§ 4. Theory of descent
1. Descent data
Let X and Y be topological spaces and let be a continuous mapping. Let be an X-space.
Definition 1 0227
A descent data relative to on the X-space is called a continuous mapping satisfying the following two properties:
(i) For every pair , the mapping induces by restriction a bijection of onto ;
(ii) For every triplet of points of X such that , one has
If is a descent data on , the family , for , is thus a law of operation (on the right) of the groupoid on the X-space (II, p. 167). In particular, one has Id for every , which is also written for every . Conversely, given a law of operation (on the right) of the groupoid on , the map from into Z defined by verifies the relations (i) and (ii) of the definition.
Let X and Y be topological spaces, a continuous mapping and let be an X-space. If is a descent data relative to on , the relation defined by « and » is the equivalence relation in Z deduced from the operation of the groupoid defined by ; it is compatible with the mapping . It is said that this is the equivalence relation associated with the descent data . The canonical bijection of the graph Γ of the equivalence relation onto is a homeomorphism, the reciprocal homeomorphism applies an element to .
Conversely, let R be an equivalence relation in Z compatible with the mapping and let Γ be the graph of R. Suppose moreover that the mapping defines a homeo-morphism of Γ onto . The mapping given by pr is continuous; it is a descent data relative to on and the relation R is the equivalence relation associated with .
Example 1 0228
Let X and Y be topological spaces, let be a continuous mapping and let be a Y-space. Put . The mapping from into Z which associates with the element is a descent datum relative to on the X-space , pr, called the canonical descent datum. For and , the relation is equivalent to .
Example 2 0229
Let Y be a topological space, a family of subsets of Y, and for each , let be a -space. Let X denote the sum topological space of the family and the X-space sum of the family . Let be the canonical mapping.
The space is then identified with the sum space of the family I, p. 4, Example 5). Let be a descent datum relative to on . For each pair , define a continuous mapping by . The family satisfies the following properties:
(i) For each , one has Id;
(ii) For each pair is an isomorphism of -spaces ;
(iii) For each triple and each , one has .
Conversely, every family possessing the above properties arises from a unique descent datum relative to on .
2. Effective descent data
Let X and Y be topological spaces, a continuous mapping and an X-space. Let be a descent datum relative to on , let be the associated equivalence relation and let be the canonical mapping. Since the relation is compatible with the mapping , there exists a unique continuous mapping such that the diagram
Z
X Y
let it be a commutative square. The Y-space is called the quotient space of by the descent datum . Let us denote by the mapping defined by for . It is continuous. Let and let be such that ; then we have and the point is the unique element of Z such that ; consequently, the mapping is bijective.
One says that the descent datum relative to on is effective if diagram (1) is a cartesian square, that is to say, if the continuous bijection is a homeomorphism. In order that the descent datum be effective, it is necessary and sufficient that the sets , where U is an open subset of X and V an open subset of Z saturated for , constitute a basis of the topology of Z. In particular, the condition for a descent datum relative to to be effective is of a local nature on Y.
Example 1 022A
Let X and Y be topological spaces and a continuous mapping. Let be a Y-space. Let Z be the X-space , endowed with the mapping pr; let us denote by its canonical descent datum relative to (IV, p. 383, Example 1). The subsets of Z of the form , where V is an open subset of T, are open in Z and saturated for the relation loc. cit.). By definition of the product topology, the sets pr, where U is open in X and V is open in T, form a basis of the topology of . Consequently, the canonical descent datum on a fibred product is effective.
The canonical mapping is injective and continuous. It is not, however, necessarily surjective, nor strict (IV, p. 462, exerc. 1).
Example 2 022B
Let us take up again the notation of Example 2 (IV, p. 383). The topological space is then the topological space obtained by gluing the spaces along the by means of the bijections (TG, I, p. 16). Consequently, if for every , the set is open (resp. closed) in Y, the set is open (resp. closed) in and the restriction of to induces a homeomorphism of onto (TG, I, p. 17, Prop. 9). The space Z is the sum space of the spaces ; the space is the sum space of the spaces . The mapping is identified with the sum mapping of the mappings . It is therefore a homeomorphism, which shows that the descent datum is effective.
