Let H be a quiver, let G be a groupoid, let φ and ψ be quiver morphisms of H into G, and let h: Som(H) → Fl(G) be a homotopy connecting φ to ψ. Let G′ be the groupoid deduced from G by contraction of the arrows of the image of h(II, p. 175, No.o11) and let β:G→G′ be the canonical morphism.
Let us denote by Γ the quiver (Som(G), Som(H), Som(φ ), Som(ψ )). By definition of a homotopy, the pair (IdSom(G),h) is a quiver morphism of Γ into G; it extends to a unique groupoid morphism η: Grp(Γ) → G. By construction, the set of vertices of G′ is the set of connected components of the quiver Γ.
In all that follows in this No.o, we shall suppose that the groupoid G is transitive. According to remark 1 of II, p. 170, this amounts to supposing that the groupoid G′ is so. We also fix a vertex a0 of G.
Let us denote by Γ the graph associated with the quiver Γ (cf. II, p. 156). The set of loops of length ⩾1 in Γ is identified with the set Ω(Γ) of finite sequences (z1,...,zn), where n is an integer such that n⩾1 and z1,...,zn are elements of Fl(Γ) such that t(zi)=o(zi+1) for every integer i such that 1⩽i<n and t(zn)=o(z1).
Let z=((b1,ε1),...,(bn,εn)) be an element of Ω(Γ). The conjugacy class of the element
Int(g)(η(z))=gh(b1)ε1...h(bn)εng−1
in Ga0 does not depend on the choice of the arrow g of G joining the vertex a0 to the origin of (b1,ε). We denote this conjugacy class by c(z).
The group morphism βa0:Ga0→Gβ(a0)′ is surjective, and its kernel is the smallest subgroup of Ga0 containing the conjugacy classes c(z) for z∈Ω(Γ).
Let K be the smallest distinguished subgroupoid of G whose set of arrows contains the image of h. The morphism Ga0→Gβ(a0)′ is surjective and its kernel is equal to Ka0(II, p. 170, prop. 2). The proposition then follows from prop. 8 of II, p. 176.
A subset Z of Ω(Γ) is said to be distinguished (relative to the pair (φ,ψ)) if it satisfies the following properties:
(i) For every arrow z of Γ, one has (z,z)∈Z ;
(ii) For every (z1,...,zn)∈ Z, one has (zn,...,z2,z1)∈ Z and (zn,z1,...,zn−1)∈Z ;
(iii) Let z=(z1,...,zn) and z′=(z1′,...,zm′) be elements of Z such that t(zn)=o(z1′). Put zz′=(z1,...,zn,z1′,...,zm′). If two elements among z,z′,zz′ belong to Z, the same is true of the third;
(iv) For every arrow f of H, put φ(f,1)=ψ(f,−1)=φ(f) and φ(f,−1)=ψ(f,1)=ψ(f). Let n be an integer ⩾ 1 and (f1,ε1),...,(fn,εn) a sequence of elements of Fl(H) ×{−1,1} such that ψ(fi,εi)=φ(fi+1,εi+1) for 1⩽i<n and ψ(fn,εn)=φ(f1,ε1); let ai denote the origin of fi and bi its term. In order that ((a1,ε1),...,(an,εn)) belong to Z, it is necessary and sufficient that ((b1,ε1),...,(bn,εn)) belong to Z.
The intersection in Ω(Γ) of every family of distinguished subsets (relative to (φ,ψ)) is again distinguished. In particular, there exists a smallest distinguished subset containing a given subset Z of Ω(Γ).
If N is a distinguished subgroup of Ga, the set of elements z∈Ω(Γ) such that c(z) is contained in N is a distinguished subset of Ω(Γ).
Let Z denote the set of elements z∈Ω(Γ) such that c(z)⊂N.
Let z∈ Fl(Γ). We have c(z,z)={ea} by definition, whence (z,z)∈Z.
Let (z1,...,zn)∈Z. By definition of c, the conjugacy class c(z1,...,zn) is equal to c(zn,z1,...,zn−1) and is formed by the inverses of the elements of c(zn,...,z1). This shows that Z satisfies condition (ii).
With the notation of condition (iii), one can choose elements u∈c(z),v∈c(z′) and w∈c(zz′) such that uv=w. If two of the elements u,v and w belong to N, the same is true of the third, therefore N satisfies (iii).
The notation being that of (iv), we have, for every integer i such that 1⩽i⩽n, the relation
φ(fi)h(bi)=h(ai)ψ(fi)
for h is a homotopy joining φ to ψ. This equality can also be written
φ(fi,εi)h(bi)εi=h(ai)εiψ(fi,εi)
Taking into account the relations ψ(fi,εi)=φ(fi+1,εi+1) for 1⩽i<n and ψ(fn,εn)=φ(f1,ε1), one deduces
so that the conjugacy classes c((a1,ε1),...,(an,εn)) and c((b1,ε1),...,(bn,εn)) are equal. This shows that Z satisfies condition (iv) and completes the proof of the proposition.
Let Z be a subset of Ω(Γ); in order that the conjugacy classes c(z), for z∈ Z, generate the kernel of the canonical homomorphism βa0:Ga0→Gβ(a0)′, it is enough that the smallest distinguished subset of Ω(Γ) containing Z be equal to Ω(Γ).
