In this article, we investigate more generally abelian varieties with an indecomposable polarization of degree smaller than the dimension.
When the degree increases, many special cases begin to appear, often due to the presence of reducible divisors that represent the polarization.
In order to avoid overly technical statements, we restrict ourselves to two cases: the case where the ambient abelian variety is
simple, and the case of polarizations of degree
$2$
(thereby completing Hacon's abovementioned results). Although we obtained almost complete results for polarizations of degree
$3$
, we chose not to inflict their very technical proofs on the unsuspecting reader.
We refer to Theorems
3 , 4 , 6 , and 7 for more precise formulations and quote only the following results. Let
$(A,\ell )$
be a polarized abelian variety of degree
$d$
and dimension
$g$
, and let
$D$
be any effective divisor that represents
$m\ell $
, with
$m>0$
.
Assume
$A$
is simple and
$g>(d+1{)}^{2}/4$
. If
$m=1$
, the divisor
$D$
is normal and has rational singularities; if
$k\ge 2$
, the set of points of multiplicity at least
$mk$
on
$D$
has codimension
$>k$
in
$A$
.
When
$d=2$
and
$(A,\ell )$
is only indecomposable, similar conclusions hold. Moreover, the set of points of multiplicity at least
$mk$
on
$D$
has codimension
$\le k$
in
$A$
if and only if
$(A,\ell )$
is a double étale cover of a product of at least
$k$
nonzero principally polarized abelian varieties.
The proofs of these results systematically use generic vanishing theorems and precise descriptions of cohomological loci attached to various situations (see § 4 ). We work over the complex numbers.
2 Singularities of pairs
We just need a quick review of the basic terminology relative to the singularities of a pair
$(A,D)$
consisting of an effective
$\mathbf{Q}$
divisor
$D$
in a smooth projective variety
$A$
.
A
log resolution of the pair
$(A,D)$
is a proper birational morphism
$\mu :{A}^{\prime}\to A$
such that the union of
${\mu}^{1}\left(D\right)$
and the exceptional locus of
$\mu $
is a divisor with simple normal crossing support. Write
$${\mu}^{*}({K}_{A}+D)={K}_{{A}^{\prime}}+\sum {a}_{i}{D}_{i}$$
where the
${D}_{i}$
are distinct prime divisors on
${A}^{\prime}$
. The pair
$(A,D)$
is

$\bullet $
log canonical if
${a}_{i}\le 1$
for all
$i$
;

$\bullet $
log terminal if
${a}_{i}<1$
for all
$i$
;

$\bullet $
canonical if
${a}_{i}\le 0$
for all
$i$
such that
${D}_{i}$
is
$\mu $
exceptional;
for some log resolution
$\mu $
. The multiplier ideal sheaf associated to the pair
$(A,D)$
is
$$\mathsf{\Im}(A,D)={\mu}_{*}\left({\omega}_{{A}^{\prime}/A}(\lfloor {\mu}^{*}D\rfloor )\right)={\mu}_{*}\left({\mathfrak{O}}_{{A}^{\prime}}(\sum \lfloor {a}_{i}\rfloor {D}_{i})\right)$$
One sees that
$$\begin{array}{ccc}(A,D)logcanonical& \u27fa& \mathsf{\Im}(A,tD)={\mathfrak{O}}_{A}forallt\in \mathbf{Q}\cap (0,1)\end{array}$$  
$$\begin{array}{ccc}(A,D)logterminal& \u27fa& \mathsf{\Im}(A,D)={\mathfrak{O}}_{A}\end{array}$$  
Assume now that
$D$
is a prime divisor in
$A$
. The adjoint ideal sheaf
$\mathfrak{J}(A,D)\subset {\mathfrak{O}}_{A}$
is defined in [
EL]
, Proposition 3.1. For any desingularization
$f:X\to D$
, it fits into an exact sequence
$$\begin{array}{c}0\to {\omega}_{A}\to {\omega}_{A}\left(D\right)\otimes \mathfrak{J}(A,D)\to {f}_{*}{\omega}_{X}\to 0\end{array}$$ 
(1)