Without a particular assumption on the subsets , it is not always true that the restriction of to induces a homeomorphism of onto its image; in this case, the descent datum is not effective (IV, p. 462, exerc. 2).
Proposition 1 022C
Let X and Y be topological spaces, let be a continuous mapping. Suppose that every point of Y possesses a neighbourhood above which there exists a continuous section of the mapping . Then every descent datum relative to on an X-space is effective.
Let be an X-space and let be a descent datum relative to on . The assertion that is an effective descent datum is local in Y, which allows us to suppose that the mapping possesses a continuous section . Let and denote the canonical mappings, and let be the mapping given by . The mapping is bijective and continuous; it is enough to prove that it is a homeomorphism.
The mapping from Z into Z which, to , associates is continuous and maps every element of Z to the unique element of Z which is equivalent to it for the relation and such that belongs to the image of . It therefore defines, by passing to the quotient, a continuous mapping which is a section of the mapping . In particular, one has .
For every , one has ; let us then put . The resulting mapping from into Z is continuous. For every , one has
because is equivalent to for the relation . This proves that the mapping is the identity mapping of . For , one then has and , whence , by definition of the equivalence relation . Thus one has and Id. The mapping is therefore a homeomorphism, which was to be proved.
Proposition 2 022D
Let Y be a continuous mapping, let be an X-space and let be a descent datum relative to on Z. The equivalence relation is closed if is proper; it is open if is open.
The mapping X Z defined by is a homeomorphism, with inverse mapping . We have = pr, where pr Z is the second projection. If is proper, pr is proper; if is open, pr is open (I, p. 17, prop. 8). It follows that is proper (resp. open) if is. The saturation of a subset A of Z with respect to the relation is the image of under . Consequently, if is proper, the saturation of a closed subset is closed; if is open, the saturation of an open subset is open.
3. Descent of morphisms
Let X and Y be topological spaces and let be a continuous mapping of X into Y. Let and be X-spaces endowed with descent data relative to , denoted respectively by and . One says that an X-morphism is compatible with the descent data and if one has
for every . It amounts to the same to say that the images under of two equivalent points with respect to the relation are equivalent with respect to the relation . Such a morphism defines, by passing to the quotients, a continuous mapping ; it is a Y-morphism of spaces.
Let us denote by the set of X-morphisms of Z into which are compatible with the descent data and .
Proposition 3 022E
Let X and Y be topological spaces and let Y be a continuous mapping. Let and be X-spaces endowed with descent data relative to , denoted respectively by and . If the descent datum is effective, the mapping is a bijection of onto .
For every X-morphism of Z into compatible with the descent data, the mapping is a Y-morphism. Conversely, let and denote the canonical mappings, and let be a Y-morphism. The mappings and are Y-morphisms. The hypothesis that the descent datum is effective means that the diagram
X Y
is a cartesian square. Hence there exists a unique continuous mapping such that and . The first equality means that is an X-morphism, the second equality means that is compatible with the descent data and that , whence the proposition.
4. Descent: case of spread spaces
Let X and Y be topological spaces and let be a continuous mapping. Let T and be Y-spaces; endow the X-spaces and with their canonical descent data and denote by the set of X-morphisms of into which are compatible with these descent data. For every Y-morphism , the X-morphism of into is compatible with the canonical descent data. We shall denote by the resulting mapping.
Proposition 4 022F
Suppose that the mapping is strict and surjective and that T is a spread Y-space. Then the mapping is bijective.
Let (resp. ) denote the equivalence relation on (resp. on ) associated with the canonical descent datum. Since the mapping is surjective, the projection pr is surjective and the canonical mapping is bijective. In particular, the mapping is injective. Let us prove that it is surjective. Let be an X-morphism compatible with the canonical descent data. Then for , one has , where is a mapping of T into .