Let N be the smallest subgroup of Ga0 containing c(z) for every z∈Z; it is distinguished. Let Z′ be the set of elements z of Ω(Γ) such that c(z)∈N. By proposition 2, Z′ is a distinguished subset of Ω(Γ). It contains Z. By hypothesis, it is therefore equal to Ω(Γ). It then follows from prop. 1 (II, p. 197) that N is the kernel of the homomorphism βa.
2. Definition of the coequalizer
Let H be a quiver, let G be a groupoid and let φ,ψ be morphisms of quivers from H into G. Let us denote by Coh(φ,ψ ) the cohomotopizer of the pair (φ,ψ) (II, p. 185, def. 3),α:G→ Coh(φ,ψ ) the canonical groupoid morphism and h the canonical homotopy joining α∘φ to α∘ψ.
Let Coeg(φ,ψ ) be the groupoid obtained from Coh(φ,ψ ) by contraction of the arrows belonging to the image of h (II, p. 196, no 1), let us denote by β: Coh(φ,ψ )→ Coeg(φ,ψ ) the canonical morphism and let us put γ=β∘α.
The pair (Coeg(φ,ψ ),γ ) has the following universal property:
a) We have γ∘φ=γ∘ψ.
b) Let G′ be a groupoid and let θ:G→G′ be a groupoid morphism such that θ∘φ=θ∘ψ. There exists a unique groupoid morphism θ: Coeg(φ,ψ )→G′ such that θ∘γ=θ.
Let a be a vertex of H. The arrow h(a) of Coh(φ,ψ ) joins α(φ(a)) to α(ψ(a)). By definition of the groupoid Coeg(φ,ψ ), the origin and the end of the arrow β(h(a)) are equal; therefore we have γ(φ(a))=γ(ψ(a)).
Let f be an arrow of H; if a denotes its origin and b its end, we thus have
α(φ(f))⋅h(b)=h(a)⋅α(ψ(b))
Taking the image by β of the two members of this equality, one obtains the relation γ(φ(f))=γ(ψ(f)). This proves assertion a).
Let us prove b). The mapping η: Som(H) → Fl(G′) which associates with every vertex a of H the arrow eθ(φ(a)) of G′ is a homotopy joining θ∘φ to θ∘ψ. By the universal property of cohomotopizers (II, p. 185, prop. 3), there exists a unique groupoid morphism θ1: Coh(φ,ψ )→G′ such that θ1∘α=θ and θ1(h(a))=eθ(φ(a)) for every vertex a of H. By prop. 7 of II, p. 176, this latter property implies the existence of a unique groupoid morphism θ: Coeg(φ,ψ )→G′ such that θ∘β=θ1. One then has θ∘γ=θ1∘α=θ.
Conversely, if θ′: Coeg(φ,ψ )→G′ is a groupoid morphism such that θ′∘γ=θ, one has (θ′∘β)∘α=(θ∘β)∘α, whence θ′∘β=θ∘β by II, p. 185, prop. 3, whence θ′=θ by prop. 7 of II, p. 176. This proves the uniqueness of θ, hence assertion b).
The groupoid Coeg(φ,ψ ) is generated by the image of the morphism γ.
Let C be the subgroupoid of Coeg(φ,ψ ) generated by the image of G; let i denote the canonical morphism of C into Coeg(φ,ψ ) and θ:G→C the morphism such that i∘θ=γ. By proposition 2, there exists a unique groupoid morphism θ: Coeg(φ,ψ )→C such that θ∘γ=θ. Then γ=i∘θ∘γ, hence i∘θ is the identity morphism of Coeg(φ,ψ ) (loc. cit.). In particular, the mappings Som(i) and Fl(i) are surjective, therefore C = Coeg(φ,ψ ).
The set of vertices of Coeg(φ,ψ ) is the set of connected components of the quiver
Γ = (Som(G), Som(H), Som(φ ), Som(ψ ))
(II, p. 197). In other words, it is the quotient set of Som(G) by the finest equivalence relation such that φ(a) is equivalent to ψ(a) for every a∈ Som(G).
The mapping Orb(γ ), deduced from γ by passing to orbits, defines by passage to the quotient a bijection of the set of connected components of the framework of the pair (φ,ψ) onto the set of orbits of the coequalizer. This results from II, p. 185, prop. 4 and from the fact that the mapping deduced from β by passing to orbits is bijective ( II, p. 170, remark 1).
Consequently, the mapping of Som(G) into Orb(Coeg(φ,ψ )) deduced from γ identifies the set of orbits of Coeg(φ,ψ ) with the quotient set of Som(G) by the equivalence relation generated by the pairs (φ(x),ψ(x)) for x∈ Som(H) and the pairs (o(f),t(f)) for f∈ Fl(G).
The mapping Fl(γ ) of Fl(G) into Fl(Coeg(φ,ψ )) is not in general surjective.
