author | Kevin Walker <kevin@canyon23.net> |
Thu, 03 Jun 2010 18:42:08 -0700 | |
changeset 330 | 8dad3dc7023b |
parent 327 | d163ad9543a5 |
child 329 | eb03c4a92f98 |
permissions | -rw-r--r-- |
215 | 1 |
%!TEX root = ../blob1.tex |
2 |
||
3 |
\section{TQFTs via fields} |
|
4 |
\label{sec:fields} |
|
5 |
\label{sec:tqftsviafields} |
|
6 |
||
319
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
7 |
In this section we review the notion of a ``system of fields and local relations". |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
8 |
For more details see \cite{kw:tqft}. From a system of fields and local relations we can readily construct TQFT invariants of manifolds. This is described in \S \ref{sec:constructing-a-TQFT}. A system of fields is very closely related to an $n$-category. In Example \ref{ex:traditional-n-categories(fields)}, which runs throughout this section, we sketch the construction of a system of fields from an $n$-category. We make this more precise for $n=1$ or $2$ in \S \ref{sec:example:traditional-n-categories(fields)}, and much later, after we've have given our own definition of a `topological $n$-category' in \S \ref{sec:ncats}, we explain precisely how to go back and forth between a topological $n$-category and a system of fields and local relations. |
215 | 9 |
|
10 |
We only consider compact manifolds, so if $Y \sub X$ is a closed codimension 0 |
|
11 |
submanifold of $X$, then $X \setmin Y$ implicitly means the closure |
|
12 |
$\overline{X \setmin Y}$. |
|
13 |
||
14 |
||
15 |
\subsection{Systems of fields} |
|
16 |
||
17 |
Let $\cM_k$ denote the category with objects |
|
18 |
unoriented PL manifolds of dimension |
|
19 |
$k$ and morphisms homeomorphisms. |
|
20 |
(We could equally well work with a different category of manifolds --- |
|
21 |
oriented, topological, smooth, spin, etc. --- but for definiteness we |
|
22 |
will stick with unoriented PL.) |
|
23 |
||
319
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
24 |
Fix a symmetric monoidal category $\cS$ whose objects are sets. While reading the definition, you should just think about the cases $\cS = \Set$ or $\cS = \Vect$. |
215 | 25 |
|
26 |
A $n$-dimensional {\it system of fields} in $\cS$ |
|
27 |
is a collection of functors $\cC_k : \cM_k \to \Set$ for $0 \leq k \leq n$ |
|
28 |
together with some additional data and satisfying some additional conditions, all specified below. |
|
29 |
||
319
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
30 |
Before finishing the definition of fields, we give two motivating examples of systems of fields. |
215 | 31 |
|
313 | 32 |
\begin{example} |
33 |
\label{ex:maps-to-a-space(fields)} |
|
319
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
34 |
Fix a target space $T$, and let $\cC(X)$ be the set of continuous maps |
327 | 35 |
from $X$ to $T$. |
313 | 36 |
\end{example} |
215 | 37 |
|
313 | 38 |
\begin{example} |
39 |
\label{ex:traditional-n-categories(fields)} |
|
40 |
Fix an $n$-category $C$, and let $\cC(X)$ be |
|
215 | 41 |
the set of sub-cell-complexes of $X$ with codimension-$j$ cells labeled by |
42 |
$j$-morphisms of $C$. |
|
43 |
One can think of such sub-cell-complexes as dual to pasting diagrams for $C$. |
|
319
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
44 |
This is described in more detail in \S \ref{sec:example:traditional-n-categories(fields)}. |
313 | 45 |
\end{example} |
215 | 46 |
|
47 |
Now for the rest of the definition of system of fields. |
|
48 |
\begin{enumerate} |
|
49 |
\item There are boundary restriction maps $\cC_k(X) \to \cC_{k-1}(\bd X)$, |
|
50 |
and these maps are a natural |
|
51 |
transformation between the functors $\cC_k$ and $\cC_{k-1}\circ\bd$. |
|
52 |
For $c \in \cC_{k-1}(\bd X)$, we will denote by $\cC_k(X; c)$ the subset of |
|
53 |
$\cC(X)$ which restricts to $c$. |
|
54 |
In this context, we will call $c$ a boundary condition. |
|
55 |
\item The subset $\cC_n(X;c)$ of top fields with a given boundary condition is an object in our symmetric monoidal category $\cS$. (This condition is of course trivial when $\cS = \Set$.) If the objects are sets with extra structure (e.g. $\cS = \Vect$ or $\Kom$), then this extra structure is considered part of the definition of $\cC_n$. Any maps mentioned below between top level fields must be morphisms in $\cS$. |