By definition of , the mapping pr is equal to pr, and is therefore continuous. Since the mapping is surjective and strict and T is a spread Y-space, the projection pr is strict (I, p. 32, remark 3). By proposition 9 of I, p. 18, the mapping is therefore continuous. It is a Y-morphism such that , which proves the proposition.
Corollary 022G
Let X and Y be topological spaces and let be a strict and surjective mapping. Let T and be spread Y-spaces. If there exists an X-isomorphism compatible with the canonical descent data, the Y-spaces T and are isomorphic.
Let be an X-isomorphism of spread spaces. By proposition 4, there exists a unique morphism of Y-spaces such that . Since is surjective, the mapping is bijective. It then follows from cor. 2 of I, p. 30 that is an isomorphism, since the Y-spaces T and are spread.
Proposition 5 022H
Let X and Y be topological spaces and let be a continuous mapping. Suppose that the mapping is proper and separated, or else open. Every descent datum relative to on a spread X-space is effective. Moreover, if is surjective, the quotient space is a spread Y-space.
Let be an X-étalé space and let be a descent datum relative to on . Let us denote by the equivalence relation in Z associated with , let be the canonical surjection and let denote the mapping of Z into defined by . It is continuous and bijective; let us prove that it is a homeomorphism.
a) Suppose first that the mapping is open.
Let be a point of Z; put . Since is étale, there exists a neighbourhood U of in X and a continuous section of over U such that . The mapping is étale, and the mappings of into defined by and are continuous sections of it over . Since they coincide at every point of of the form , they therefore coincide on an open neighbourhood V of contained in . Let be a neighbourhood of in X such that is contained in V. Let be a point of ; let be such that . We have therefore , whence and .
Since the mapping is open (IV, p. 386, prop. 2), the set is an open subset of on which the mapping is equal to the continuous mapping pr. The mapping is therefore a homeomorphism.
b) Suppose now that the mapping is proper and separated.
Let be a point of Z; put and . The mapping given by is a section of over .
The set is compact (TG, I, p. 75, Theorem 1 and I, p. 26, Remark 2), and two distinct points of have disjoint neighbourhoods in X, since is separated (I, p. 25, Proposition 1). By Theorem 2 of I, p. 37, there therefore exists a neighbourhood of in X and a section of over extending . The set is an open set of Z, since is étale (I, p. 30, Corollary 3), and contains the saturated set of for the relation . The mapping is closed (IV, p. 386, Proposition 2). There then exists an open set V of such that is a neighbourhood of contained in (I, p. 75, Lemma).
Let now , and let be the unique point of Z such that ; by definition, one has . Since , one has . This shows that the restriction of to the open set of is equal to pr. Consequently, is a homeomorphism.
Thus we have shown that the descent datum is effective. The mapping is universally strict (I, p. 20, corollary). Under the assumption that is surjective, it then follows from Proposition 8 of I, p. 31 that is étale.
5. Descent: case of coverings
Let X, Y be topological spaces, and let be a surjective continuous mapping. Let be a covering of X, and let be a descent datum relative to on Z.
If is proper and separated (resp. if is open), the descent datum is effective and the Y-space is an étale space (IV, p. 389, Proposition 5). It is even a covering of Y if has a continuous section in a neighbourhood of each point (I, p. 72, Proposition 3) or if Z is a locally finite covering of X (I, p. 77, Corollary 4). This number is devoted to bringing out other conditions under which is a covering of Y.
Let us first prove a lemma.
Lemma 1 022I
Let B, be topological spaces, and let be a continuous mapping. If the fibred square is locally connected, the topological space is locally connected.
Let be a point of and let V be a neighbourhood of . Suppose that the fibre square is locally connected and let W be a connected neighbourhood of in which is contained in . Put U = pr. The set U is contained in V, and is connected since the image of a connected set under a continuous mapping is connected. If denotes the diagonal of , the mapping pr is a homeomorphism. Since U contains pr and is a neighbourhood of in , U is a neighbourhood of in . This proves the lemma.