3. Comparison of the isotropy groups of the cohomotopizer and the coequalizer
Let H be a quiver, let G be a groupoid, and let φ and ψ be morphisms of quivers from H into G. Let us denote by α:G→ Coh(φ,ψ ), β: Coh(φ,ψ )→ Coeg(φ,ψ ) and γ:G→ Coeg(φ,ψ ) the canonical morphisms of groupoids, and by h the canonical homotopy joining α∘φ to α∘ψ. These notations are summarized by the following diagram
γ
H ψφ G α Coh(φ,ψ )β Coeg(φ,ψ ) .
In the whole remainder of this no, we suppose that the framework of the pair (φ,ψ) is a connected quiver and we fix a vertex a0 of G. Consequently, the groupoids Coh(φ,ψ ) and Coeg(φ,ψ ) are transitive.
Let Γ denote the quiver (Som(G), Som(H), Som(φ ), Som(ψ )), Γ the graph associated with Γ, and let Ω(Γ) be the set of finite sequences (z1,...,zn) of arrows of Γ, with n⩾1, such that the terminus of zi is the origin of zi+1 if 1⩽i<n and that the terminus of zn is the origin of z1. In No. 1 there was constructed a mapping z↦c(z) of the set Ω(Γ) into the set of conjugacy classes of the group Coh(φ,ψ )a0.
Let H×GH denote the equalizer in H×H of the quiver morphisms ψ∘pr1 and φ∘pr2. The set of its vertices is the set of pairs (a,b) of vertices of H such that ψ(a)=φ(b); the set of its arrows is the set of pairs (f,g) of arrows of H such that ψ(f)=φ(g); the origin of an arrow (f,g) is the vertex (o(f),o(g)) and its term is the vertex (t(f),t(g)) (cf. II, p. 153).
Finally, let Ker(φ,ψ ) denote the equalizer of φ and ψ; recall (II, p. 165, example 2) that this is the subquiver of H whose vertices a are those such that φ(a)=ψ(a) and whose arrows are the f∈ Fl(H) such that φ(f)=ψ(f).
Let μ:H×GH→H be a quiver morphism such that φ∘μ=φ∘ pr1 and ψ∘μ=ψ∘ pr2. Suppose that, for every pair (a,b) of vertices of H such that φ(a)=φ(b), there exists a vertex c of H such that φ(c)=ψ(b) and a=μ(b,c).
Let A1 be a set of vertices of Ker(φ,ψ ) meeting each of its connected components and let Z1 be the set of elements of Ω(Γ) of the form ((a,1)), where a∈A1. Let also A2 be a set of vertices of H×GH meeting each connected component of H×GH and let Z2 be the set of triples of the form ((a,1),(b,1),(μ(a,b),−1)), as (a,b) ranges over A2.
Then Ω(Γ) is the smallest distinguished subset of Ω(Γ) containing Z1∪Z2. In particular, the conjugacy classes c(z) in Coh(φ,ψ )a0, where z ranges over Z1∪Z2, generate the kernel of the canonical homomorphism βa of Coh(φ,ψ )a0 into Coeg(φ,ψ )β(a0).
Let Z1′,Z2′ and Z′ denote the smallest distinguished subsets of Ω(Γ) containing respectively Z1,Z2 and Z1∪Z2(II, p. 197, definition 1). It is required to prove that Z′ is equal to Ω(Γ); by II, p. 199, Corollary 1 of Proposition 1 of II, p. 197, the last assertion of the proposition will follow.
a) For every vertex a of Ker(φ,ψ ), ((a,1)) belongs to Z1′.
b) For every vertex (a,b) of H×GH, ((a,1),(b,1),(μ(a,b),−1)) belongs to Z2′.
a) Let A1′ be the set of vertices a of Ker(φ,ψ ) such that ((a,1)) belongs to Z1′. We have A1⊂A1′, by definition of Z1′. Let f be an arrow of Ker(φ,ψ ), let a be its origin and b its end. It follows from property (iv) in the definition of a distinguished subset (II, p. 197, definition 1) applied to the sequence ((f,1)) that a∈A1′ is equivalent to b∈A1′. Since A1 meets every connected component of Ker(φ,ψ ), we have A1′= Som(Ker(φ,ψ )), which was to be proved.
b) Let A2′ be the set of vertices (a,b) of H×GH such that the triple ((a,1),(b,1),(μ(a,b),−1)) belongs to Z2′. By hypothesis, we have A2⊂A2′. Since A2 meets each connected component of H×GH, it is enough to establish that, if there exists an arrow (f,f′) joining a vertex (a′,b′) to a vertex (a,b), then (a,b) belongs to A2′ if and only if analogously for (a′,b′).
Put f′′=μ(f,f′); this is an arrow joining μ(a′,b′) to μ(a,b), and one has φ(f′′)=φ(f) and ψ(f′′)=ψ(f′). By condition (iv) in the definition of a distinguished subset of Ω(Γ) (loc. cit.) applied to the sequence ((f,1),(f′,1),(f′′,−1)), the two conditions
(i) the triple ((a,1),(b,1),(μ(a,b),−1)) belongs to Z′;
(ii) the triple ((a′,1),(b′,1),(μ(a′,b′),−1)) belongs to Z′;
are equivalent. This shows that (a,b) belongs to A2′ if and only if (a′,b′) belongs to A2′ and completes the proof of the lemma.