of sheaves on
$A$
and ([
EL]
, Proposition 3.1; [
Ko2]
, Corollary 7.9.2, Theorem 7.9, and Theorem (11.1.1))
$$\begin{array}{ccc}(A,D)canonical& \u27fa& \mathfrak{J}(A,D)={\mathfrak{O}}_{A}\end{array}$$  
$$\begin{array}{ccc}& \u27fa& Disnormalandhasrationalsingularities\end{array}$$  
Furthermore, for any positive integers
$m$
and
$k$
, we have
$$\begin{array}{ccc}(A,\frac{1}{m}D)logcanonical& \u27f9& codi{m}_{A}(Sin{g}_{mk}D)\ge k\end{array}$$  
$$\begin{array}{ccc}(A,\frac{1}{m}D)logterminal& \u27f9& codi{m}_{A}(Sin{g}_{mk}D)>k\end{array}$$ 
(2)

$$\begin{array}{ccc}(A,D)canonicalandk\ge 2& \u27f9& codi{m}_{A}(Sin{g}_{k}D)>k\end{array}$$  
(The pair
$(A,D)$
is log terminal only when
$D=0$
.)
3 Polarized abelian varieties
Let
$A$
be an abelian variety of dimension
$g$
. Any line bundle
$L$
on
$A$
induces a morphism
${\phi}_{L}:A\to Pic0\left(A\right)$
defined by
${\phi}_{L}\left(a\right)={\tau}_{a}^{*}L\otimes {L}^{1}$
, where
${\tau}_{a}:A\to A$
is the translation
$x\mapsto xa$
. This morphism only depends on the numerical equivalence class
$\left[L\right]$
of
$L$
and will also be denoted by
${\phi}_{\left[L\right]}$
. We denote its kernel by
$K\left(L\right)$
or
$K\left(\right[L\left]\right)$
. If
$D$
is a divisor on
$A$
, we write
$\left[D\right]$
for
$\left[{\mathfrak{O}}_{A}\right(D\left)\right]$
and
$K\left(D\right)$
for
$K\left(\right[D\left]\right)$
. The line bundle
$L$
is ample if and only if
$K\left(L\right)$
is finite, in which case this group has order
${d}^{2}$
, where
$$d={h}^{0}(A,L)=\frac{1}{g!}{c}_{1}(L{)}^{g}$$
is the degree of
$L$
. A polarization on
$A$
is a numerical equivalence class of ample line bundles on
$A$
. A polarization of degree
$1$
is called principal and a divisor representing it is called a theta divisor. A polarization
$\ell $
is of type
$\left(d\right)$
if
$K(\ell )\simeq (\mathbf{Z}/d\mathbf{Z}{)}^{2}$
. If
$d$
is prime, any polarization of degree
$d$
is of type
$\left(d\right)$
.
A polarized abelian variety
$(A,\ell )$
is indecomposable if it is not the product of nonzero polarized abelian varieties. If
$g\ge 2$
, a general element of
$\ell $
is then prime.
4 Cohomological loci in
$Pic0\left(A\right)$
Let
$A$
be an abelian variety. For any coherent sheaf
$\mathfrak{F}$
on
$A$
and integer
$i$
, we define reduced subvarieties of
$Pic0\left(A\right)$
by setting
$$\begin{array}{ccc}{V}_{i}(\mathfrak{F})& =& \{P\in Pic0(A\left)\right{H}^{i}(A,\mathfrak{F}\otimes P)\ne 0\}\end{array}$$  
$$\begin{array}{ccc}{V}_{>0}(\mathfrak{F})& =& {\bigcup}_{i>0}{V}_{i}(\mathfrak{F})\end{array}$$  
We investigate the geometry of these loci when
$\mathfrak{F}$
is the tensor product of an ample line bundle with an ideal sheaf.
Lemma 1.
Let
$L$
be an ample line bundle of degree
$d$
on an abelian variety
$A$
of dimension
$g$
, with base locus
$Bs\leftL\right$
, and let
$Z$
be a subscheme of
$A$
, with ideal sheaf
$\mathsf{\Im}$
. Set
${V}_{i}={V}_{i}(L\otimes \mathsf{\Im})$
and
$$h={h}^{0}(A,L\otimes \mathsf{\Im}\otimes P)$$
for
$P$
general in
$Pic0\left(A\right)$
. We have the following.