Proposition 6 022J
Let
E
B
be a cartesian square. Suppose that the mapping is proper, separated and surjective, and make one of the following assumptions:
(i) The fibres of are locally connected and the fibre square is locally connected.
(ii) The fibres of are finite, the diagonal of is open in and is a locally connected space.
Then, if is a covering, is a covering.
Since the mapping is universally strict (I, p. 20, corollary), the mapping is étale (I, p. 31, prop. 8) and separated (I, p. 27, prop. 4). We shall suppose, which is permissible, that .
Let be a point of B; it is a question of proving that the point has a neighbourhood W such that the W-space is a covering
which is trivializable. Put = and denote by = . The mapping defined by is a -isomorphism (I, p. 9, prop. 4), hence is a trivializable covering of and is a trivialization of this covering.
Let us prove that there exists a neighbourhood of in and a continuous trivialization of the covering extending . Under assumption (ii), is finite and its points have pairwise disjoint open neighbourhoods over which the covering is trivializable, whence the assertion in this case. Under assumption (i), is locally connected, and so is IV, p. 390, lemma 1); since the mapping is proper and separated, is compact and two distinct points have disjoint neighbourhoods in , so that the pair satisfies property (PCV) ( I, p. 37, lemma 1). The assertion therefore follows from Cor. 2 of I, p. 90.
As is proper, there exists a neighbourhood V of in B such that contains (lemma, I, p. 75). We may thus suppose that
Let us endow the -spaces and with their canonical descent data relative to . We shall now prove that, after shrinking V and , the isomorphism of -spaces which we have just defined is compatible with the descent data, that is to say that one has if and . Let denote the mapping pr.
Put ; let us consider it as a -space by means of the first projection. The mapping of into is an isomorphism of -spaces; this shows that is a covering of .
For , 2, define a mapping by setting ; these are trivializations of the covering . Let be the set of points over which these trivializations and coincide; it contains , as well as the diagonal . Let us prove that is a neighbourhood of . Under assumption (i), this follows from Cor. 2 of I, p. 80, since is locally connected. Under assumption (ii), contains a neighbourhood of in , for this set is locally connected (loc. cit.). As is open in is a neighbourhood of in .
As is proper, the canonical mapping of into B is proper, for it is the composite of the projection pr and the mapping .
By the lemma of I, p. 75, there exists a neighbourhood W of in V
such that is contained in ; let us put ; it is a subset of and the isomorphism of étale spaces is compatible with the canonical descent data relative to the mapping . It follows from corollary (IV, p. 388) that the W-étale spaces and are isomorphic. In particular, is a trivializable covering, whence the proposition.
Corollary 1 022K
Let E and B be topological spaces, a continuous mapping and a locally finite closed covering of B such that for every pair , the intersection is a locally connected space. Then, in order that the B-space be a covering, it is necessary and sufficient that, for every , the -space be a covering of .
The condition is necessary (cf. I, p. 69). Conversely, let denote the topological sum of the family and the canonical mapping. The mapping is closed (TG, I, p. 6, prop. 4), separated (I, p. 27, remark 5), with finite fibres, hence proper (TG, I, p. 75, th. 1); it is also surjective. The diagonal , being equal to , is open in . Finally, the space is homeomorphic to the topological sum of the family , , ; it is therefore locally connected. Hypothesis (ii) of proposition 6 is satisfied. If for every is a covering of is then a covering of , hence E is a covering of B.
Corollary 2 022L
Let X and Y be topological spaces and let be a proper, separated and surjective mapping. Assume in addition one of the following hypotheses:
(i) The fibres of , as well as the space , are locally connected;
(ii) The fibres of are finite, the diagonal of is open in and the space is locally connected. Then every descent datum relative to on a covering of X is effective, and the quotient space is a covering of Y.
Let Z be a covering of X and let be a descent datum relative to on Z. By proposition 5 (IV, p. 389), the descent datum is effective, in other words the square
Z
X Y
is a cartesian square. The hypotheses of proposition 6 (IV, p. 391) are then satisfied. Consequently, is a covering of Y.