Let us now prove by induction on the integer n⩾1 that every element ((a1,ε1),(a2,ε2),...,(an,εn)) of Ω(Γ) belongs to Z′.
A) Case in which n=1.
Let ((a,ε)) be an element of Ω(Γ) of length 1. We have φ(a)=ψ(a), whence ((a,1))∈Z′ by lemma 1. Condition (ii) in the definition of a distinguished subset then implies that ((a,−1)) belongs to Z′.
B) Case in which n⩾2.
By virtue of condition (ii) for a distinguished subset, it is enough to treat the case in which ε2=1. Suppose that ε1=1; then (a1,a2) is a vertex of H×GH; put a=μ(a1,a2); by lemma 1, the triple ((a1,1),(a2,1),(a,−1)) therefore belongs to Z′. In the case in which ε1=−1, we have φ(a1)=φ(a2); we can therefore choose a vertex a of H such that μ(a2,a)=a1 and the triple ((a2,1),(a,1),(a1,−1)) belongs to Z′, hence the triple ((a1,−1),(a2,1),(a,1)) also, by virtue of condition (ii) in the definition of a distinguished subset.
Then ((a,ε1),(a3,ε3),...,(an,εn)) belongs to Ω(Γ) and is of length n−1. By induction, it is an element of Z′. By virtue of conditions (i), (ii) and (iii) in the definition of a distinguished subset, we deduce successively that the elements
a) With the notation of proposition 4, suppose moreover that the mapping (Som(φ ), Som(ψ )) of Som(H) into Som(G) × Som(G) is injective and that its image is the graph of an equivalence relation in Som(G). Then, for every vertex (a,b) of H×GH, there exists a unique vertex ca,b of H such that φ(ca,b)=φ(a) and ψ(ca,b)=ψ(b).
b) Suppose further that for every arrow (f,f′) of H×GH, there exists an arrow f′′ of H such that φ(f′′)=φ(f) and ψ(f′′)=ψ(f′).
Let A be a set of vertices of H×GH meeting each of its connected components. Then, the kernel of the morphism βa0 is the smallest distinguished subgroup of Coh(φ,ψ )a0 containing the conjugacy classes c((a,1),(b,1),(ca,b,−1)), for (a,b)∈A.
Let R be the equivalence relation in Som(G) whose graph is the image of the mapping (Som(φ ), Som(ψ )). Let (a,b) be a vertex of H×GH. Then we have R{φ(a),ψ(a)} and R{φ(b),ψ(b)}, whence R{φ(a),ψ(b)} since ψ(a)=φ(b). Consequently, there exists a unique vertex c of H such that φ(c)=φ(a) and ψ(c)=ψ(b); we denote it by μ(a,b).
For every arrow (f,f′) of H×GH, choose an arrow f′′ of H such that φ(f′′)=φ(f) and ψ(f′′)=ψ(f) and denote it by μ(f,f′).
Thus a quiver morphism μ:H×GH→H has been defined such that φ∘μ=φ∘ pr1 and ψ∘μ=ψ∘ pr2.
Let (a,b) be a pair of vertices of H such that φ(a)=φ(b). We have R{φ(a),ψ(a)} and R{φ(b),ψ(b)}, hence R{ψ(b),ψ(a)}. Consequently there exists a unique vertex c of H such that φ(c)=ψ(b) and ψ(c)=ψ(a). The vertex μ(b,c) satisfies φ(μ(b,c))=φ(b)=φ(a) and ψ(μ(b,c))=ψ(c)=ψ(a). Hence μ(b,c)=a because the mapping (Som(φ ), Som(ψ )) of Som(H) into Som(G) × Som(G) is injective.
Let A1 be the set of vertices of Ker(φ,ψ ) and let Z1 be the set of elements of Ω(Γ) of the form ((a,1)), for a∈A1. Put A2=A and let Z2 be the set of elements ((a,1),(b,1),(μ(a,b),−1)) of Ω(Γ), for (a,b)∈A2. Let Z (resp. Z1, resp. Z2) denote the smallest distinguished subset of Ω(Γ) containing Z1∪Z2 (resp. Z1, resp. Z2).
Let a∈A1; one has φ(a)=ψ(a) ; let us put x=φ(a). The vertex a is the unique vertex of H such that φ(a)=ψ(a)=x; in particular, one has μ(a,a)=a. The triplet ((a,1),(a,1),(a,−1)) therefore belong to Z2. It then follows from conditions (i) and (iii) in the definition of a distinguished subset that ((a,1)) belong to Z2. Consequently, one has Z1⊂Z2, whence Z=Z2.
By Proposition 4, Ω(Γ) is the smallest distinguished subset of Ω(Γ) containing Z2 and the kernel of βa0 is the smallest distinguished subgroup of Coh(φ,ψ )a0 containing the conjugacy classes c(z), for z∈Z2, whence the corollary.
Let X and R be topological spaces, let o and t be continuous mappings of R into X, let C be the set of pairs (f,g)∈R2 such that t(f)=o(g) in R, and let m be a continuous mapping of C into R making the quiver (X,R,o,t) into a groupoid; suppose moreover that the mapping f↦f−1 of R into R is continuous. Then the hypotheses of the proposition are satisfied if one puts H=ϖ(R), G=ϖ(X),φ=ϖ(o),ψ=ϖ(t) and μ=ϖ(m).