a)
For
$i>dimZ+1$
, the set
${V}_{i}$
is empty.

b)
If
$h=0$
, the set
${V}_{>0}$
is nonempty.

c)
We have
$h\le d$
, and
$h=d$
if and only if
$Z$
is empty.

d)
If
$h=d1>0$
, and if the polarized abelian variety
$(A,[L\left]\right)$
is indecomposable, the scheme
$Z$
is finite and either
${V}_{1}=Pic0\left(A\right)$
, or
$Z$
is a single (reduced) point
$z$
and
${V}_{1}={\phi}_{L}(BsLz)$
, so that
$dim{V}_{1}\ge gd$
.

e)
If
$A$
is simple and
$0<h<d$
, we have
$dimZ\le d1h$
and
$dim{V}_{>0}\ge g(d+1{)}^{2}/4$
.
Note that
$h={h}^{0}(A,L\otimes \mathsf{\Im}\otimes P)=\chi (A,L\otimes \mathsf{\Im})$
for
$P/\in {V}_{>0}(L\otimes \mathsf{\Im})$
.

Proof.
For
$i>dimZ+1$
, we have
${H}^{i}(A,L\otimes \mathsf{\Im}\otimes P)\simeq {H}^{i}(A,L\otimes P)=0$
for all
$P\in Pic0\left(A\right)$
. This proves a).
If
${V}_{>0}$
is empty,
$$h={h}^{0}(A,L\otimes \mathsf{\Im}\otimes P)=\chi (A,L\otimes \mathsf{\Im})$$
for all
$P\in Pic0\left(A\right)$
. If
$h=0$
, we have
${H}^{i}(A,L\otimes \mathsf{\Im}\otimes P)=0$
for all integers
$i$
and all
$P\in Pic0\left(A\right)$
. This is impossible by [
M]
, Corollary 2.4, and b) is proved.
We have
$h=d$
if and only if all sections of
$L$
vanish on general translates of
$Z$
; this happens if and only if
$Z$
is empty. This proves c). Let us now prove d). Set
$$J=\left\{\right(s,a)\in \mathbf{P}{H}^{0}(A,L)\times As{}_{Z+a}\equiv 0\}$$
The fiber of a point
$a$
of
$A$
for the second projection
$q:J\to A$
is isomorphic to
$\mathbf{P}{H}^{0}(A,L\otimes {P}_{{\phi}_{L}\left(a\right)}\otimes \mathsf{\Im})$
. If
$h>0$
, a unique irreducible component
$I$
of
$J$
dominates
$A$
, and
$dimI=g+h1$
.
Let
$p:I\to \mathbf{P}{H}^{0}(A,L)$
be the first projection. Since any nonempty
${F}_{s}=q\left({p}^{1}\left(s\right)\right)$
satisfies
$Z+{F}_{s}\subset div\left(s\right)$
, we have
$$g1\ge dim{F}_{s}\ge dimIdimp\left(I\right)\ge g+h1dimp\left(I\right)\ge g(dh)$$
Assume
$h=d1>0$
. Then
$p$
is surjective and
${F}_{s}$
has dimension
$g1$
. The divisor of a general section
$s$
being prime, the inclusion
$z+{F}_{s}\subset div\left(s\right)$
is an equality for all
$z$
in
$Z$
. This implies that
$Z$
is finite. If
${V}_{1}\ne Pic0\left(A\right)$
, the length of
$Z$
is
$dh=1$
, so that
$Z=\left\{z\right\}$
, and an element
$P={P}_{{\phi}_{L}\left(a\right)}$
of
$Pic0\left(A\right)$
satisfies
${H}^{1}(A,L\otimes P\otimes \mathsf{\Im})\ne 0$
if and only if the restriction
$${H}^{0}(A,L\otimes P)\to {H}^{0}(Z,L\otimes {\mathfrak{O}}_{Z}\otimes P)\simeq {\mathbf{C}}_{z}$$
is not surjective; in other words, if all sections of
$L\otimes P$
vanish at
$z$
, i.e. ,
$z\in BsL\otimes P=Bs\leftL\righta$
. This proves d).
Assume
$A$
is simple and
$0<h<d$
. Then
$Z$
