ORBIFOLD QUOTIENTS OF SYMMETRIC DOMAINS

OF TUBE TYPE

FABRIZIO CATANESE

Dedicated to Lucian Badescu, in memoriam, a dear friend and a very

arXiv:2406.03027v2 [math.AG] 6 Oct 2024

esteemed colleague.

Abstract. In this paper we characterize the compact orbifolds,

quotients X = D/Γ of a bounded symmetric domain D of tube
type by the action of a discontinuous group Γ, as those projec-
tive orbifolds with ample canonical divisor possessing a slope zero
tensor of ‘orbifold type’.

Contents
Introduction 1
1. Complex orbifolds, Deligne-Mostow orbifolds, orbifold fundamental groups, orbifold cov
2. Locally symmetric manifolds of tube type and descent of slope zero tensors 5
3. Proof of the Main Theorems 7
3.1. Proof of Theorem 3 8
3.2. Proof of Theorem 4 9
3.3. Final remarks 10
References 10

Introduction
Let M be a simply connected complex manifold, and Γ be a properly
discontinuous group of automorphisms (biholomorphic self maps) of M.
Then the quotient complex analytic space X = M/Γ is a normal
complex space.
In the case where the action of Γ is quasi-free, namely, Γ acts freely
outside of a closed complex analytic set of codimension at least 2, we
just consider the normal complex space X; but, in the case where the
set Σ of points z ∈ M whose stabilizer is nontrivial has codimension
1, it is convenient to replace X by the complex global orbifold X ,
consisting of the datum of X and of the irreducible Weil divisors Di ⊂
Date: October 8, 2024.
AMS Classification: 32Q15, 32Q30, 32Q55, 14K99, 14D99, 20H15, 20K35.
Key words: Symmetric bounded domains, properly discontinuous group actions,
Complex orbifolds, Orbifold fundamental groups, Orbifold classifying spaces.

1
2 FABRIZIO CATANESE

X, whose union is the codimension 1 part of the branch locus B of

p : M → X (B = p(Σ) is the set of critical values of p): each divisor
Di is marked with the integer mi which is the order of the stabilizer
group at a general point of the inverse image of Di .
The more general case where Γ′ is a properly discontinuous group
of automorphisms of any (connected) complex manifold M ′ reduces to
the previous one by taking M to be the universal covering of M ′ and
letting Γ be the group of lifts to M of elements of Γ′ .
One says that the above global orbifold X is good if Γ admits a finite
index normal subgroup Λ which acts freely (this holds if Λ is torsion
free), with quotient a compact complex manifold Y = M/Λ: in this
case X = Y /G, where G is the finite quotient group G := Γ/Λ.
Particularly interesting are the cases where M is a contractible do-
main M = D ⊂ Cn : in this case Y is a classifying space K(Λ, 1) for
the group Λ, whereas (see section 1) X is an orbifold classifying space
for the orbifold fundamental group Γ of X .
The easiest example is the case where D = Cn , Y is a complex torus
Y = Cn /Λ, and X is a finite quotient of a complex torus: this case
was considered in [Cat24], and can be classified by simply saying that
Γ is an arbitrary abstract even cristallographic group, endowed with a
complex structure on Λ ⊗ C.
A more difficult case is the case where D ⊂⊂ Cn is a bounded sym-
metric domain (see [Koba59], [Helga78] and also [Mok89], [CaDS12]).
Again we have a good global orbifold, by virtue of the so-called Sel-
berg’s Lemma ([Selb60], [Cass76]).
In the case where Γ acts freely, we have a so-called locally-symmetric
manifold, and there is a vast literature devoted to their possible char-
acterizations, some final touch with rather explicit criteria being con-
tained in our work with Di Scala, [CaDS12], [CaDS12].
The purpose of this note is to apply the idea, as in [Cat24], to use
orbifolds in order to deal with the case of a non free action of Γ. At
least what is easier here is that necessarily X must be projective, and
indeed by [Kod54], the canonical divisor KY and the orbifold canonical
divisor
X mi − 1
KX := KX + D̂ := KX + Di
i
m i