Suppose further that R is the graph of an equivalence relation in X, the mappings o and t being deduced from the projections of X×X into X by passing to the subspaces. Then the hypotheses of the corollary are satisfied.
More generally, the hypotheses of the proposition are satisfied when (G,H,φ,ψ,μ) is a "groupoid of groupoids." This means that G and H are groupoids, φ and ψ are groupoid morphisms of H into G, making the quadruplet (G,H,φ,ψ) into a "quiver in groupoids"; finally, μ is a law of composition in this "quiver" given by a groupoid morphism H×GH→H, satisfying a certain number of properties expressing the associativity of the law and the fact that every "arrow" is invertible. The reader will moreover remark that the mappings to the fundamental groupoid studied in this work all fall under this case.
4. Isotropy Group of a Coequalizer
Let G and H be groupoids; let φ and ψ be groupoid morphisms of H into G. The purpose of this No. is to summarize the calculus of the isotropy groups of the cohomotopizer (II, p. 193, prop. 6) and the comparison of the isotropy groups of the cohomotopizer and of the coequalizer made in the preceding No. in order to deduce from it the calculus of the isotropy groups of the coequalizer Coeg(φ,ψ ).
Let us denote by α:G→ Coh(φ,ψ ),β: Coh(φ,ψ )→ Coeg(φ,ψ ) and γ=β∘α:G→ Coeg(φ,ψ ) the canonical groupoid morphisms. Let us denote by h: Som(H) → Fl(Coh(φ,ψ )) the canonical homotopy; let us recall that the set of vertices of Coh(φ,ψ ) is equal to Som(G).
Let Γ denote the quiver (Som(G), Som(H), Som(φ ), Som(ψ )); let then φ0 and ψ0 denote the mappings deduced from φ and ψ by passing to orbits and Γ0 the framework (Orb(G), Orb(H),φ0,ψ0) of the pair (φ,ψ). We shall assume that the quiver Γ0 is connected and that the set of its vertices is not empty, which amounts to assuming that the groupoids Coh(φ,ψ ) and Coeg(φ,ψ ) are transitive.
Recall (cf. II, p. 192, definition 4) that a basic equipment consists in the data of a family (a,b,c1,c2,T,i0) where: for every i∈ Orb(G),a(i) is a vertex in the orbit i of G; for every j∈ Orb(H),b(j) is a vertex in the orbit j of H,c1(j) and c2(j) are arrows of G joining respectively φ(b(j)) to a(φ0(j)) and ψ(b(j)) to a(ψ0(j)); T is a subquiver of Γ0 whose associated tree is a maximal tree of the graph Γ0; finally, i0 is an orbit of G and one sets a0=a(i0).
One then defines a morphism of quivers τ0 from Γ0 into Coh(φ,ψ ) such that τ0(i)=β(a(i)) and τ0(j)=α(c1(j))−1h(b(j))α(c2(j)) for i∈ Orb(G) and j∈ Orb(H). If i belongs to Orb(G), let di be the unique class of paths in the graph T joining i0 to i, and let δi denote its image in Coh(φ,ψ ) under the morphism of groupoids τ0.
From these data, Prop. 6 (II, p. 193) yields a surjective homomorphism
and describes generators of its kernel, thus providing a presentation of the group Coh(φ,ψ )a0.
The set of loops of length ⩾1 in the graph Γ associated to Γ is identified with the set Ω(Γ) of sequences (z1,...,zn) of arrows of Γ, indexed by Z/nZ, where n runs through the set of integers ⩾1, such that for every k∈Z/nZ, the terminus of zk is the origin of zk+1. Let Z be a subset of Ω(Γ) such that the conjugacy classes c(z), for z∈Z, generate the kernel of the surjective homomorphism βa0. The homomorphism βa0∘Λ is surjective; to deduce its kernel, that is to say a presentation of the group Coeg(φ,ψ )β(a0), it remains to choose, for every z∈Z, an element C(z) in the group (∗Ga(i))∗F(π0(H))
i∈Orb(G)
such that Λ(C(z)) belongs to the conjugacy class c(z).
Let therefore z=(z1,...,zn) be an element of Ω(Γ); put zk=(yk,εk), where yk∈ Fl(Γ) = Som(H) and εk∈{±1}. By definition, c(z) is the conjugacy class of the element
gh(y1)ε1...h(yn)εng−1
of the groupoid Coh(φ,ψ )a0, where g is an arbitrary arrow in Coh(φ,ψ ) joining a0 to the origin of z1. For k∈Z/nZ, let jk be the orbit of yk in H and choose an arrow fk of H joining yk to the vertex b(jk). By definition of a homotopy, one then has the relation
h(yk)(α∘ψ)(fk)=(α∘φ)(fk)h(b(jk))
in the groupoid Coh(φ,ψ ). Consequently, using the definition of the quiver morphism τ0, one has
uk=α(φ(fk)c1(jk))⋅δφ0(jk)−1 and vk=δψ0(jk)⋅α(ψ(fk)c2(jk))−1. For every element k∈Z/nZ, define arrows uk,vk in G by
h(yk)εk=ukΛ(jk)εkvk
so that
(uk,vk) if εk=1 ;
(uk,vk)=−−
(vk1,uk1) if εk=−1.