Proposition 7 022M
Let X and Y be topological spaces and let be a continuous and surjective mapping. Suppose that the space Y is unloopable and that the mapping is open and has the path-lifting property. Then every descent datum relative to on a covering of X is effective, and the quotient space is a covering of Y.
Let be a covering of X and let be a descent datum relative to on Z. Since the mapping is surjective and open, it follows from prop. 5 of IV, p. 389, that the descent datum is effective. Let ; it is the quotient Y-space; its projection is étale (loc. cit.); it is also separated (I, p. 27, prop. 4).
By hypothesis, the mapping has the path-lifting property, and so does the mapping , since Z is a covering of X (III, p. 302, Corollary 2 of Prop. 3). Hence the same is true of the mapping , and therefore of the mapping . Consequently (cf. IV, p. 341, Remark 2), T is a covering of Y. The proposition is thus proved.
Remark 022N
Let X and Y be topological spaces and let be a mapping. In order that every descent datum relative to on a covering of X be effective, and that the quotient space be a covering of Y, it is necessary that be strict and that be an open and closed subset of X.
In fact, let us identify the X-space X with and endow it with its canonical descent datum relative to . The quotient space is identified with , endowed with the quotient topology of the topology of X for the equivalence relation defined by . If this is a covering of Y, the space is then identified with an open and closed subset of Y and the mapping is strict.
6. Descent of groupoids
Let X and Y be topological spaces and let be a continuous mapping. Let and denote the two projections of into X, Coeg() the coequalizer groupoid of the pair of groupoid morphisms from into , and Coeg() the canonical groupoid morphism (II, p. 199, Def. 2). Since , we have . It then follows from the universal property of coequalizers (II, p. 199, Prop. 3) that there exists a unique groupoid morphism Coeg() such that .
The set of vertices of Coeg() is the quotient set of the set X = Som() by the equivalence relation defined by II, p. 200, Remark 1). It is identified with .
Proposition 8 022O
Let X and Y be non-empty topological spaces and let be a continuous mapping of X into Y. Assume that the space X is locally arcwise connected, that the space Y is connected, and that the mapping is strict and surjective. Then the groupoid Coeg() is transitive.
Let Γ be the quiver whose set of vertices is X and whose set of arrows is the sum set of Fl() and , the origin and target mappings being those of on Fl() and the mappings and on . By the definition of the framework of the pair (pr)(pr)) (II, p. 185, Def. 3) and Remark 2 of II, p. 200, the set of orbits of Coeg() is identified with the set of connected components of the quiver Γ. Since the space X is non-empty, it is enough to prove that the graph Γ is connected.
The connected components of Γ are saturated with respect to the equivalence relation “there exists a path joining to ”, and therefore are open in X, since X is locally arcwise connected. They are then also closed. They are also saturated with respect to the equivalence relation R defined by . By hypothesis, the mapping induces a homeomorphism of onto Y, so that the image under of every connected component of Γ is an open and closed subset of Y, and therefore is equal to Y since the space Y is assumed connected.
Let C be a connected component of Γ, and let be a point of X. From the preceding, there exists a point such that . By definition of the quiver Γ, we have . Thus C = X and the quiver Γ is therefore connected.
Proposition 9 022P
Let X and Y be topological spaces and let be a continuous mapping of X into Y. Suppose that the space X is locally arcwise connected, that the space Y is simply connected, and that the mapping is strict and surjective. Then the groupoid is generated by the image of under .
We may suppose that the spaces X and Y are non-empty. Let G denote the subgroupoid of generated by the image of under . Since the mapping is surjective, the set of vertices of G is equal to Y. By prop. 8, Coeg() is a transitive groupoid. Since induces the identity on the vertices, the image of Coeg() under is a transitive subgroupoid of . The image of under generates Coeg() (II, p. 200, corollary); since one has , the groupoid G is transitive.
Let be a point of Y and let H denote the subgroup of . By Th. 1 of IV, p. 342, there exists a connected covering of Y and a point of the fibre of which H is the stabilizer.