must be ample.
We have a priori two options for the assumptions to be made, for in-
stance we can consider the more general orbifolds introduced in [Cat00]
(see also [Cat08], 5.5 and 5.8, and 6.1 of [Cat15]) or the more special
Deligne-Mostow orbifolds ([Del-Mos93] Section 14), locally modelled
as quotients of a smooth manifold by a finite group; in the quasi-free
case, where all the multiplicities mi = 1, the Orbifold fundamental
group π1orb (X ) is the fundamental group π1 (X ∗ ) of the smooth locus
X ∗ of X, while in general Γ := π1orb (X ) is a quotient of π1 (X \ D).
 ORBIFOLDS QUOTIENTS OF TUBE DOMAINS 3

On the differential geometric side, the input is simpler, see [CaFr09],

[CaDS12], in the case where D is of tube type, which means that D is
biholomorphic to a tube domain T = V + iC, where V is a real vector
space and C ⊂ V is an open self-dual convex cone containing no lines.
In fact, the main concept in the tube case is the one of a nontrivial
slope zero tensor
(1) 0 6= ψY ∈ H 0 (S mn (Ω1Y )(−mKY )),
(here n := dim(Y )) which characterizes the locally symmetric mani-
folds of tube type, together with the property that KY is ample.
Now, ψ descends to a meromorphic section 0 6= ψX on X of
(2) S mn (Ω1X (logD))(−m(KX + D)).
Conversely, such a tensor ψX on X lifts to a holomorphic tensor ψY
only if it is orbifold type, meaning that some vanishing conditions
are to be imposed (see section 2 for precise definitions).
Our first result is the following:
Theorem 3. The global compact complex orbifolds X of bounded sym-
metric domains D of tube type (i.e., D is a product of irreducible
bounded symmetric domains of tube type) are the projective complex
orbifolds such that:
(1) their orbifold fundamental group Γ admits a torsion free normal
finite index subgroup Λ,
(2) X admits a meromorphic slope zero tensor 0 6= ψX ( a meromor-
phic section of S mn (Ω1X (logD))(−m(KX + D))) which is of orbifold
type, P
(3) KX := KX + i mmi −1 i
Di is ample,
and
(i) the corresponding Galois covering Y → X = X/G (G := Γ/Λ) is
smooth, equivalently, the orbifold universal cover of X is smooth, or
(i’) Y has singularities which are 2-homologically connected, that is,
they have a resolution of singularities π : Y ′ → Y with Rj f∗ (ZY ′ ) =
0, for j = 1, 2.
Moreover, X should be an orbifold classifying space.
Our main result is instead:
Theorem 4. The global compact complex orbifolds X of bounded sym-
metric domains D of tube type (i.e., D is a product of irreducible
bounded symmetric domains of tube type) are the projective complex
orbifolds such that:
(1) their orbifold fundamental group Γ admits a torsion free normal
finite index subgroup Λ,
(2) X admits a meromorphic slope zero tensor 0 6= ψX ( a meromor-
phic section of S mn (Ω 1
PX m(logD))(−m(KX + D))) of orbifold type,
(3) KX := KX + i mi −1 i
Di is ample,
4 FABRIZIO CATANESE