Let xk denote the origin of the arrow h(yk)εk; its end is then xk+1; let ik be the orbit of xk. We then define a loop λk(z) at a(ik) in the groupoid G by the formula
(1)
c2(jk−1)−1ψ(fk−1)−1φ(fk)c1(jk) if (εk−1,εk)=(1,1);
c2(jk−1)−1ψ(fk−1)−1ψ(fk)c2(jk) if (εk−1,εk)=(1,−1); λk(z)=
c1(jk−1)−1φ(fk−1)−1φ(fk)c1(jk) if (εk−1,εk)=(−1,1);
c1(jk−1)−1φ(fk−1)−1ψ(fk)c2(jk) if (εk−1,εk)=(−1,−1). By construction, one has
Λ(λk(z))=vk−1uk
so that the image under the homomorphism Λ of the element
Let G and H be groupoids; let φ and ψ be morphisms of groupoids from H into G. Suppose that the groupoid Coeg(φ,ψ ) is transitive. Let (a,b,c1,c2,T,i0) be a basic equipment of the pair (φ,ψ).
A complementary equipment is the data of a subset Z of Ω(Γ) such that the conjugacy classes c(z) for z∈ Z generate the kernel of the homomorphism βa(i0) and, for every element z=((y1,ε1),...,(yn,εn)) of Z, of a sequence f(z)=(f1,...,fn) of arrows in H such that fk joins yk to the vertex b(jk), jk denoting the orbit of yk in H.
A complete equipment of the pair (φ,ψ) is the data of a basic equipment and of a complementary equipment.
Let G and H be groupoids; let φ and ψ be morphisms of groupoids from H into G. Suppose that the groupoid Coeg(φ,ψ ) is transitive. Let γ:G→ Coeg(φ,ψ ) denote the canonical morphism of groupoids.
Let us endow the pair (φ,ψ) with a complete equipment
(a,b,c1,c2,T,i0,Z,(f(z))z∈Z)
For j∈ Orb(H), let φj:Hb(j)→Ga(φ0(j)) and ψj:Hb(j)→Ga(ψ0(j)) be the group homomorphisms Int(c1(j))−1∘φb(j) and Int(c2(j))−1∘ψb(j) respectively, where φ0 and ψ0 denote the canonical mappings deduced from φ and ψ by passing to orbits. Let τ be the morphism of the framework Γ0 of the pair (φ,ψ) into Coeg(φ,ψ ) defined by Som(τ )(i)=γ(a(i)) if i∈ Orb(G) and such that Fl(τ )(j) is the path γ(c1(j))−1γ(c2(j)) in Coeg(φ,ψ ). If i is an orbit of G, let ci be the unique class of paths in T joining i0 to i and put δi=τ(ci), where τ: Grp(G) → Coeg(φ,ψ ) is the canonical groupoid morphism deduced from τ.
There then exists a unique group homomorphism
λ:(∗Ga(i))∗ F(Orb(H)) → Coeg(φ,ψ )γ(a(i0))
i∈Orb(G)
such that
λ(f)=δiγa(i)(f)δi−1 for i∈ Orb(G) and f∈Ga(i),
λ(j)=δφ0(j)τ(j)δψ0(j)−1 for j∈ Orb(H).
The homomorphism λ is surjective; its kernel is the smallest normal subgroup containing the following elements:
where the loops λi(z) are defined by formula (1), p. 208.
The existence and uniqueness of such a morphism result from the universal property of the free product of a family of groups (A, I, p. 85, prop. 8). Let us denote by α:G→ Coh(φ,ψ ) and β: Coh(φ,ψ )→ Coeg(φ,ψ ) the canonical morphisms, so that γ = β∘α. Let also h: Som(H) → Fl(Coh(φ,ψ )) be the canonical homotopy joining α∘φ to α∘ψ. Since β is the groupoid morphism obtained by contraction of the arrows in the image of h, we have
Fl(τ )(j)=γ(c1(j))−1γ(c2(j))=γ(c1(j)−1)β(h(j))γ(c2(j))=β(τ0(j)) in Coeg(φ,ψ ), where τ0:Γ0→ Coh(φ,ψ ) denotes the quiver morphism deduced from the basic equipment (a,b,c1,c2,T,i0). Consequently, the homomorphism λ is the composite of the homomorphism Λ defined in prop. 6 (II, p. 193) and the surjective homomorphism βa(i0). In particular it is surjective.
Let z∈Z; by construction, Λ(r3(z)) belongs to the conjugacy class c(z). By definition of a complementary equipment, the kernel of the homomorphism βa(i0) is therefore the smallest invariant subgroup of Coh(φ,ψ )a(i0) containing the elements Λ(r3(z)) for z∈Z. Since the homomorphism Λ is surjective, the kernel of the homomorphism λ=βa(i0)∘Λ is therefore the smallest invariant subgroup of the group (∗Ga(i))∗F(π0(H)) containing the generators of the kernel
i∈Orb(G)
of Λ given by formulae (R1),(R2), together with the elements defined by formulae (R3). The proposition is thus proved.