If is a point of X, the set Fl is not empty, since G is transitive. For Fl, the point is a point of the fibre , independent of since the group H, the isotropy group of G at , fixes . Let us denote this point by ; let denote the resulting mapping and the mapping . By construction, the mapping is compatible with the canonical operations of in X and . It then follows from lemma 1 of III, p. 312 that is continuous. The mapping is therefore continuous; moreover it is compatible with the equivalence relation defined by , for depends only on . Since is strict and surjective, there exists a unique continuous mapping such that , so that the covering T admits a section. Since T is connected, the mapping is a homeomorphism (I, p. 31, cor. 4 of prop. 6), and one has , whence . Since G is transitive, it follows that .
Proposition 10 022Q
Let X and Y be topological spaces and let be a continuous mapping of X into Y. Suppose that the space X is simply connected, that the space is locally arcwise connected, and that every descent datum relative to on a covering of X is effective and the quotient space is a covering of Y. Then the groupoid morphism of Coeg() into is injective.
Under the hypotheses of the proposition, is strict ( IV, p. 394, remark); moreover, one may suppose that it is surjective.
Lemma 2 022R
Let us retain the notation and hypotheses of the proposition. Let T be a set endowed with an operation of the groupoid Coeg() relative to a mapping . Then there exists a unique topology on T for which makes T a covering of Y, such that for every class of paths with origin in X and every point of T, one has . In particular, the operation of Coeg() on T factors through an operation of the groupoid .
The uniqueness of such a topology on T follows from Lemma 1 (III, p. 312); let us prove its existence. Endow the set with a law of operation of by putting for every and every arrow with origin in . Since the space X is simply connected, this law of operation of is without local monodromy (III, p. 313, Remark). There therefore exists on the set a unique topology which makes the X-space , pr a covering of X and such that the canonical operation of in this covering coincides with the given operation (III, p. 313, Prop. 3). Let us denote this covering by . Let us prove that the mapping of into Z is a descent datum on Z relative to the mapping . It is enough to verify that it is continuous, the other conditions of Definition 1 of IV, p. 382, being evident.
The mapping is a lifting of the mapping pr to the covering Z of X. Since the space is locally arcwise connected, the space , which is a covering of it, is also locally arcwise connected. By the Corollary (III, p. 269), in order to prove that the mapping is continuous, it is enough to prove that it is arcwise continuous.
Let therefore be a path in and let us prove that the mapping of into Z is continuous. For every , let us denote by and the paths in X defined by and . For every , one thus has , and , where denotes the strict homotopy class of a path III, p. 304, Remark). The mapping is a path in ; by definition of the coequalizer Coeg(), one therefore has and . By definition of the operation of on Z, one therefore has
This shows that is a continuous lifting to Z of the path loc. cit.). Consequently, the mapping is arcwise continuous, hence continuous; it is a descent datum relative to on the X-space Z.
Let denote the equivalence relation defined by the descent datum . Since the mapping is surjective, the mapping pr induces, by passing to the quotient, a bijection of onto T. Let us endow T with the topology deduced from that of by transport of structure, so that is a Y-space. By hypothesis, T is therefore a covering of Y and the diagram
Z T
X Y
is a cartesian square.
Let be a point of Z, a path with origin in X, and let be the continuous lifting of to Z with origin . The path pr is the lifting with origin of the path in Y, so that . This completes the proof of the lemma.
Let us now prove the proposition. Let and be two arrows of Coeg() whose images in are equal. Since the groupoid morphism is the identity on the vertex sets, the arrows and have the same origin and the same target. Let us denote by the origin of ; let T be the set of arrows of Coeg() with origin , and let be the restriction to T of the target mapping. The groupoid Coeg() operates by composition on the right on the set T, relative to . By lemma 2, the actions of and on T are identical. Hence . This proves that the groupoid morphism is injective.