is nonempty and the inclusion
$Z+{F}_{s}\subset div\left(s\right)$
implies ([
D2]
, Corollaire 2.7)
$$dimZ\le g1dim{F}_{s}\le d1h$$
For
$a$
general in
$A$
, the subvariety
$p\left({q}^{1}\right(a\left)\right)$
of
$\mathbf{P}{H}^{0}(A,L)$
is a linear subspace of dimension
$h1$
. It must vary with
$a$
, because a nonzero
$s$
does not vanish on all translates of
$Z$
. It follows that the linear span of
$p\left(I\right)$
has dimension at least
$h$
. For
${s}_{1},...,{s}_{h+1}$
general elements in
$p\left(I\right)$
, one has ([
D2]
, Corollaire 2.4)
$$dim({F}_{{s}_{1}}\cap \cdot \cdot \cdot \cap {F}_{{s}_{h+1}})\ge g(h+1)(dh)\ge g(d+1{)}^{2}/4$$
For
$a\in {F}_{{s}_{1}}\cap \cdot \cdot \cdot \cap {F}_{{s}_{h+1}}$
, the sections
${s}_{1},...,{s}_{h+1}$
all vanish on
$Z+a$
, hence
${h}^{0}(A,L\otimes \mathsf{\Im}\otimes {P}_{{\phi}_{L}\left(a\right)})\ge h+1$
. This implies
${P}_{{\phi}_{L}\left(a\right)}\in {V}_{>0}$
and proves e). □
Assume now that there is a smooth variety
$X$
with a morphism
$f:X\to A$
such that the sheaf
$\mathfrak{F}$
on
$A$
is a direct summand of
${f}_{*}{\omega}_{X}$
.
Let
$B$
be an abelian variety with a morphism
$\pi :A\to B$
. For all integers
$i$
and
$j$
, and any torsion point
${P}_{0}\in Pic0\left(A\right)$
, every irreducible component of
${V}_{i}\left({R}^{j}{\pi}_{*}\right(\mathfrak{F}\otimes {P}_{0}\left)\right)$
is an abelian subvariety of
$Pic0\left(B\right)$
of codimension at least
$i$
translated by a torsion point. This applies in particular to the loci
${V}_{i}(\mathfrak{F})$
in
$Pic0\left(A\right)$
.
When
${P}_{0}=0$
, this is a particular case of [
HP]
, Theorem 2.2. For the general case, associate to the torsion element
${f}^{*}{P}_{0}\in Pi{c}^{0}\left(X\right)$
a cyclic étale cover
$p:{X}^{\prime}\to X$
. Then
${\omega}_{X}\otimes {f}^{*}{P}_{0}$
is a direct summand of
${p}_{*}{\omega}_{{X}^{\prime}}$
, hence
$\mathfrak{F}\otimes {P}_{0}$
is a direct summand of
${f}_{*}{p}_{*}{\omega}_{{X}^{\prime}}$
.
Lemma 2.
Under the hypotheses and notation of Lemma 1 , assume further that there is an exact sequence
$$0\to {\mathfrak{O}}_{A}^{\oplus \varepsilon}\to L\otimes \mathsf{\Im}\to \mathfrak{F}\to 0$$
for some
$\varepsilon \in \mathbf{N}$
, where
$\mathfrak{F}$
is a direct summand of a pushforward of a dualizing sheaf. The following properties hold.

a)
Every irreducible component of
${V}_{i}(L\otimes \mathsf{\Im})$
is an abelian subvariety of
$Pic0\left(A\right)$
of codimension at least
$i$
translated by a torsion point.

b)
If the support of
$\mathfrak{F}$
is not contained in any nonample divisor of
$A$
, we have, for any
$i>0$
such that
${V}_{i}(L\otimes \mathsf{\Im})$
is nonempty,
$dimZ\ge i1+dim{V}_{i}(L\otimes \mathsf{\Im})$
.