|
56 |
\item $\cC_k$ is compatible with the symmetric monoidal |
|
57 |
structures on $\cM_k$, $\Set$ and $\cS$: $\cC_k(X \du W) \cong \cC_k(X)\times \cC_k(W)$, |
|
58 |
compatibly with homeomorphisms, restriction to boundary, and orientation reversal. |
|
59 |
We will call the projections $\cC(X_1 \du X_2) \to \cC(X_i)$ |
|
60 |
restriction maps. |
|
61 |
\item Gluing without corners. |
|
62 |
Let $\bd X = Y \du -Y \du W$, where $Y$ and $W$ are closed $k{-}1$-manifolds. |
|
63 |
Let $X\sgl$ denote $X$ glued to itself along $\pm Y$. |
|
64 |
Using the boundary restriction, disjoint union, and (in one case) orientation reversal |
|
65 |
maps, we get two maps $\cC_k(X) \to \cC(Y)$, corresponding to the two |
|
66 |
copies of $Y$ in $\bd X$. |
|
67 |
Let $\Eq_Y(\cC_k(X))$ denote the equalizer of these two maps. |
|
68 |
Then (here's the axiom/definition part) there is an injective ``gluing" map |
|
69 |
\[ |
|
70 |
\Eq_Y(\cC_k(X)) \hookrightarrow \cC_k(X\sgl) , |
|
71 |
\] |
|
72 |
and this gluing map is compatible with all of the above structure (actions |
|
73 |
of homeomorphisms, boundary restrictions, orientation reversal, disjoint union). |
|
74 |
Furthermore, up to homeomorphisms of $X\sgl$ isotopic to the identity, |
|
75 |
the gluing map is surjective. |
|
76 |
From the point of view of $X\sgl$ and the image $Y \subset X\sgl$ of the |
|
77 |
gluing surface, we say that fields in the image of the gluing map |
|
78 |
are transverse to $Y$ or splittable along $Y$. |
|
79 |
\item Gluing with corners. |
|
80 |
Let $\bd X = Y \cup -Y \cup W$, where $\pm Y$ and $W$ might intersect along their boundaries. |
|
81 |
Let $X\sgl$ denote $X$ glued to itself along $\pm Y$. |
|
82 |
Note that $\bd X\sgl = W\sgl$, where $W\sgl$ denotes $W$ glued to itself |
|
83 |
(without corners) along two copies of $\bd Y$. |
|
84 |
Let $c\sgl \in \cC_{k-1}(W\sgl)$ be a be a splittable field on $W\sgl$ and let |
|
85 |
$c \in \cC_{k-1}(W)$ be the cut open version of $c\sgl$. |
|
86 |
Let $\cC^c_k(X)$ denote the subset of $\cC(X)$ which restricts to $c$ on $W$. |
|
87 |
(This restriction map uses the gluing without corners map above.) |
|
88 |
Using the boundary restriction, gluing without corners, and (in one case) orientation reversal |
|
89 |
maps, we get two maps $\cC^c_k(X) \to \cC(Y)$, corresponding to the two |
|
90 |
copies of $Y$ in $\bd X$. |
|
91 |
Let $\Eq^c_Y(\cC_k(X))$ denote the equalizer of these two maps. |
|
92 |
Then (here's the axiom/definition part) there is an injective ``gluing" map |
|
93 |
\[ |
|
94 |
\Eq^c_Y(\cC_k(X)) \hookrightarrow \cC_k(X\sgl, c\sgl) , |
|
95 |
\] |
|
96 |
and this gluing map is compatible with all of the above structure (actions |
|
97 |
of homeomorphisms, boundary restrictions, orientation reversal, disjoint union). |
|
98 |
Furthermore, up to homeomorphisms of $X\sgl$ isotopic to the identity, |
|
99 |
the gluing map is surjective. |
|
100 |
From the point of view of $X\sgl$ and the image $Y \subset X\sgl$ of the |
|
101 |
gluing surface, we say that fields in the image of the gluing map |
|
102 |
are transverse to $Y$ or splittable along $Y$. |
|
103 |
\item There are maps $\cC_{k-1}(Y) \to \cC_k(Y \times I)$, denoted |
|
104 |
$c \mapsto c\times I$. |
|
105 |
These maps comprise a natural transformation of functors, and commute appropriately |
|
106 |
with all the structure maps above (disjoint union, boundary restriction, etc.). |
|
107 |
Furthermore, if $f: Y\times I \to Y\times I$ is a fiber-preserving homeomorphism |
|
108 |
covering $\bar{f}:Y\to Y$, then $f(c\times I) = \bar{f}(c)\times I$. |
|
109 |
\end{enumerate} |
|
110 |
||
111 |
There are two notations we commonly use for gluing. |
|
112 |
One is |
|
113 |
\[ |
|
114 |
x\sgl \deq \gl(x) \in \cC(X\sgl) , |
|
115 |
\] |
|
116 |
for $x\in\cC(X)$. |
|
117 |
The other is |
|
118 |
\[ |
|
119 |
x_1\bullet x_2 \deq \gl(x_1\otimes x_2) \in \cC(X\sgl) , |
|
120 |
\] |
|
121 |