5. Quotient of a groupoid by the action of a group
Let G be a transitive groupoid, let K be a group, and let θ:K→ Aut(G)∘ be a group homomorphism of K into the opposite group of the automorphism group of the groupoid G. We shall say that the group K acts on the right on G. If k∈K, we shall sometimes denote by k∗x (resp. k∗f) the image of a vertex x (resp. of an arrow f) of G under the groupoid automorphism θ(k).
Let ∣K∣ be the groupoid whose set of vertices is K and whose set of arrows joining two vertices is empty if these vertices are distinct, and is reduced to one element otherwise. Let H be the product groupoid G×∣K∣; a vertex of H is a pair (a,k), where a is a vertex of G and k is an element of K; if f is an arrow of G joining a vertex a to a vertex b, we shall denote by (f,k) the unique arrow of H joining (a,k) to (b,k). Let φ:H→G be the groupoid morphism given by the first projection and let ψ:H→G be the groupoid morphism such that Som(ψ )((a,k))=k∗a and Fl(ψ )((f,k))=k∗f if k∈K,a∈ Som(G) and f∈ Fl(G).
Let us denote by G/K the coequalizer Coeg(φ,ψ ) and let γ:G→G/K be the canonical groupoid morphism.
Let o be a vertex of G. For k∈K, choose an arrow ck joining k∗o to o in G; there exists one since G is transitive. For k∈K, let us denote by Fix(k) the subgroupoid of G whose vertices (resp. arrows) are the elements of Som(G) (resp. of Fl(G)) fixed by k; choose a set Ak of vertices of Fix(k) meeting all the orbits of this groupoid. For k∈K and a∈Ak, choose also an arrow f(a,k) in G joining the vertex a to the vertex o.
such that λ(f)=γo(f) for f∈Go and λ([k])=γ(ck) for k∈K is surjective. Its kernel is the smallest normal subgroup of Go∗F(K) containing the following elements:
Since G is transitive, the mapping which to an element k∈K associates the orbit of (o,k) is a bijection of K onto the set of orbits of H. Thus Orb(G) is identified with {o} and Orb(H) with K. The framework Γ0 of the pair (φ,ψ) is then identified with the quiver having a unique vertex o and whose set of arrows is K. Let T be the subquiver of Γ0 with set of vertices {o} and whose set of arrows is empty; the associated graph is the unique maximal tree of the graph Γ0.
The family (o,(o,k)k∈K,(eo)k∈K,(ck)k∈K,T,o) is a basic equipment of the pair (φ,ψ).
The mapping f↦(f,k) defines an isomorphism of the isotropy group Go onto the isotropy group H(o,k); by means of this isomorphism, the homomorphisms φk and ψk of H(o,k) into Go defined by Prop. 5 of II, p. 208 are given by
(3) φk(f,k)=f et ψk(f,k)=ck−1k∗(f)ck,
for k∈K and f∈Go.
The quiver H×GH is the equalizer in H×H of the quiver morphisms ψ∘ pr1 and φ∘ pr2. Its vertices are the pairs ((a,k),(b,h)) where a and b are vertices of G and k and h elements of K such that b=k∗a, and its arrows are the pairs ((f,k),(g,h)) where f and g are arrows of G, and k and h elements of K such that g=k∗f; the origin of the arrow ((f,k),(g,h)) is equal to ((o(f),k),(o(g),h)); its end is equal to ((t(f),k),(t(g),h)).
One then defines a quiver morphism μ from H×GH into H by putting μ((a,k),(b,h))=(a,kh) and μ((f,k),(g,h))=(f,kh). We have φ∘μ=φ∘ pr1 and ψ∘μ=ψ∘ pr2.
Let x and y be vertices of H such that φ(x)=φ(y). Then there exist a vertex a of G and elements k and h of K such that x=(a,k) and y=(a,h). Put z=(h∗a,h−1k); then μ(y,z)=x. This shows that the pair (φ,ψ) satisfies the hypotheses of Prop. 4 (II, p. 202).
A vertex (a,k) of H belong to the groupoid Ker(φ,ψ ), equalizer of φ and ψ, if and only if k∗a=a, that is to say, if k belong to the stabilizer of the vertex a in the group K. Let A1 be the set of pairs (a,k), for k∈K and a∈Ak; it meets all the orbits of the subgroupoid Ker(φ,ψ ) of H. Let Z1 be the subset of Ω(Γ) formed by the sequences of the form (((a,k),1)), for (a,k)∈A1. For z=((a,k),1)∈Z1,(f(a,k),k) is an arrow of H joining (a,k) to (o,k). Put f(z)=((f(a,k),k)).
The set A2 of vertices of H×GH of the form ((o,k),(k∗o,h)), for (k,h)∈K2, meets all the orbits of H×GH. Observe also that one has μ((o,k),(k∗o,h))=(o,kh). The arrows (eo,k), (ck,h), (eo,kh) in H join respectively (o,k), (k∗o,h), (o,kh) to (o,k), (o,h) and (o,kh). Let Z2 denote the set of sequences of the form (((o,k),1),((k∗o,h),1),((o,kh),−1)) in Ω(Γ); for such an element z of Z2, put f(z)=((eo,k),(ck,h),(eo,kh)).