and
(ii) X is a Deligne-Mostow complex projective orbifold, and X is an
orbifold classifying space.
One may speculate/ask whether condition (ii) may be replaced by
the weaker assumption that X has KLT singularities.
1. Complex orbifolds, Deligne-Mostow orbifolds,
orbifold fundamental groups, orbifold coverings
This section is an abridged version of the corresponding section of
[Cat24], so we shall be here quicker in the exposition.
Definition 5. (compare 5.5 in [Cat08], and section 4 of [Del-Mos93])
Let Z be a normal complex space, let D be a closed analytic set of Z
containing Sing(Z), and let {Di |i ∈ I} be the irreducible components
of D of codimension 1.
(1) Attaching to each Di a positive integer mi ≥ 1, we obtain a
complex orbifold (Z, D, {mi |i ∈ I}).
(2) The orbifold fundamental group π1orb (Z\D, (m1 , . . . , mr , . . . ))
is defined as the quotient
π1orb (Z \ D, (m1 , . . . , mr , . . . )) := π1 (Z \ D)/hh(γ1m1 , . . . , γrmr , . . . ii
of the fundamental group of (Z \ D) by the subgroup normally gen-
erated by simple geometric loops γi going each around a smooth point
of the divisor Di (and counterclockwise).
(3) The orbifold is said to be quasi-smooth or geometric if moreover
Di is smooth outside of Sing(Z).
(4) The orbifold is said to be a Deligne-Mostow orbifold if more-
over for each point z ∈ Z there exists a local chart φ : Ω → U = Ω/G,
where 0 ∈ Ω ⊂ Cn , G is a finite subgroup of GL(n, C), φ(0) = z, U
is an open neighbourhood of z, and the orbifold structure is induced by
the quotient map. That is, D ∩ U is the branch locus of Φ, and the
integers mi are the ramification multiplicities.
(5) An orbifold is said to be reduced (or impure) if all the multi-
plicites mi = 1.
(6) We identify two orbifolds under the equivalence relation generated
by forgetting the divisors Di with multiplicity 1.
Remark 6. (i) A D-M (= Deligne-Mostow) orbifold is quasi-smooth,
and the underlying complex space Z has only quotient singularities
(these are rational singularities).
(ii) In the case where we have a reduced orbifold, that is, there is no
divisorial part, then the orbifold fundamental group is the fundamental
group of Z \ Sing(Z).
(iii) If Z = M/Γ is the quotient of a complex manifold by a properly
discontinuous subgroup Γ, then Z is a D-M orbifold, since any stabi-
lizer subgroup is finite (Γ being properly discontinuous) and by Cartan’s
 ORBIFOLDS QUOTIENTS OF TUBE DOMAINS 5

lemma ([Cart57]) the action of the stabilizer subgroup becomes linear

after a local change of coordinates.
(iv) one could more generally consider a wider class of orbifolds al-
lowing also the multiplicity mi = ∞: this means that the relation
γimi = 1 is a void condition; this more general case is useful to deal
with the compactifications of finite volume quotients X = D/Γ (see for
instance [AMRT75]).
Now, to a subgroup of the orbifold fundamental group corresponds a
connected orbifold covering of orbifolds, (see for instance [Del-Mos93]),
in particular to the trivial subgroup corresponds the orbifold universal
cover
(Z̃, D̃, {m̃j }).
Definition 7. We say that an orbifold (Z, D, (mj )) is an orbifold clas-
sifying space if its universal covering (Z̃, D̃, {m̃j }) satisfies the proper-
ties
(a) either Z̃ is contractible and the multiplicities m̃j are all equal to
1, or
(b) there is a homotopy retraction of Z̃ to a point which preserves the
subdivisor D̃ ′ consisting of the irreducible components with multiplicity
m̃j > 1.
Definition 8. The orbifold canonical divisor is defined as
X mi − 1
KX := KX + D̂ := KX + Di .
i
mi
It satisfies the property that, for an orbifold covering f : Y → X , we
have
f ∗ (KX ) = KY .
2. Locally symmetric manifolds of tube type and descent
of slope zero tensors
As mentioned in the introduction, a bounded symmetric domain D
is of tube type if D is biholomorphic to a tube domain T = V + iC,
where V is a real vector space and C ⊂ V is an open self-dual convex
cone containing no lines.
Recall the notation for the classical domains:
• In,p is the domain D = {Z ∈ Mn,p (C) : Ip −t Z · Z > 0}.

• IIn is the intersection of the domain In,n with the subspace of

skew symmetric matrices.
• IIIn is the intersection of the domain In,n with the subspace of
symmetric matrices.
• IVn is the Lie Ball in Cn ,
{z||z12 + · · · + zn2 | < 1, 1 + |z12 + · · · + zn2 |2 − 2(|z1 |2 + · · · + |zn |2 ) > 0
6 FABRIZIO CATANESE