in the case that $X = X_1 \du X_2$, with $x_i \in \cC(X_i)$. |
|
122 |
||
123 |
\medskip |
|
124 |
||
125 |
Using the functoriality and $\cdot\times I$ properties above, together |
|
126 |
with boundary collar homeomorphisms of manifolds, we can define the notion of |
|
127 |
{\it extended isotopy}. |
|
128 |
Let $M$ be an $n$-manifold and $Y \subset \bd M$ be a codimension zero submanifold |
|
129 |
of $\bd M$. |
|
130 |
Let $x \in \cC(M)$ be a field on $M$ and such that $\bd x$ is splittable along $\bd Y$. |
|
131 |
Let $c$ be $x$ restricted to $Y$. |
|
132 |
Let $M \cup (Y\times I)$ denote $M$ glued to $Y\times I$ along $Y$. |
|
133 |
Then we have the glued field $x \bullet (c\times I)$ on $M \cup (Y\times I)$. |
|
134 |
Let $f: M \cup (Y\times I) \to M$ be a collaring homeomorphism. |
|
135 |
Then we say that $x$ is {\it extended isotopic} to $f(x \bullet (c\times I))$. |
|
136 |
More generally, we define extended isotopy to be the equivalence relation on fields |
|
137 |
on $M$ generated by isotopy plus all instance of the above construction |
|
138 |
(for all appropriate $Y$ and $x$). |
|
139 |
||
140 |
\nn{should also say something about pseudo-isotopy} |
|
141 |
||
142 |
||
143 |
\nn{remark that if top dimensional fields are not already linear |
|
144 |
then we will soon linearize them(?)} |
|
145 |
||
319
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
146 |
For top dimensional ($n$-dimensional) manifolds, we're actually interested |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
147 |
in the linearized space of fields. |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
148 |
By default, define $\lf(X) = \c[\cC(X)]$; that is, $\lf(X)$ is |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
149 |
the vector space of finite |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
150 |
linear combinations of fields on $X$. |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
151 |
If $X$ has boundary, we of course fix a boundary condition: $\lf(X; a) = \c[\cC(X; a)]$. |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
152 |
Thus the restriction (to boundary) maps are well defined because we never |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
153 |
take linear combinations of fields with differing boundary conditions. |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
154 |
|
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
155 |
In some cases we don't linearize the default way; instead we take the |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
156 |
spaces $\lf(X; a)$ to be part of the data for the system of fields. |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
157 |
In particular, for fields based on linear $n$-category pictures we linearize as follows. |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
158 |
Define $\lf(X; a) = \c[\cC(X; a)]/K$, where $K$ is the space generated by |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
159 |
obvious relations on 0-cell labels. |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
160 |
More specifically, let $L$ be a cell decomposition of $X$ |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
161 |
and let $p$ be a 0-cell of $L$. |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
162 |
Let $\alpha_c$ and $\alpha_d$ be two labelings of $L$ which are identical except that |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
163 |
$\alpha_c$ labels $p$ by $c$ and $\alpha_d$ labels $p$ by $d$. |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
164 |
Then the subspace $K$ is generated by things of the form |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
165 |
$\lambda \alpha_c + \alpha_d - \alpha_{\lambda c + d}$, where we leave it to the reader |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
166 |
to infer the meaning of $\alpha_{\lambda c + d}$. |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
167 |
Note that we are still assuming that $n$-categories have linear spaces of $n$-morphisms. |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
168 |
|
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
169 |
