text/ncat.tex
changeset 552 c2091a3ebcc3
parent 551 9dfb5db2acd7
child 555 11532ce39ec0
equal deleted inserted replaced
551:9dfb5db2acd7 552:c2091a3ebcc3
   827 %\nn{say something about cofibrant replacements?}
   827 %\nn{say something about cofibrant replacements?}
   828 In fact, there is also a trivial, but mostly uninteresting, way to do this: 
   828 In fact, there is also a trivial, but mostly uninteresting, way to do this: 
   829 we can think of each vector space associated to an $n$-ball as a chain complex concentrated in degree $0$, 
   829 we can think of each vector space associated to an $n$-ball as a chain complex concentrated in degree $0$, 
   830 and take $\CD{B}$ to act trivially. 
   830 and take $\CD{B}$ to act trivially. 
   831 
   831 
   832 Be careful that the ``free resolution" of the topological $n$-category $\pi_{\leq n}(T)$ is not the $A_\infty$ $n$-category $\pi^\infty_{\leq n}(T)$.
   832 Beware that the ``free resolution" of the topological $n$-category $\pi_{\leq n}(T)$ 
       
   833 is not the $A_\infty$ $n$-category $\pi^\infty_{\leq n}(T)$.
   833 It's easy to see that with $n=0$, the corresponding system of fields is just 
   834 It's easy to see that with $n=0$, the corresponding system of fields is just 
   834 linear combinations of connected components of $T$, and the local relations are trivial.
   835 linear combinations of connected components of $T$, and the local relations are trivial.
   835 There's no way for the blob complex to magically recover all the data of $\pi^\infty_{\leq 0}(T) \iso C_* T$.
   836 There's no way for the blob complex to magically recover all the data of $\pi^\infty_{\leq 0}(T) \iso C_* T$.
   836 
   837 
   837 \begin{example}[The bordism $n$-category, $A_\infty$ version]
   838 \begin{example}[The bordism $n$-category, $A_\infty$ version]
   901 \end{example}
   902 \end{example}
   902 
   903 
   903 
   904 
   904 \subsection{From balls to manifolds}
   905 \subsection{From balls to manifolds}
   905 \label{ss:ncat_fields} \label{ss:ncat-coend}
   906 \label{ss:ncat_fields} \label{ss:ncat-coend}
   906 In this section we describe how to extend an $n$-category $\cC$ as described above 
   907 In this section we show how to extend an $n$-category $\cC$ as described above 
   907 (of either the plain or $A_\infty$ variety) to an invariant of manifolds, which we denote by $\cl{\cC}$.
   908 (of either the plain or $A_\infty$ variety) to an invariant of manifolds, which we denote by $\cl{\cC}$.
   908 This extension is a certain colimit, and we've chosen the notation to remind you of this.
   909 This extension is a certain colimit, and the arrow in the notation is intended as a reminder of this.
   909 Thus we show that functors $\cC_k$ satisfying the axioms above have a canonical extension 
   910 
   910 from $k$-balls to arbitrary $k$-manifolds.
       
   911 Recall that we've already anticipated this construction in the previous section, 
       
   912 inductively defining $\cl{\cC}$ on $k$-spheres in terms of $\cC$ on $k$-balls, 
       
   913 so that we can state the boundary axiom for $\cC$ on $k+1$-balls.
       
   914 In the case of plain $n$-categories, this construction factors into a construction of a 
   911 In the case of plain $n$-categories, this construction factors into a construction of a 
   915 system of fields and local relations, followed by the usual TQFT definition of a 
   912 system of fields and local relations, followed by the usual TQFT definition of a 
   916 vector space invariant of manifolds given as Definition \ref{defn:TQFT-invariant}.
   913 vector space invariant of manifolds given as Definition \ref{defn:TQFT-invariant}.
   917 For an $A_\infty$ $n$-category, $\cl{\cC}$ is defined using a homotopy colimit instead.
   914 For an $A_\infty$ $n$-category, $\cl{\cC}$ is defined using a homotopy colimit instead.
   918 Recall that we can take a plain $n$-category $\cC$ and pass to the ``free resolution", 
   915 Recall that we can take a plain $n$-category $\cC$ and pass to the ``free resolution", 
   919 an $A_\infty$ $n$-category $\bc_*(\cC)$, by computing the blob complex of balls 
   916 an $A_\infty$ $n$-category $\bc_*(\cC)$, by computing the blob complex of balls 
   920 (recall Example \ref{ex:blob-complexes-of-balls} above).
   917 (recall Example \ref{ex:blob-complexes-of-balls} above).
   921 We will show in Corollary \ref{cor:new-old} below that the homotopy colimit invariant 
   918 We will show in Corollary \ref{cor:new-old} below that the homotopy colimit invariant 
   922 for a manifold $M$ associated to this $A_\infty$ $n$-category is actually the 
   919 for a manifold $M$ associated to this $A_\infty$ $n$-category is actually the 
   923 same as the original blob complex  for $M$ with coefficients in $\cC$.
   920 same as the original blob complex for $M$ with coefficients in $\cC$.
       