There are moreover the exceptional domains D16 of dimension 16

and D27 of dimension 27 (related to 2 × 2 and 3 × 3 matrices over the
Cayley algebra).
The tube domain condition excludes the domains In,p with n 6= p.
The following result was shown in [CaDS12], based on the concept
of a slope zero tensor mentioned in the introduction, see (1):
0 6= ψY ∈ H 0 (S mn (Ω1Y )(−mKY )).
Theorem 9. Let X be a compact complex manifold of dimension n.
Then the following two conditions:
(1) KX is ample
(2) X admits a nontrivial slope zero tensor ψX ∈ H 0 (S mn (Ω1X )(−mKX )),
(here m is a positive integer);
hold if and only if X ∼
= Ω/Γ , where Ω is a bounded symmetric domain
of tube type and Γ is a cocompact discrete subgroup of Aut(Ω) acting
freely.
Moreover, the degrees and the multiplicities of the irreducible fac-
tors of the polynomial ψp (the evaluation of ψ at a point p) determine
uniquely the universal covering X e = Ω.
In particular, for m = 2, we get that the universal covering X e is
a polydisk if and only if ψp is the square of a squarefree polynomial
(indeed, of a product of linear forms).
Now, if X is a global orbifold of a symmetric bounded domain D
of tube type, associated to a properly discontinuous subgroup Γ of
biholomorphisms of D, then by Selberg’s Lemma the group Γ admits
a finite index normal subgroup Λ and the compact complex manifold
Y := D/Λ fulfills conditions (1) and (2) of Theorem 9, in particular it
admits a nontrivial slope zero tensor ψY .
Since X = Y /G, G = Γ/Λ, we want to show that ψY , which is clearly
∗
G- invariant, descends to some meromorphic tensor ψX on the smooth
∗
locus X of X. Then we define
∗
ψX := i∗ (ψX ), for i : X ∗ → X being the inclusion.
In order to achieve our goal, let us consider now the following local
situation.
Lemma 10. Consider the action of the cyclic group µq of roots of unity
at the origin in Y := Cn , via the action, for ζ ∈ µq :
(x, z) := (x1 , . . . , xn−1 , z) 7→ (x, ζz),
with quotient map π : Y → X,
π : (x, z) := (x1 , . . . , xn−1 , z) 7→ (x, y), y := z q .
Then a G-invariant slope zero holomorphic tensor
0 6= ψ ∈ H 0 (S mn (Ω1Y )(−mKY ))
 ORBIFOLDS QUOTIENTS OF TUBE DOMAINS 7

descends to a tensor 0 6= φ, a meromorphic section of (here D = div(y))

S mn (Ω1X (logD))(−m(KX + D)).
Proof. Write ψ as (here r + |I| = mn)
mn
X X (dx)I dz r
AI (x, z) =
r=0 |I|=mn−r
(dx1 ∧ · · · ∧ dxn−1 ∧ dz)m
mn X
X (dx)I dz r
= BI,h (x)z h =
r=0 I,h
(dx1 ∧ · · · ∧ dxn−1 ∧ dz)m
mn X
X (dx)I d(logy)r
= BI,h (x)q −r+m z h+r−m .
r=0 I,h
(dx1 ∧ · · · ∧ dxn−1 ∧ d(logy))m
G-invariance is equivalent to the condition that q divides h + r − m,
hence h = bq + m − r, and we have, downstairs on X, the tensor
mn X
X (dx)I d(logy)r
φ := q −r+m BI,h (x) y b .
r=0
(dx1 ∧ · · · ∧ dxn−1 ∧ d(logy))m
I,h

The holomorphic part of the tensor φ is the sum of the series where
b ≥ 0, that is, corresponding to the terms with
h + r ≥ m.
Its order of pole on D is at most −b = [ mq ].

Definition 11. We shall say that the meromorphic tensor φ is of orb-
ifold type if conversely its pull back ψ to any orbifold covering with
multiplicities ≤ 1 is holomorphic: this means that we only have terms
with b such that bq + m ≥ r.
Remark 12. (i) Since in the case of a good global orbifold quotient
the slope tensor of Y cannot vanish on a divisor, the case h = 0 must
occur, hence it follows that q|(m − r) if we have a nonzero term with
h = 0 and given r.
(ii) In the case of a polydisk D, we have semispecial tensors (see
[CaDS12]), hence we may assume that m = 1 in the previous Lemma,
and it follows that these descend to the quotient as holomorphic tensors.