\nn{Maybe comment further: if there's a natural basis of morphisms, then no need; |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
170 |
will do something similar below; in general, whenever a label lives in a linear |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
171 |
space we do something like this; ? say something about tensor |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
172 |
product of all the linear label spaces? Yes:} |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
173 |
|
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
174 |
For top dimensional ($n$-dimensional) manifolds, we linearize as follows. |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
175 |
Define an ``almost-field" to be a field without labels on the 0-cells. |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
176 |
(Recall that 0-cells are labeled by $n$-morphisms.) |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
177 |
To each unlabeled 0-cell in an almost field there corresponds a (linear) $n$-morphism |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
178 |
space determined by the labeling of the link of the 0-cell. |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
179 |
(If the 0-cell were labeled, the label would live in this space.) |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
180 |
We associate to each almost-labeling the tensor product of these spaces (one for each 0-cell). |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
181 |
We now define $\lf(X; a)$ to be the direct sum over all almost labelings of the |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
182 |
above tensor products. |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
183 |
|
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
184 |
|
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
185 |
\subsection{Systems of fields from $n$-categories} |
121c580d5ef7
editting all over the place
Scott Morrison <scott@tqft.net>
parents:
313
diff
changeset
|
186 |
\label{sec:example:traditional-n-categories(fields)} |
327 | 187 |
We now describe in more detail Example \ref{ex:traditional-n-categories(fields)}, systems of fields coming from sub-cell-complexes labeled |
215 | 188 |
by $n$-category morphisms. |
189 |
||
190 |
Given an $n$-category $C$ with the right sort of duality |
|
327 | 191 |
(e.g. a pivotal 2-category, 1-category with duals, star 1-category, disklike $n$-category), |
215 | 192 |
we can construct a system of fields as follows. |
193 |
Roughly speaking, $\cC(X)$ will the set of all embedded cell complexes in $X$ |
|
194 |
with codimension $i$ cells labeled by $i$-morphisms of $C$. |
|
195 |
We'll spell this out for $n=1,2$ and then describe the general case. |
|
196 |
||
197 |
If $X$ has boundary, we require that the cell decompositions are in general |
|
198 |
position with respect to the boundary --- the boundary intersects each cell |
|
327 | 199 |
transversely, so cells meeting the boundary are mere half-cells. Put another way, the cell decompositions we consider are dual to standard cell |
215 | 200 |
decompositions of $X$. |
201 |
||
202 |
We will always assume that our $n$-categories have linear $n$-morphisms. |
|
203 |
||
204 |
For $n=1$, a field on a 0-manifold $P$ is a labeling of each point of $P$ with |
|
205 |
an object (0-morphism) of the 1-category $C$. |
|
206 |
A field on a 1-manifold $S$ consists of |
|
207 |
\begin{itemize} |
|
327 | 208 |
\item a cell decomposition of $S$ (equivalently, a finite collection |
215 | 209 |
of points in the interior of $S$); |
210 |
\item a labeling of each 1-cell (and each half 1-cell adjacent to $\bd S$) |
|
211 |
by an object (0-morphism) of $C$; |
|
212 |
\item a transverse orientation of each 0-cell, thought of as a choice of |
|
213 |
``domain" and ``range" for the two adjacent 1-cells; and |
|
214 |
\item a labeling of each 0-cell by a morphism (1-morphism) of $C$, with |
|
215 |
domain and range determined by the transverse orientation and the labelings of the 1-cells. |
|
216 |
\end{itemize} |
|
217 |
||
218 |
If $C$ is an algebra (i.e. if $C$ has only one 0-morphism) we can ignore the labels |