Proof.
Since
${V}_{i}(L\otimes \mathsf{\Im})\left\{0\right\}={V}_{i}(\mathfrak{F})\left\{0\right\}$
, item a) holds. Let us prove b). Since b) follows from Lemma 1 .a) when
${V}_{i}(L\otimes \mathsf{\Im})$
is finite, we may pick a common irreducible component
$V$
of
${V}_{i}(\mathfrak{F})$
and of
${V}_{i}(L\otimes \mathsf{\Im})$
of maximal positive dimension. Let
$B=Pic0\left(V\right)$
and let
$\pi :A\to B$
be the induced morphism. Let
${P}_{0}\in Pic0\left(A\right)$
be a torsion point such that
$V={P}_{0}+{\pi}^{*}Pic0\left(B\right)$
.
We know that
${V}_{k}\left({R}^{j}{\pi}_{*}\right(\mathfrak{F}\otimes {P}_{0}\left)\right)$
has codimension at least
$k$
in
$Pic0\left(B\right)$
. It follows that for general
$P\in Pic0\left(B\right)$
, and all
$k>0$
and
$j\ge 0$
,
$${H}^{k}(B,{R}^{j}{\pi}_{*}(\mathfrak{F}\otimes {P}_{0})\otimes P)=0$$
hence
$$\begin{array}{c}{H}^{0}(B,{R}^{i}{\pi}_{*}(\mathfrak{F}\otimes {P}_{0})\otimes P)\simeq {H}^{i}(A,\mathfrak{F}\otimes {P}_{0}\otimes {\pi}^{*}P)\ne 0\end{array}$$ 
(3)

because
${P}_{0}\otimes {\pi}^{*}P$
is in
$V\subset {V}_{i}(\mathfrak{F})$
. We have an exact sequence
$${R}^{i}{\pi}_{*}({P}_{0}{)}^{\oplus \varepsilon}\u27f6{R}^{i}{\pi}_{*}(L\otimes \mathsf{\Im}\otimes {P}_{0})\u27f6{R}^{i}{\pi}_{*}(\mathfrak{F}\otimes {P}_{0}\left)\stackrel{\delta}{\u27f6}{R}^{i+1}{\pi}_{*}\right({P}_{0}{)}^{\oplus \varepsilon}$$
The sheaves
${R}^{j}{\pi}_{*}\left({P}_{0}\right)$
are direct sums of numerically trivial line bundles on
$B$
(this follows from the proof of [
Ke]
, Theorem 1). By a result of Kollár ([
Ko3]
, Theorem 3.4; [
HP]
, Theorem 2.1), the sheaf
${R}^{i}{\pi}_{*}(\mathfrak{F}\otimes {P}_{0})$
is torsionfree on
$\pi (Supp\mathfrak{F})$
, which is
$B$
by hypothesis. It follows that the support of the sheaf
${R}^{i}{\pi}_{*}(L\otimes \mathsf{\Im}\otimes {P}_{0})$
is
$B$
: if it is not, the map
$\delta $
is generically injective, hence injective; but a twist of
${R}^{i+1}{\pi}_{*}({P}_{0}{)}^{\oplus \varepsilon}$
by a general
$P\in Pic0\left(B\right)$
has no nonzero section, contradicting ( 3 ).
Since
${R}^{i}{\pi}_{*}(L\otimes {P}_{0})=0$
, the short exact sequence
$$0\to L\otimes \mathsf{\Im}\otimes {P}_{0}\to L\otimes {P}_{0}\to L\otimes {\mathfrak{O}}_{Z}\otimes {P}_{0}\to 0$$
yields a surjection
$${R}^{i1}{\pi}_{*}(L\otimes {\mathfrak{O}}_{Z}\otimes {P}_{0})\twoheadrightarrow {R}^{i}{\pi}_{*}(L\otimes \mathsf{\Im}\otimes {P}_{0})$$
hence the support of
${R}^{i1}{\pi}_{*}(L\otimes {\mathfrak{O}}_{Z}\otimes {P}_{0})$
is also
$B$
. In particular, all fibers of
$\pi {}_{Z}:Z\to B$
have dimension at least
$i1$
, and b) follows. □
5 Singularities of ample divisors in abelian varieties
This section contains the central results of this article. Let
$(A,\ell )$
be a polarized abelian variety. We study the singularities of a divisor in
$m\ell $
, with
$m>0$
, when the dimension is large enough with respect to the degree.
Theorem 3.
Let
$(A,\ell )$
be a simple polarized abelian variety of degree
$d$
and dimension
$g>(d+1{)}^{2}/4$
.