3. Proof of the Main Theorems

We begin with some general observations, valid also for the specula-
tion made in the introduction.
First of all we concentrate especially on the necessity parts of the
statements; observe that X = Y /G is a Deligne-Mostow orbifold and
the singularities of X are quotient singularities, since, at any point
y ∈ Y having a nontrivial stabilizer Gy < G, the group Gy acts linearly
8 FABRIZIO CATANESE

by Cartan’s Lemma [Cart57]. Since D is contractible, X is an orbifold

classifying space.
By [K-M98] (Prop. 5.15, page 158) quotient singularities (X, x) are
rational singularities, that is, they are normal and, if f : Z → X is a lo-
cal resolution, then Ri f∗ OZ = 0 for i ≥ 1. They enjoy also the stronger
property of being KLT (Kawamata Log Terminal) singularities.
Indeed Prop. 5.22 of [K-M98] (where dlt=KLT if there is no bound-
ary divisor ∆, ∆′ ) says that KLT singularities are rational singularities,
while Prop. 5.20, page 160, says that if we have a finite morphism be-
tween normal varieties, F : Y → X, then X is KLT if and only if Y is
KLT).
Hence, conversely, if we start with a Deligne-Mostow orbifold X, the
corresponding finite covering Y is Deligne-Mostow, and if X is KLT,
then also Y is KLT.
First important principle: in both cases (ii), (iii), the nor-
mal complex space Y has rational singularities.
Moreover, Y is projective if and only if X is projective (by averaging,
we can find on Y a G-invariant very ample divisor).
We pass now to the converse implications.
Key argument: we consider the orbifold covering Y associ-
ated to the normal torsion free subgroup
Λ < Γ := π1orb (X),
and we show that Y is a locally symmetric manifold.
Lemma 13. The orbifold Y is just a normal complex space, that is,
there are no marked divisors with multiplicity mi ≥ 2.
Proof. Consider the exact sequence
1 → Λ → Γ → G → 1.
Then the generators γi have finite order mi , hence their image in G
has order exactly mi , because Λ is torsion free.
This means that the covering Y → X is ramified with multiplicity
mi at the divisor Di , and therefore its inverse image in Y is a reduced
divisor with multiplicity 1.


3.1. Proof of Theorem 3. Case (i): if we assume that Y is smooth,

then Y admits a nontrivial slope zero tensor, and we may directly
invoke Theorem 9, using that by (3) X and Y have ample canonical
divisor, since KY = π ∗ (KX ).
Similarly, if the universal covering is smooth, also Y is smooth, be-
cause by assumption Λ is torsion free and the stabilizers are finite,
whence Λ acts freely.
Case (i’): as in [Cat24] we prove that Y must be smooth.
 ORBIFOLDS QUOTIENTS OF TUBE DOMAINS 9

Let Y ′ be a resolution of Y . Since by assumption R1 f∗ (ZY ′ ) = 0 (this

is ture for instance if Y has rational singularities) and R2 f∗ (ZY ′ ) = 0,
we have an isomorphism
H j (Y ′ , Q) ∼
= H j (Y, Q), j = 1, 2.
Hence the degree 1 morphism α : Y ′ → Y yields an isomorphism of
first and second cohomology groups.
We follow a similar argument to the one used in [Cat02], proof of
Proposition 4.8: it suffices to show that α is finite, because then α,
being finite and birational, is an isomorphism Y ′ ∼ = Y by normality,
hence we have shown that Y is smooth and we proceed as for case (i).
Now, if α is not finite, there is a curve C which is contracted by
α, hence its homology class c ∈ H2 (Y ′ , Q) maps to zero in H2 (Y, Q).
And, since H 2 (Y, Q) ∼= H 2 (Y ′ , Q), the class c of C, by the projection
formula, is orthogonal to the whole of H 2 (Y ′ , Q), which is the pull back
of H 2 (Y, Q).
This is a contradiction because, Y ′ being projective, the class c of C
cannot be trivial.