|
219 |
of 1-cells, so a field on a 1-manifold $S$ is a finite collection of points in the |
|
220 |
interior of $S$, each transversely oriented and each labeled by an element (1-morphism) |
|
221 |
of the algebra. |
|
222 |
||
223 |
\medskip |
|
224 |
||
225 |
For $n=2$, fields are just the sort of pictures based on 2-categories (e.g.\ tensor categories) |
|
226 |
that are common in the literature. |
|
227 |
We describe these carefully here. |
|
228 |
||
229 |
A field on a 0-manifold $P$ is a labeling of each point of $P$ with |
|
230 |
an object of the 2-category $C$. |
|
231 |
A field of a 1-manifold is defined as in the $n=1$ case, using the 0- and 1-morphisms of $C$. |
|
232 |
A field on a 2-manifold $Y$ consists of |
|
233 |
\begin{itemize} |
|
327 | 234 |
\item a cell decomposition of $Y$ (equivalently, a graph embedded in $Y$ such |
215 | 235 |
that each component of the complement is homeomorphic to a disk); |
236 |
\item a labeling of each 2-cell (and each partial 2-cell adjacent to $\bd Y$) |
|
237 |
by a 0-morphism of $C$; |
|
238 |
\item a transverse orientation of each 1-cell, thought of as a choice of |
|
239 |
``domain" and ``range" for the two adjacent 2-cells; |
|
240 |
\item a labeling of each 1-cell by a 1-morphism of $C$, with |
|
241 |
domain and range determined by the transverse orientation of the 1-cell |
|
242 |
and the labelings of the 2-cells; |
|
243 |
\item for each 0-cell, a homeomorphism of the boundary $R$ of a small neighborhood |
|
244 |
of the 0-cell to $S^1$ such that the intersections of the 1-cells with $R$ are not mapped |
|
245 |
to $\pm 1 \in S^1$; and |
|
246 |
\item a labeling of each 0-cell by a 2-morphism of $C$, with domain and range |
|
247 |
determined by the labelings of the 1-cells and the parameterizations of the previous |
|
248 |
bullet. |
|
249 |
\end{itemize} |
|
250 |
\nn{need to say this better; don't try to fit everything into the bulleted list} |
|
251 |
||
252 |
For general $n$, a field on a $k$-manifold $X^k$ consists of |
|
253 |
\begin{itemize} |
|
254 |
\item A cell decomposition of $X$; |
|
255 |
\item an explicit general position homeomorphism from the link of each $j$-cell |
|
256 |
to the boundary of the standard $(k-j)$-dimensional bihedron; and |
|
257 |
\item a labeling of each $j$-cell by a $(k-j)$-dimensional morphism of $C$, with |
|
258 |
domain and range determined by the labelings of the link of $j$-cell. |
|
259 |
\end{itemize} |
|
260 |
||
261 |
%\nn{next definition might need some work; I think linearity relations should |
|
262 |
%be treated differently (segregated) from other local relations, but I'm not sure |
|
263 |
%the next definition is the best way to do it} |
|
264 |
||
265 |
\medskip |
|
266 |
||
267 |
||
268 |
||
269 |
||
270 |
\subsection{Local relations} |
|
271 |
\label{sec:local-relations} |
|
313 | 272 |
Local relations are certain subspaces of the fields on balls, which form an ideal under gluing. Again, we give the examples first. |
215 | 273 |
|
313 | 274 |
\addtocounter{prop}{-2} |
275 |
\begin{example}[contd.] |
|
276 |
For maps into spaces, $U(B; c)$ is generated by fields of the form $a-b \in \lf(B; c)$, |
|
277 |
where $a$ and $b$ are maps (fields) which are homotopic rel boundary. |
|
278 |
\end{example} |
|
215 | 279 |
|
313 | 280 |
\begin{example}[contd.] |
281 |
For $n$-category pictures, $U(B; c)$ is equal to the kernel of the evaluation map |
|
282 |
$\lf(B; c) \to \mor(c', c'')$, where $(c', c'')$ is some (any) division of $c$ into |
|
283 |
domain and range. |
|
284 |
\end{example} |
|
285 |
||
286 |
These motivate the following definition. |
|
287 |
||
288 |
\begin{defn} |
|
215 | 289 |
A {\it local relation} is a collection subspaces $U(B; c) \sub \lf(B; c)$, |
290 |
for all $n$-manifolds $B$ which are |
|
291 |
homeomorphic to the standard $n$-ball and all $c \in \cC(\bd B)$, |
|
292 |
satisfying the following properties. |
|
293 |
\begin{enumerate} |
|
294 |
\item functoriality: |
|
295 |
$f(U(B; c)) = U(B', f(c))$ for all homeomorphisms $f: B \to B'$ |
|
296 |
\item local relations imply extended isotopy: |
|
297 |