By Prop. 4 of II, p. 202, the set Z=Z1∪Z2 and the family (f(z))z∈Z is a complementary equipment.
Let k∈K and let a∈Ak. The element C((a,k),1) of the group Go∗F(K) defined by formula (2) of II, p. 208 is equal to
Let k and h∈K be given. One verifies that the element
C(((o,k),1),((k∗o,h),1),((o,kh),−1))
of the group Go∗F(K) defined by formula (2) of II, p. 208 is equal to
[k][h](ch−1h∗(ck−1)ckh)[kh]−1(5)
The elements r3′(e,a) given by the relations (R3′) for k=e are all equal to [e]ce−1, an element of the group Go∗F(K) which one obtains by applying the relation (R3′′) to k=h=e. Taking account of relations (3), (4), and (5), the proposition therefore follows from II, p. 208, Prop. 5.
Suppose that the group K is generated by the stabilisers of the vertices of G. Then the group homomorphism γo:Go→(G/K)γ(o) is surjective. Moreover, if the groupoid G is simply transitive, the same is true of the groupoid G/K.
The relation (R3′′) implies λ([e])=λ(ce)=γo(ce). The relations (R3′′) then imply that the set of k∈K such that λ([k]) belong to the image of γo is a subgroup of K. Finally, the relations (R3′) show that, for every element k∈K whose stabiliser is not empty, λ([k]) belong to the image of γo. The corollary follows.
One can give another description, sometimes more convenient, of the group (G/K)γ(o). For this, let M=K×Go and define a law of composition in M by the formula
(k,a)⋅(h,b)=(kh,ckh−1h∗(cka)chb)
for k,h∈K and a,b∈Go. One verifies that this law of composition is associative, that (e,ce−1) is an identity element, and that the element (k−1,ck−−11(k−1)∗(cka)−1ce) is the inverse of (k,a). It therefore endows M with a group structure. Moreover, the mapping λ′:M→(G/K)γ(o) defined by (k,a)↦λ([k]a) is a group homomorphism. Let α′ be the unique group morphism of Go∗F(K) into M such that α′(f)=(e,f) if f∈Go and α′([k])=(k,eo) if k∈K; one has λ′∘α′=λ.
The relations (R2) and (R3′′) show that every element of (G/K)γ(o) is the image under the homomorphism λ of an element of Go∗F(K) of the form [k]f, with f∈Go and k∈K. Hence the homomorphism λ′ is surjective. One verifies moreover that the image under α′ of an element of Go∗ F(K) of the form (R2) or (R3′′) is equal to the identity element. Consequently, the kernel of the homomorphism λ′ is the smallest normal subgroup of M containing the images under α′ of the elements of Go∗F(K) of the form (R3′).
Suppose that the group K operates freely on Som(G). There then exists a unique group homomorphism π:(G/K)γ(o)→ K whose kernel contains the image of γo and such that π(λ([k]))=k for every k∈K. Moreover, Go⟶γo(G/K)γ(o)−→π K is an extension of K by Go.
If such a group homomorphism π exists, the group homomorphism π∘λ is necessarily equal to the unique group homomorphism p from Go∗F(K) into K such that p(f)=e for f∈Go and p([k])=k. It is immediate to verify that the elements of Go∗F(K) defined by formulas (R2) and (R3′′) belong to the kernel of p. By assumption, there is no element of type (R3′). Thus, the kernel of the morphism λ contains that of p. Consequently, there exists a unique group homomorphism π:(G/K)γ(o)→K such that π∘λ=p.
It is evident that the homomorphism π is surjective. To prove that the homomorphism γo is injective and that its image is exactly the kernel of π, let us remark that the homomorphism λ′:M→(G/K)γ(o) (II, p. 213, remark 1) is an isomorphism, since it has been assumed that K operates freely on Som(G). The composite homomorphism (λ′)−1∘γo from Go into M is given by f↦(e,f), whereas the homomorphism π∘λ′:M→K maps (k,f) to k. The corollary follows.
Suppose that the groupoid G is simply transitive. Let K0 be the subgroup of K generated by the stabilizers of the vertices of G. The mapping from K into (G/K)γ(o) which to k∈K associates γ(ck) is a surjective group homomorphism, with kernel K0.
If an element k∈K fixes a vertex a of G, the element g−1kg fixes the vertex g∗a; this implies that K0 is a normal subgroup of K.
By proposition 5, the unique homomorphism λ:F(K)→(G/K)γ(o) such that λ([k])=γ(ck) is surjective, and its kernel is the smallest normal subgroup of F(K) containing the elements [k], where k is an element of K which fixes a vertex of G, and the elements [kh]−1[k][h], where (k,h)∈K2. In particular, the mapping λ′:K→(G/K)γ(o), defined by λ′(k)=γ(ck)=λ([k]) for k∈K, is a group homomorphism. One has λ=λ′∘p, where p:F(K)→K denotes the canonical surjective group homomorphism. Consequently, the homomorphism λ′ is surjective and its kernel is the smallest normal subgroup of K which contains the elements k, for k fixing a vertex of G, that is to say K0. This proves the proposition.