3.2. Proof of Theorem 4. By (ii) X is a Deligne-Mostow orbifold,

hence also Y is a Deligne-Mostow orbifold, therefore Y is a normal
space with quotient singularities (and these are rational singularities)1.
We need now to mimic the proof in the case where Y is smooth, for
instance the proof of Theorem 9, extending it to the case of a normal
space Y with quotient singularities 2.
The first ingredient is: the existence of a complete Kähler-Einstein
metric on the orbifold Y with ample KY . This was first proven in
dimension 2 in [KobR85] (see also [Siu87] for the techniques used) and
was proven later on in a more general situation in [Cas-LN06] and
[E-G-Z09].
The second ingredient is Proposition 5.4 of [Camp04]: take the orb-
ifold universal covering Ỹ of Y , and let Y ′ be its smooth part: then
there exists a De Rham decomposition of Riemannian manifolds
Y
Y′ = Yi′ ,
i

where the holonomy action on each factor is irreducible.

Now, the slope zero tensor is parallel for the Levi Civita connec-
tion (by the Bochner principle, since the slope is zero), as proven by
Shoschichi Kobayashi in [Koba80], hence all the factors have holonomy
different from the Unitary group.
1Similarly, under assumption (iii), Y has KLT singularities, which are also ra-
tional singularities.
2Pay attention: the orbifolds of [Camp04] are the D-M orbifolds with all m ≤ 1,
i
that we call here of reduced type!
10 FABRIZIO CATANESE

Since KY is ample, there are no flat factors, and by the Theorem of

Berger [Ber53] and Simons the holonomy of each factor is the holonomy
of an irreducible bounded symmetric domain.
Now, the orbifold metric on Ỹ is complete, since the metric on the
orbifold Y is complete; and for each i we can take the completion Ỹi ,
hereby obtaining a decomposition for Ỹ .
Now Yi′ ⊂ Ỹi admits a holomorphic map fi′ to an irreducible bounded
symmetric domain Di : by the Hartogs property, fi extends to a holo-
morphic map
fi : Ỹi → Di ,
which is an isometry when restricted to Yi′ .
Because of completeness, fi is surjective, and since Ỹi is normal, its
singular locus has codimension 2, hence fi′ : Yi′ → fi′ (Yi′ ) must be an
isomorphism as the target is simply connected, and we have a covering
space.
Again by Hartogs, the inverse of fi , defined on fi′ (Yi′ ), extends to
yield an isomorphism; hence fi is an isomorphism.
In particular, it follows that Y is smooth, hence we only need now
to invoke Theorem 9.
3.3. Final remarks.
Remark 14. (a) As already discussed, if X has KLT singularities, by
Prop. 5.20 of [K-M98] also Y has KLT singularities (these are also
rational singularities). In this case, we need again to find a Kähler
Einstein metric on Y , and to use the Bochner principle.
(b) One may also ask whether one can replace the condition of KLT
singularities for X by the condition that X has rational singularities,
proving then that also Y has rational singularities.

References
[AMRT75] Ash, Avner; Mumford, David; Rapoport, Michael; Tai,
Yungsheng : Smooth compactification of locally symmetric vari-
eties. Lie Groups: History, Frontiers and Applications. Vol. IV.
Brookline, Mass.: Math Sci Press. IV, 335 p. (1975).
[Aub78] Aubin, Thierry.: Équations du type Monge-Ampére sur les variétés
kählériennes compactes. Bull. Sci. Math. (2) 102 (1978), no. 1, 63-95.
[Ber53] Berger, Michel: Sur les groupes d’holonomie homogéne des
variétés á connexion affine et des variétés riemanniennes. Bull. Soc.
Math. France 83 (1953), 279-330
[Bog78] Bogomolov, Fedor A. : Holomorphic tensors and vector bundles
on projective manifolds. Izv. Akad. Nauk SSSR Ser. Mat. 42 (1978),
no. 6, 1227 –1287, 1439.
[Bore63] Borel, Armand: Compact Clifford-Klein forms of symmetric
spaces, Topology 2 1963 111 - 122.
[Cart57] Cartan, Henri: Quotient d’ un espace analytique par un groupe
d’automorphismes. Princeton Math. Ser. 12, 90–102 (1957).
 ORBIFOLDS QUOTIENTS OF TUBE DOMAINS 11