Theorem 1 022S
Let X and Y be topological spaces, and let Y be a continuous surjective mapping. Suppose that the spaces X and Y are locally path-connected and semi-locally simply connected. Finally, suppose that one of the following properties is satisfied:
(i) The mapping is proper, separated, with locally connected fibres, and the space is locally path-connected;
(ii) The mapping is proper, separated, with finite fibres, the diagonal is open in , and its complement is locally connected;
(iii) The mapping is open and has the path-lifting property. Then the groupoid morphism is an isomorphism of the groupoid Coeg() onto the Poincaré groupoid .
First let us note that under these hypotheses, the mapping is surjective and strict (I, p. 18, example 2). Moreover, every descent datum relative to on a covering of X is effective, and the quotient space is a covering of Y; this follows indeed from IV, p. 393, corollary 2 of prop. 4 under hypotheses (i) and (ii), and from prop. 7 of IV, p. 394 under hypothesis (iii). By prop. 10, the groupoid morphism is therefore injective and its image is a subgroupoid of .
By prop. 9 of IV, p. 395, this image is equal to under hypotheses (i) and (ii), but also under hypothesis (iii), since the groupoid morphism is then surjective.
Consequently, is an isomorphism.
Example 022T
Here are two examples in which the hypotheses of theorem 1 are satisfied.
1) Let Y be an unlaceable topological space. Let be a locally finite covering of Y by closed sets. Suppose that, for every , the space is unlaceable and that, for every pair , the space is locally path-connected. One may take for X the sum space of the family and for the mapping deduced from the family of canonical injections.
2) Let G be a discrete group operating properly in an unlaceable topological space X, put and let be the canonical mapping. It is open (TG, III, p. 10, lemma 2) and has the path-lifting property by virtue of Theorem 4 of III, p. 287. The space Y is unlaceable by IV, p. 349, Prop. 8, b). By hypothesis, the mapping from into given by is proper, hence strict (I, p. 18, Example 2) and its image is . It then follows from III, p. 261, Prop. 8 that is locally path-connected.
7. Descent by an étale and surjective mapping
Let X and Y be topological spaces and let be a continuous mapping from X into Y. We retain the notation of the preceding No.
Theorem 2 022U
Suppose that every point of Y possesses a neighbourhood above which there exists a continuous section of the mapping . The groupoid morphism from Coeg() into is an isomorphism.
By hypothesis, there exists a covering of Y by open sets and, for every , a continuous section of .
If is a path in X, we denote by its strict homotopy class in and by the image of in Coeg() by the groupoid morphism . If and are two paths in X and are composable in Coeg() if and only if the paths and in Y are juxtaposed.
Let be a path in Y. By Lemma 4 of III, p. 272, applied
to the compact space and to the covering () of , there exists an integer such that, for every integer satisfying , the image of the interval by is contained in an open set . For every integer , let be the path in Y defined by ; one has (cf. III, p. 291, remark 1), and is the path denoted by . For every , let us denote by the path in X and put . Since for every , the paths and can be juxtaposed, the sequence is composable in Coeg(). By construction,
which proves that the groupoid morphism is surjective.
Let and be arrows of Coeg(). Since the groupoid Coeg() is generated by the image of (II, p. 200, corollary), there exist finite sequences and of paths in X such that one has and . The paths in Y can then be juxtaposed and one has ; analogously, .
Suppose that one has . There then exists a strict homotopy joining to . By lemma 4 of III, p. 272, applied to the compact space and to the covering of , there exists an integer such that, for every pair of integers satisfying and , the image of by is contained in an open set of the covering .
Every path in X is of the form , where is the path . Replacing the integers and by their product if necessary, one may therefore suppose that .
For every pair of integers of and every pair , set
For , also set , , and . By lemma 1 of III, p. 295, the paths and are strictly homotopic, whence the relation
in , for every pair . On the other hand, for every pair of integers , with 2 and 1 , one has , whence the relation
in Coeg(). Analogously, for every pair of integers such that and , one has
For , the paths and coincide. By definition of the coequalizer Coeg(), one therefore has . Analogously, for every . Hence one has the relations
By lemma 3 below, one has
Since the paths and are constant, one has, for and where and are the origin and the end of the arrows and of Coeg(). It follows that
that is, .