   921 
       
   922 Recall that we've already anticipated this construction in the previous section, 
       
   923 inductively defining $\cl{\cC}$ on $k$-spheres in terms of $\cC$ on $k$-balls, 
       
   924 so that we can state the boundary axiom for $\cC$ on $k+1$-balls.
       
   925 
       
   926 \medskip
   924 
   927 
   925 We will first define the ``decomposition" poset $\cell(W)$ for any $k$-manifold $W$, for $1 \leq k \leq n$. 
   928 We will first define the ``decomposition" poset $\cell(W)$ for any $k$-manifold $W$, for $1 \leq k \leq n$. 
   926 An $n$-category $\cC$ provides a functor from this poset to the category of sets, 
   929 An $n$-category $\cC$ provides a functor from this poset to the category of sets, 
   927 and we  will define $\cl{\cC}(W)$ as a suitable colimit 
   930 and we  will define $\cl{\cC}(W)$ as a suitable colimit 
   928 (or homotopy colimit in the $A_\infty$ case) of this functor. 
   931 (or homotopy colimit in the $A_\infty$ case) of this functor. 
  1083 One can show that the usual hocolimit and the local hocolimit are homotopy equivalent using an 
  1086 One can show that the usual hocolimit and the local hocolimit are homotopy equivalent using an 
  1084 Eilenberg-Zilber type subdivision argument.
  1087 Eilenberg-Zilber type subdivision argument.
  1085 
  1088 
  1086 \medskip
  1089 \medskip
  1087 
  1090 
  1088 $\cl{\cC}(W)$ is functorial with respect to homeomorphisms of $k$-manifolds. Restricting the $k$-spheres, we have now proved Lemma \ref{lem:spheres}.
  1091 $\cl{\cC}(W)$ is functorial with respect to homeomorphisms of $k$-manifolds. 
       
  1092 Restricting the $k$-spheres, we have now proved Lemma \ref{lem:spheres}.
  1089 
  1093 
  1090 It is easy to see that
  1094 It is easy to see that
  1091 there are well-defined maps $\cl{\cC}(W)\to\cl{\cC}(\bd W)$, and that these maps
  1095 there are well-defined maps $\cl{\cC}(W)\to\cl{\cC}(\bd W)$, and that these maps
  1092 comprise a natural transformation of functors.
  1096 comprise a natural transformation of functors.
  1093 
  1097 
  1139 But $a'_0$ and $a'_k$ are both elements of $\psi(x'_0)$ (because $x'_k = x'_0$).
  1143 But $a'_0$ and $a'_k$ are both elements of $\psi(x'_0)$ (because $x'_k = x'_0$).
  1140 So by the injectivity clause of the composition axiom, we must have that $a'_0 = a'_k$.
  1144 So by the injectivity clause of the composition axiom, we must have that $a'_0 = a'_k$.
  1141 But this implies that $a = a_0 = a_k = \hat{a}$, contrary to our assumption that $a\ne \hat{a}$.
  1145 But this implies that $a = a_0 = a_k = \hat{a}$, contrary to our assumption that $a\ne \hat{a}$.
  1142 \end{proof}
  1146 \end{proof}
  1143 
  1147 
  1144 \nn{need to finish explaining why we have a system of fields;
  1148 %\nn{need to finish explaining why we have a system of fields;
  1145 define $k$-cat $\cC(\cdot\times W)$}
  1149 %define $k$-cat $\cC(\cdot\times W)$}
  1146 
  1150 
  1147 \subsection{Modules}
  1151 \subsection{Modules}
  1148 
  1152 
  1149 Next we define plain and $A_\infty$ $n$-category modules.
  1153 Next we define plain and $A_\infty$ $n$-category modules.
  1150 The definition will be very similar to that of $n$-categories,
  1154 The definition will be very similar to that of $n$-categories,
  2225 	f: \cS(X; c; E) \to \cS(X'; f(c); f(E)) .
  2229 	f: \cS(X; c; E) \to \cS(X'; f(c); f(E)) .
  2226 \]
  2230 \]
  2227 It is easy to show that this is independent of the choice of $E$.
  2231 It is easy to show that this is independent of the choice of $E$.
  2228 Note also that this map depends only on the restriction of $f$ to $\bd X$.
  2232 Note also that this map depends only on the restriction of $f$ to $\bd X$.
  2229 In particular, if $F: X\to X$ is the identity on $\bd X$ then $f$ acts trivially, as required by
  2233 In particular, if $F: X\to X$ is the identity on $\bd X$ then $f$ acts trivially, as required by
  2230 Axiom \ref{axiom:extended-isotopies} of \S\ref{ss:n-cat-def}.
  2234 Axiom \ref{axiom:extended-isotopies}.
  2231 
  2235 
  2232 We define product $n{+}1$-morphisms to be identity maps of modules.
  2236 We define product $n{+}1$-morphisms to be identity maps of modules.
  2233 
  2237 
  2234 To define (binary) composition of $n{+}1$-morphisms, choose the obvious common equator
  2238 To define (binary) composition of $n{+}1$-morphisms, choose the obvious common equator
  2235 then compose the module maps.
  2239 then compose the module maps.