[Cas-LN06] Cascini, Paolo, La Nave, Gabriele: Kähler-Ricci Flow and

the Minimal Model Program for Projective Varieties. Preprint,
arXiv:math/0603064 [math.AG] (2006).
[Camp04] Campana, Frédéric: Orbifolds with trivial first Chern class. (Orb-
ifoldes à première classe de Chern nulle.) Collino, Alberto (ed.) et al.,
The Fano conference. Papers of the conference organized to commem-
orate the 50th anniversary of the death of Gino Fano (1871–1952),
Torino, Italy, September 29–October 5, 2002. Torino: Università di
Torino, Dipartimento di Matematica. 339–351 (2004).
[Cass76] Cassels, John William Scott: An embedding theorem for fields.
Bull. Aust. Math. Soc. 14, 193-198 (1976), Addendum 479-480
(1976).
[Cat02] Catanese, Fabrizio: Deformation types of real and complex man-
ifolds. Contemporary trends in algebraic geometry and algebraic
topology (Tianjin, 2000), 195–238, Nankai Tracts Math., 5, World
Sci. Publ., River Edge, NJ,(2002).
[Cat04] Catanese, Fabrizio: Deformation in the large of some complex
manifolds, I Ann. Mat. Pura Appl. (4) 183, Volume in Memory of
Fabio Bardelli, (2004), no. 3, 261–289.
[Cat00] Catanese, Fabrizio: Fibred surfaces, varieties isogenous to a prod-
uct and related moduli spaces. Am. J. Math. 122, No. 1, 1–44 (2000).
[Cat08] Catanese, Fabrizio: Differentiable and deformation type of alge-
braic surfaces, real and symplectic structures. Catanese, Fabrizio
(ed.) et al., Symplectic 4–manifolds and algebraic surfaces. Lectures
given at the C.I.M.E. summer school, Cetraro, Italy, September 2–
10, 2003. Berlin: Springer; Florence: Fondazione C.I.M.E Lecture
Notes in Mathematics 1938, 55–167 (2008).
[Cat15] Catanese, Fabrizio: Topological methods in moduli theory. Bull.
Math. Sci. 5 (2015), no. 3, 287– 449.
[Cat24] Catanese, Fabrizio: Orbifold classifying spaces and quotients of
complex tori. arXiv 2403.06044.
[CaFr09] Catanese, Fabrizio and Franciosi, Marco: On varieties whose
universal cover is a product of curves. With an appendix by Antonio
J. Di Scala. Contemp. Math., 496, Interactions of classical and nu-
merical algebraic geometry, 157-179, Amer. Math. Soc., Providence,
RI, (2009).
[CaDS12] Catanese, Fabrizio and Di Scala, Antonio José: A charac-
terization of varieties whose universal cover is the polydisk or a tube
domain, Math. Annalen, 394, (2012) doi:10.1007/s00208-012-0841-x.
[CaDS12] Catanese, Fabrizio and Di Scala, Antonio José: A charac-
terization of varieties whose universal cover is a bounded symmetric
domain without ball factors. Adv. Math. 257, 567–580 (2014).
[Del-Mos93] Deligne, Pierre; Mostow, George Daniel: Commensurabili-
ties among lattices in PU (1, n). Annals of Mathematics Studies. 132.
Princeton, NJ: Princeton University Press. 183 p. (1993).
[E-G-Z09] Eyssidieux, Philippe, Guedj, Vincent, Zeriahi, Ahmed: Sin-
gular Kähler-Einstein metrics. J. Am. Math. Soc. 22, No. 3, 607–639
(2009).
[G-R58] Grauert, Hans; Remmert, Reinhold: Komplexe Räume. Math.
Ann. 136, 245–318 (1958).
[Gra62] Grauert, Hans: Über Modifikationen und exzeptionelle analytische
Mengen. Math. Ann. 146, 331–368 (1962).
12 FABRIZIO CATANESE

[GKP23] Greb, Daniel; Kebekus, Stefan; Peternell, Thomas:

Miyaoka-Yau inequalities and the topological characterization of cer-
tain klt varieties arXiv:2309.14121, to appear in a Comptes Rend.
Ac. Sci. volume in memory of jean Pierre Demailly.
[Helga78] Helgason, Sigurdur: Differential geometry, Lie groups, and sym-
metric spaces. Pure and Applied Mathematics, 80. Academic Press,
Inc. [Harcourt Brace Jovanovich, Publishers], New York-London,
(1978) xv+628 pp.
[Koba59] Kobayashi, Shoshichi:Geometry of bounded domains. Trans.
Amer. Math. Soc. 92 (1959) 267-290.
[Koba80] Kobayashi, Shoshichi: First Chern class and holomorphic tensor
fields. Nagoya Math. J. 77 (1980), 5 - 11.
[KobNom63] Kobayashi, Shoshichi and Nomizu, Katsumi.: Foundations of
differential geometry. Vol I. Interscience Publishers, a division of
John Wiley & Sons, New York-Lond on (1963) xi+329 pp.
[KobOchi81] Kobayashi, Shoshichi and Ochiai, Takushiro: Holomor-
phic structures modeled after compact hermitian symmetric spaces.
in‘Manifolds and Lie grous’, Hano et al. ed., ‘Papers in honor of Yozo
Matsushima’, Progr. Math. 14, Birkhäuser, Basel (1981), 207–221.
[KobR85] Kobayashi, Ryoichi: Einstein-Kähler V-metrics on open Satake
V-surfaces with isolated quotient singularities. Math. Ann. 272, 385-
398 (1985).
[Kod54] Kodaira, Kunihiko: On Kähler varieties of restricted type (an
intrinsic characterization of algebraic varieties). Ann. of Math. (2)
60, (1954). 28-48.
[K-M98] Kollár, János; Mori, Shigefumi: Birational geometry of alge-
braic varieties. With the collaboration of C. H. Clemens and A. Corti.
Cambridge Tracts in Mathematics. 134. Cambridge: Cambridge Uni-
versity Press. viii, 254 p. (1998).
[Mok89] Mok, Ngaiming : Metric rigidity theorems on Hermitian locally
symmetric manifolds. Series in Pure Mathematics, 6. World Scientific
Publishing Co., Inc., Teaneck, NJ, (1989) xiv+278 pp.
[Mok02] Mok, Ngaiming : Characterization of certain holomorphic geodesic
cycles on quotients of bounded symmetric domains in terms of tan-
gent subspaces. Compositio Math. 132 (2002), no. 3, 289-309.
[Olm05] Olmos, Carlos: A geometric proof of the Berger holonomy theo-
rem, Ann. of Math. (2) 161 (2005), no. 1, 579-588.
[Selb60] Selberg, Atle: On discontinuous groups in higher-dimensional
symmetric spaces. 1960 Contributions to function theory (internat.
Colloq. Function Theory, Bombay, 1960) pp. 147–164 Tata Institute
of Fundamental Research, Bombay
[Siu87] Siu, Yum-Tong: Lectures on Hermitian-Einstein metrics for sta-
ble bundles and Kähler- Einstein metrics. Delivered at the German
Mathematical Society Seminar in Düsseldorf (FRG) in June, 1986.
DMV Seminar, Bd. 8. Basel-Boston: Birkhäuser Verlag. 171 p.(1987).
[Sie48] Siegel, Carl Ludwig: Analytic Functions of Several Complex
Variables. Notes by P. T. Bateman. Institute for Advanced Study,
Princeton, N.J., (1950) ii+200 pp.
[Yau77] Yau, Shing-Tung: Calabi’s conjecture and some new results in
algebraic geometry. Proc. Nat. Acad. Sci. U.S.A. 74 (1977), no. 5,
1798–1799.
 ORBIFOLDS QUOTIENTS OF TUBE DOMAINS 13

[Yau93] Yau, Shing-Tung:: A splitting theorem and an algebraic geomet-

ric characterization of locally Hermitian symmetric spaces., Comm.
Anal. Geom. 1 (1993), no. 3-4, 473 – 486.

Lehrstuhl Mathematik VIII, Mathematisches Institut der Univer-

sität Bayreuth, NW II, Universitätsstr. 30, 95447 Bayreuth,

Korea Institute for Advanced Study, Hoegiro 87, Seoul, 133-722,

South Korea
Email address: [email protected]