (∞,2)(\infty,2)-category of stable, presentable kk-linear ∞\infty-categories, with 1-morphisms consisting of continuous exact functors.

Note that Cat∞ex{\rm Cat}_{\infty}^{\rm ex} has a symmetric monoidal structure, the tensor product of presentable stable categories as studied in [L4, 4.2], and the unit of the monoidal structure is the ∞\infty-category of kk-modules. Since the empty 11-manifold is the unit of 2​Cob2{\rm Cob}, a topological field theory 𝒵\mathcal{Z} sends its endomorphisms, which are closed surfaces Σ\Sigma, to endomorphisms of the unit of the target ∞\infty-category 𝒞\mathcal{C}. When 𝒞=Cat∞ex\mathcal{C}={\rm Cat}_{\infty}^{\rm ex}, the unit is Modk\Mod_{k} and 𝒵⁡(Σ)\mathcal{Z}(\Sigma) is a kk-module.

0NYS

Proposition 6.3. For a perfect stack XX, there is a 2d TFT 𝒵X:2​Cob→Cat∞ex\mathcal{Z}_{X}:2{\rm Cob}\to{\rm Cat}_{\infty}^{\rm ex} with 𝒵X​(S1)=QC⁡(ℒ​X)\mathcal{Z}_{X}(S^{1})=\qc(\mathcal{L}X) and 𝒵X​(Σ)=Γ⁡(XΣ,𝒪XΣ)\mathcal{Z}_{X}(\Sigma)=\Gamma(X^{\Sigma},\mathcal{O}_{X^{\Sigma}}), for closed surfaces Σ\Sigma.

0NYT

Proof. We define 𝒵X\mathcal{Z}_{X} on objects by assigning 𝒵X​(m)=QC⁡((ℒ​X)m)\mathcal{Z}_{X}(m)=\qc((\mathcal{L}X)^{m}).

To define 𝒵X\mathcal{Z}_{X} on morphisms, observe first that since XX is perfect, (ℒ​X)m(\mathcal{L}X)^{m} is perfect, and for Σ∈2​Cob∘​(m,n)\Sigma\in 2{\rm Cob}^{\circ}(m,n), the mapping stack XΣ=Map⁡(Σ,X)X^{\Sigma}=\Map(\Sigma,X) is also perfect (special cases of Corollary 3.25). Now for each Σ∈2​Cob∘​(m,n)\Sigma\in 2{\rm Cob}^{\circ}(m,n), consider the correspondence

XΣ\textstyle{\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces X^{\Sigma}}(ℒ​X)m\textstyle{(\mathcal{L}X)^{m}}(ℒ​X)n\textstyle{(\mathcal{L}X)^{n}}

Since all of the stacks involved are perfect, pullback and pushforward of quasi-coherent sheaves along this correspondence defines a colimit-preserving functor

𝒵X​(Σ):QC⁡((ℒ​X)m)→QC⁡((ℒ​X)n).\mathcal{Z}_{X}(\Sigma):\qc((\mathcal{L}X)^{m})\to\qc((\mathcal{L}X)^{n}).

Thus applying this construction in families, we obtain a map of spaces

𝒵X​(m,n):2​Cob∘​(m,n)→Fun⁡(QC⁡((ℒ​X)m),QC⁡((ℒ​X)n)).\mathcal{Z}_{X}(m,n):2{\rm Cob}^{\circ}(m,n)\to\Fun(\qc((\mathcal{L}X)^{m}),\qc((\mathcal{L}X)^{n})).

Suppose now that Σ=Σ1​∐∐kS1Σ2∈2​Cob​(m,n)\Sigma=\Sigma_{1}\coprod_{\coprod_{k}{S^{1}}}\Sigma_{2}\in 2{\rm Cob}(m,n) is obtained by sewing two surfaces Σ1∈2​Cob​(m,k)\Sigma_{1}\in 2{\rm Cob}(m,k) and Σ2∈2​Cob​(k,n)\Sigma_{2}\in 2{\rm Cob}(k,n). Then we have a diagram of correspondences

XΣ\textstyle{X^{\Sigma}\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces}XΣ1\textstyle{\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces X^{\Sigma_{1}}}XΣ2\textstyle{X^{\Sigma_{2}}\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces}(ℒ​X)m\textstyle{(\mathcal{L}X)^{m}}(ℒ​X)k\textstyle{(\mathcal{L}X)^{k}}(ℒ​X)n\textstyle{(\mathcal{L}X)^{n}}

Since all of the stacks involved are perfect, base change provides a canonical equivalence of functors

𝒵X​(Σ)≃𝒵X​(Σ2)∘𝒵X​(Σ1):QC⁡((ℒ​X)m)→QC⁡((ℒ​X)n).\mathcal{Z}_{X}(\Sigma)\simeq\mathcal{Z}_{X}(\Sigma_{2})\circ\mathcal{Z}_{X}(\Sigma_{1}):\qc((\mathcal{L}X)^{m})\to\qc((\mathcal{L}X)^{n}).