Lemma 3 022V
Let G be a groupoid and let be integers . For every pair of integers such that and , let be a vertex of G; for every pair such that and , let be an arrow of G joining to ; for every pair such that and , let be an arrow of G joining to .
Assume that, for every pair of integers such that and , the arrows and are composable, analogously the arrows and , and that one has . Then,
Let us first treat the particular case where and prove the result by induction on . If , the assertion to be proved is true by assumption; suppose it true for ; one then has
by the induction hypothesis, whence the relation for .
Let us now prove the result by induction on . It is true for by what precedes; if it is true for , one then has
From the case , one has
whence the required relation.
Example 1 022W
The theorem applies when the mapping is étale and surjective.
Example 2 022X
It also applies when the space X is the sum space of a family of parts of Y whose interiors cover Y, and when is the mapping induced by the canonical injections of each of the into Y.
8. Poincaré Groupoid of a Quotient Space
Let X be a topological space endowed with a continuous action of a discrete group G; set and denote by Y the canonical mapping. Let denote the groupoid ; its set of vertices is G; for , there exists a unique arrow joining to if , and none otherwise. By passing to the fundamental groupoids, the action induces a morphism of groupoids . Let be the coequalizer of the two morphisms of groupoids and (pr) from to ; let denote the canonical morphism of groupoids. We have pr, hence (pr). By the universal property of coequalizers, there therefore exists a unique morphism of groupoids such that .
Theorem 3 022Y
Let X be a delaceable topological space and let G be a discrete group operating properly in X; let be the canonical surjection. The canonical morphism of groupoids introduced above is an isomorphism.
Let Coeg() denote the coequalizer of the two morphisms of groupoids from to induced by the projections pr and pr; let Coeg() be the canonical morphism of groupoids. Since the image of the mapping , pr is the subspace of , the two morphisms of groupoids and (pr), from to Coeg(), are equal. By the universal property of , there exists a unique morphism of groupoids Coeg() such that .
Let also denote the unique morphism of groupoids from Coeg() to such that . By IV, p. 399, Example 2, the hypotheses of Theorem 1 of IV, p. 398 are satisfied and the morphism is an isomorphism. Since
we have . Thus, to prove Theorem 3, it is enough to prove that the morphism is an isomorphism.
Lemma 4 022Z
For every path in , we have .
Let be such a path.
Let , and let be its stabilizer in G. By TG, III, p. 32, prop. 8, there exists an open neighbourhood of in X such that = for every G - , and such that the mapping induces a homeomorphism of onto an open neighbourhood of in Y. As X is unlaceable and the restriction of the mapping to is open and closed, we may further suppose that is connected and that the image of the canonical homomorphism is reduced to the identity element. The open sets thus constructed form an open covering of Y. By lemma 4 of III, p. 272, applied to the compact space and to the open sets for , there exists an integer such that for every , there exists a point in X such that
is contained in . As is also
contained in .
For or 2 and for , let denote the path in X defined by ; we have III, p. 291, remark 1). Hence, to prove that , it is enough to prove that for every integer . Replacing the pair of paths by the pair if necessary, we thus suppose that there exists such that .
The inverse image of by is the disjoint union of the connected components , where runs through a system of representatives in G of . For or 2, let be an element of G such that the point belongs to . The image of the path is then contained in . By definition of the groupoid , we have , which makes it possible to suppose that the images of the paths and are contained in and that .
For or 1, let be a path in joining to and let be an element of such that . The paths and are loops at in . They are therefore strictly homotopic in X to the constant loop at , because the image of the canonical homomorphism reduces to the identity element. Their classes have in particular the same image under the groupoid morphism , whence
Since for every element and every path in X, it follows that , as was to be proved.
By the lemma, the two groupoid morphisms (pr) and (pr) of into Coeg() are equal. By the universal property of the coequaliser, there therefore exists a unique groupoid morphism Coeg() such that . The morphism is the unique groupoid morphism of into itself such that ; hence = Id. Analogously, Id. Consequently, is an isomorphism.
Exercises
See the exercises for § 4.