if $x, y \in \cC(B; c)$ and $x$ is extended isotopic |
|
298 |
to $y$, then $x-y \in U(B; c)$. |
|
299 |
\item ideal with respect to gluing: |
|
300 |
if $B = B' \cup B''$, $x\in U(B')$, and $c\in \cC(B'')$, then $x\bullet r \in U(B)$ |
|
301 |
\end{enumerate} |
|
313 | 302 |
\end{defn} |
303 |
See \cite{kw:tqft} for further details. |
|
215 | 304 |
|
305 |
||
306 |
\subsection{Constructing a TQFT} |
|
222
217b6a870532
committing changes from loon lake - mostly small blobs
scott@6e1638ff-ae45-0410-89bd-df963105f760
parents:
215
diff
changeset
|
307 |
\label{sec:constructing-a-tqft} |
215 | 308 |
|
309 |
In this subsection we briefly review the construction of a TQFT from a system of fields and local relations. |
|
310 |
(For more details, see \cite{kw:tqft}.) |
|
311 |
||
312 |
Let $W$ be an $n{+}1$-manifold. |
|
313 |
We can think of the path integral $Z(W)$ as assigning to each |
|
314 |
boundary condition $x\in \cC(\bd W)$ a complex number $Z(W)(x)$. |
|
315 |
In other words, $Z(W)$ lies in $\c^{\lf(\bd W)}$, the vector space of linear |
|
316 |
maps $\lf(\bd W)\to \c$. |
|
317 |
||
318 |
The locality of the TQFT implies that $Z(W)$ in fact lies in a subspace |
|
319 |
$Z(\bd W) \sub \c^{\lf(\bd W)}$ defined by local projections. |
|
320 |
The linear dual to this subspace, $A(\bd W) = Z(\bd W)^*$, |
|
321 |
can be thought of as finite linear combinations of fields modulo local relations. |
|
322 |
(In other words, $A(\bd W)$ is a sort of generalized skein module.) |
|
323 |
This is the motivation behind the definition of fields and local relations above. |
|
324 |
||
325 |
In more detail, let $X$ be an $n$-manifold. |
|
326 |
%To harmonize notation with the next section, |
|
327 |
%let $\bc_0(X)$ be the vector space of finite linear combinations of fields on $X$, so |
|
328 |
%$\bc_0(X) = \lf(X)$. |
|
329 |
Define $U(X) \sub \lf(X)$ to be the space of local relations in $\lf(X)$; |
|
330 |
$U(X)$ is generated by things of the form $u\bullet r$, where |
|
331 |
$u\in U(B)$ for some embedded $n$-ball $B\sub X$ and $r\in \cC(X\setmin B)$. |
|
332 |
Define |
|
333 |
\[ |
|
334 |
A(X) \deq \lf(X) / U(X) . |
|
335 |
\] |
|
336 |
(The blob complex, defined in the next section, |
|
337 |
is in some sense the derived version of $A(X)$.) |
|
338 |
If $X$ has boundary we can similarly define $A(X; c)$ for each |
|
339 |
boundary condition $c\in\cC(\bd X)$. |
|
340 |
||
341 |
The above construction can be extended to higher codimensions, assigning |
|
342 |
a $k$-category $A(Y)$ to an $n{-}k$-manifold $Y$, for $0 \le k \le n$. |
|
343 |
These invariants fit together via actions and gluing formulas. |
|
344 |
We describe only the case $k=1$ below. |
|
345 |
(The construction of the $n{+}1$-dimensional part of the theory (the path integral) |
|
346 |
requires that the starting data (fields and local relations) satisfy additional |
|
347 |
conditions. |
|
348 |
We do not assume these conditions here, so when we say ``TQFT" we mean a decapitated TQFT |
|
349 |
that lacks its $n{+}1$-dimensional part.) |
|
350 |
||
351 |
Let $Y$ be an $n{-}1$-manifold. |
|
352 |
Define a (linear) 1-category $A(Y)$ as follows. |
|
353 |
The objects of $A(Y)$ are $\cC(Y)$. |
|
354 |
The morphisms from $a$ to $b$ are $A(Y\times I; a, b)$, where $a$ and $b$ label the two boundary components of the cylinder $Y\times I$. |
|
355 |
Composition is given by gluing of cylinders. |
|
356 |
||
357 |
Let $X$ be an $n$-manifold with boundary and consider the collection of vector spaces |
|
358 |
$A(X; \cdot) \deq \{A(X; c)\}$ where $c$ ranges through $\cC(\bd X)$. |
|
359 |
This collection of vector spaces affords a representation of the category $A(\bd X)$, where |
|
360 |
the action is given by gluing a collar $\bd X\times I$ to $X$. |
|
361 |
||
362 |
Given a splitting $X = X_1 \cup_Y X_2$ of a closed $n$-manifold $X$ along an $n{-}1$-manifold $Y$, |
|
363 |
we have left and right actions of $A(Y)$ on $A(X_1; \cdot)$ and $A(X_2; \cdot)$. |
|
364 |
The gluing theorem for $n$-manifolds states that there is a natural isomorphism |
|
365 |
\[ |
|
366 |
A(X) \cong A(X_1; \cdot) \otimes_{A(Y)} A(X_2; \cdot) . |
|
367 |
\] |
|
368 |