An application of the \(h\)-principle to manifold calculus (Q777237)

Let \(M\) be a smooth manifold, and let \({\mathcal{O}}\) denote the poset of open subsets of \(M\). Manifold calculus is a form of functor calculus that studies contravariant homotopy functors \(F\colon {\mathcal{O}}\to\mathrm{Top}\). To such a functor \(F\) it associates a Taylor tower of functors and natural transformations \[ F(-)\to \bigl( {\mathcal{T}}_\infty F(-)\to \cdots \to {\mathcal{T}}_k F(-)\to{\mathcal{T}}_{k-1} F(-)\to\cdots \to {\mathcal{T}}_0 F(-)\bigr), \] where \({\mathcal{T}}_\infty(F)\) denotes the inverse limit of the tower. The tower is said to converge if the map \(F\to {\mathcal{T}}_\infty F\) is a weak homotopy equivalence. In this case \(F\) is called an analytic functor. Manifold calculus was originally introduced to study the embedding functor \(\mathrm{Emb}(M,N)\), where \(M\) and \(N\) are smooth manifolds. An important result, proved by \textit{T. G. Goodwillie} and \textit{M. Weiss} [Geom. Topol. 3, 103--118 (1999; Zbl 0927.57028)] and \textit{T. G. Goodwillie} and \textit{J. R. Klein} [J. Topol. 1, No. 4, 761--803 (2008; Zbl 1168.57017)] says that \(\mathrm{Emb}(M,N)\) is analytic if \(\mathrm{dim}(N) -\mathrm{dim}(M)\geq 3\). In the paper under review, the author applies manifold calculus to study functors that arise from symplectic geometry. To a symplectic manifold \(N\) one can associate its Fukaya category whose objects are Lagrangian submanifolds of \(N\). A submanifold \(M\) of \(N\) is called totally real, if for each point \(p\) in \(M\), \(T_pM\) is a totally real subspace of \(T_pN\). The author defines the Lagrangian embeddings functor \(\mathrm{Emb}_\mathrm{Lag}(M,N)\) and the totally real embeddings functor \(\mathrm{Emb}_\mathrm{TR}(M,N)\). He then uses the technique of the \(h\)-principle to show that the analytic approximation of \(\mathrm{Emb}_\mathrm{Lag}(M,N)\) is weakly homotopy equivalent to \(\mathrm{Emb}_\mathrm{TR}(M,N)\), if \(\mathrm{dim}(N)-\mathrm{dim}(M)\geq 3\). Let \(\mathrm{Gr}(m, TN)\) denote the \(m\)-plane Grassmannian bundle over \(N\), and let \({\mathcal{A}}\) be a subset of \(\mathrm{Gr}(m,TN)\). An embedding \(e\colon M \hookrightarrow N\) is called \({\mathcal{A}}\)-directed, if the image of the induced map \(\mathrm{Gr}(m,e)\colon M\to \mathrm{Gr}(m, TN)\) lies in \({\mathcal{A}}\). The author proves that if \(\mathrm{dim}(N)-\mathrm{dim}(M)\geq 3\) and the \(h\)-principle holds for \({\mathcal{A}}\)-directed embeddings, then the space \(\mathrm{Emb}_{\mathcal{A}}(M,N)\) is weakly homotopy equivalent to its analytic approximation \({\mathcal{T}}_\infty\mathrm{Emb}_{\mathcal{A}}(M,N)\).