Similar diagrams define the higher compositions.

To complete the construction, note that 𝒵X\mathcal{Z}_{X} comes equipped with a canonical symmetric monoidal structure. Namely, by Theorem 4.7, there is a canonical equivalence

𝒵X​(m)=QC⁡((ℒ​X)m)≃QC⁡(ℒ​X)⊗m=𝒵X​(1)⊗m,\mathcal{Z}_{X}(m)=\qc((\mathcal{L}X)^{m})\simeq\qc(\mathcal{L}X)^{\otimes m}=\mathcal{Z}_{X}(1)^{\otimes m},

and it clearly extends to a symmetric monoidal structure. ∎

As an example, take X=B​GX=BG (in characteristic zero). Then 𝒵B​G​(S1)\mathcal{Z}_{BG}(S^{1}) is the ∞\infty-category of sheaves on the adjoint quotient G/G=LocG​(S1)G/G={\rm Loc}_{G}(S^{1}), while 𝒵B​G​(Σ)\mathcal{Z}_{BG}(\Sigma) is the cohomology of the structure sheaf of the moduli stack B​GΣ=LocG​(Σ)BG^{\Sigma}={\rm Loc}_{G}(\Sigma) of GG-local systems on Σ\Sigma.

More generally, the particular topological field theory operations we are considering are not sensitive to the structure of manifolds, and the same construction provides a field theory living over “bordisms” of spaces. In what follows, by a space we will mean a space homotopy equivalent to a finite simplicial set.

0NYU

Definition 6.4. Let BordSpaces{\rm Bord}_{\rm Spaces} denote the bordism (∞,1)(\infty,1)-category of spaces, with 1-morphisms from UU to VV given by spaces TU​VT_{UV} with maps U→TU​V←VU\rightarrow T_{UV}\leftarrow V, and composition of T:U→VT:U\to V and T′:V→WT^{\prime}:V\to W given by the homotopy pushout of spaces

T′∘T=T∐VT′.T^{\prime}\circ T=T\amalg_{V}T^{\prime}.

We endow BordSpaces{\rm Bord}_{\rm Spaces} with a symmetric monoidal structure given by disjoint union.

The proof of the previous proposition immediately extends to give the following:

0NYV

Proposition 6.5. For any perfect stack XX, there is a symmetric monoidal functor

SX:BordSpaces→Cat∞exS_{X}:{\rm Bord}_{\rm Spaces}\to{\rm Cat}_{\infty}^{\rm ex}

with SX​(U)=QC⁡(XU)S_{X}(U)=\qc(X^{U}) and SX(T:U→V)S_{X}(T:U\to V) the functor QC⁡(U)→QC⁡(V)\qc(U)\to\qc(V) given by pullback and pushforward along the correspondence

XU←XT→XV.X^{U}\leftarrow X^{T}\rightarrow X^{V}.

The 2d topological field theory 𝒵X\mathcal{Z}_{X} above is a categorified analogue of the 2d TFTs defined by string topology on a compact oriented manifold, or the topological B-model defined by a Calabi-Yau variety [C]. Namely, we assign to the circle the ∞\infty-category of quasi-coherent sheaves, rather than the complex of chains, on the loop space. This has the advantage that we may construct maps for arbitrary correspondences without the assumption that XX is a Calabi-Yau or satisfies Poincaré duality. On the other hand, we have gone up a level of categoricity, pushing off the problems of duality and orientation to the definition of operations between Γ⁡(XΣ,𝒪XΣ)\Gamma(X^{\Sigma},\mathcal{O}_{X^{\Sigma}}) for cobordisms between surfaces or invariants for 3-manifolds. Recall [BK] that an extended 3-dimensional topological field theory assigns a braided (in fact ribbon) category to the circle. This category must however satisfy a strong finiteness and nondegeneracy condition (modularity) coming from its extension to three-manifolds. It would be interesting to see what conditions need to be imposed on the sheaves we consider and on the stack XX in order to extend 𝒵X\mathcal{Z}_{X} in such a fashion. Note that Corollary 4.8 shows that QC⁡(X)\qc(X) for XX perfect satisfies an analogue of the Calabi-Yau or Frobenius property: namely it is self-dual as a Modk\Mod_{k}-module.

Original mathematics by the credited authors. Source collection and HTML conversion remain in progress.

David Ben-Zvi, John Francis, David Nadler

Original source: arXiv:0805.0157v5