a)
Every divisor in
$\ell $
is prime, normal, and has rational singularities.

b)
If
$m\ge 2$
and
$D$
is a divisor in
$m\ell $
, the pair
$(A,\frac{1}{m}D)$
is log terminal unless
$D=mE$
, with
$E\in \ell $
.
For polarizations of degree
$2$
, these results can be improved: we can drop the assumption that
$A$
be simple.
Theorem 4 (Degrees
$1$
and
$2$
).
Let
$(A,\ell )$
be an indecomposable polarized abelian variety of degree
$d\le 2$
and dimension
$g$
.

a)
If
$g>d$
, every prime divisor in
$\ell $
is normal and has rational singularities.
Let
$D$
be a divisor in
$m\ell $
, with
$m\ge 2$
.

b)
If
$g\ge d$
, the pair
$(A,\frac{1}{m}D)$
is log canonical.

c)
If
$g>d$
and
$\lfloor \frac{1}{m}D\rfloor =0$
, the pair
$(A,\frac{1}{m}D)$
is log terminal.
Note that
$\ell $
may very well contain reducible elements (see § 6 ). As to c), we explain in Corollary 10 when the assumption
$\lfloor \frac{1}{m}D\rfloor =0$
fails to hold.

Proof of the theorems.
We set up the notation in order to give a uniform presentation for all cases.
Let
$L$
be an ample line bundle on
$A$
that represents
$\ell $
, let
$E$
be a prime divisor in
$\leftL\right$
, and let
${\mathsf{\Im}}_{0}={\mathsf{\Im}}_{{Z}_{0}}$
be the adjoint ideal
$\mathfrak{J}(A,E)$
.
Let also
${\mathsf{\Im}}_{t}={\mathsf{\Im}}_{{Z}_{t}}$
, for
$t\in (0,1]\cap \mathbf{Q}$
, be the multiplier ideal
$\mathsf{\Im}(A,\frac{t}{m}D)$
.
We have:
$$\begin{array}{ccc}{Z}_{0}=\mathbb{\varnothing}& \mathbb{\u27fa}& thepair(A,E)iscanonical\end{array}$$  
$$\begin{array}{ccc}{Z}_{t}=\mathbb{\varnothing}forallt\in (0,1)\cap \mathbf{Q}& \mathbf{\u27fa}& thepair(A,\frac{1}{m}D)islogcanonical\end{array}$$  
$$\begin{array}{ccc}{Z}_{1}=\mathbb{\varnothing}& \mathbb{\u27fa}& thepair(A,\frac{1}{m}D)islogterminal\end{array}$$  
We set as above, for
$P$
general in
$Pic0\left(A\right)$
,
$${h}_{t}={h}^{0}(A,L\otimes {\mathsf{\Im}}_{t}\otimes P)\in \{0,...,d\}$$
The point is to prove
${h}_{t}=d$
(Lemma 1 .c)). Note that in case b) of Theorem 3 , we may assume
$\lfloor \frac{1}{m}D\rfloor =0$
(Corollary 10 .a)) and that in case b) of Theorem 4 , we have
$\lfloor \frac{1}{m+1}D\rfloor =0$
(Corollary 10 .b)).
Case
$t\in (0,1)$
. It follows from the Kawamata–Viehweg vanishing theorem that
${V}_{>0}(L\otimes {\mathsf{\Im}}_{t})$
is empty ([
E]
, Proposition 1.4), hence
${h}_{t}>0$
(Lemma 1 .b)). If
$A$
is simple and
$g>(d+1{)}^{2}/4$
, Lemma 1 .e) yields
${h}_{t}=d$
. If
$d\le 2$
and
$(A,\ell )$
is indecomposable, we have
${h}_{t}\ne d1$
since
$g\ge d$
(Lemma 1 .d)), hence
${h}_{t}=d$
again. This proves Theorems 3 and 4 in that case.
Case
$t=0$
. We have
${V}_{0}(L\otimes {\mathsf{\Im}}_{0})=Pic0\left(A\right)$
, i.e. ,
${h}_{0}>0$
, because
$E$
is not fibered by (nonzero) abelian varieties ([
EL]
, Theorem 3). Moreover, the exact sequence ( 1 ) shows that Lemma 2 applies with
$\varepsilon =1$
and
$\mathfrak{F}$
with support
$E$
.
Case
$t=1$
. In this case,
$L\otimes {\mathsf{\Im}}_{1}$
is a direct summand of the pushforward of a dualizing sheaf,^{1 }
so that Lemma 2 applies with
$\varepsilon =0$
. Moreover, since
$g>d$
and
$\lfloor \frac{1}{m}D\rfloor =0$
, following the proof of [
H1]
, Theorem 1, one obtains
${h}_{1}>0$
.
So we may assume
$t\in \{0,1\}$
,
${h}_{t}>0$
, and that the conclusions of Lemma 2 hold.
If
$A$
is simple,
${V}_{>0}(L\otimes \mathsf{\Im})$
is finite, and Lemma 1 .e) implies Theorem 3 .
We now prove Theorem
4 . Since
${h}_{t}>0$
, we need only consider the case
$d=2$
.
If
${h}_{t}=1$
, since
${V}_{1}(L\otimes {\mathsf{\Im}}_{t})$
has codimension at least
$1$
, the scheme
${Z}_{t}$
is a single point and
${V}_{1}(L\otimes {\mathsf{\Im}}_{t})$
has dimension
$\ge g2$
(Lemma 1 .d)).
This contradicts the fact that
${V}_{1}(L\otimes {\mathsf{\Im}}_{t})$
has dimension
$0$
(Lemma 2 .b)).
Hence
${h}_{t}=2$
and Theorem 4 is proved. □
In view of the above results and further investigations, it is natural to make the following conjecture.
Conjecture 5.
Let
$(A,\ell )$
be a simple polarized abelian variety of degree
$d$
and dimension
$g>d$
.

a)
Every divisor in
$\ell $
is prime, normal, and has rational singularities.

b)
For any divisor
$D$
in
$m\ell $
, with
$m\ge 2$
, the pair
$(A,\frac{1}{m}D)$
is log terminal unless
$D=mE$
, with
$E\in \ell $
.
We can prove the conjecture for
$d\le 3$
. For
$g>4=d$
, we can prove that the pair
$(A,\frac{1}{m}D)$
is log terminal when
$A$
is general. For
$5\ge g\ge d$
, we can prove that the pair
$(A,\frac{1}{m}D)$
is log canonical when
$A$
is simple.
However, because of the technical nature of our arguments, we do not pursue this here.
We now interpret our results in terms of dimensions of loci of singularities.
Theorem 6.
Let
$(A,\ell )$
be a simple polarized abelian variety of degree
$d$
and dimension
$g>(d+1{)}^{2}/4$
. Let
$m$
and
$k$
be positive integers. For all
$D\in m\ell $
, we have
$$dimSin{g}_{mk}D<gk$$
unless
$k=1$
and
$D=mE$
, with
$E\in \ell $
.

Proof.
According to Theorem 3 , the hypotheses imply that the pair
$(A,D)$
is canonical for
$m=1$
and the pair
$(A,\frac{1}{m}D)$
is log terminal for
$m\ge 2$
, unless
$D=mE$
, with
$E\in \ell $
. Since the pair
$(A,E)$
is then also canonical, the theorem follows from ( 2 ). □
In the case of a polarization of degree
$2$
, we get a more precise result, analogous to [EL] , Corollary 2, and [H1] , Corollary 2.
Theorem 7.
Let
$(A,\ell )$
be an indecomposable polarized abelian variety of degree
$2$
and dimension
$g>2$
and let
$m$
and
$k$
be positive integers. The following properties are equivalent:

(i)
for some
$D$
in
$m\ell $
, the locus
$Sin{g}_{mk}D$
contains an irreducible component of codimension
$k$
in
$A$
;

(ii)
the polarized abelian variety
$(A,\ell )$
is a double étale cover of a product of
$k$
nonzero principally polarized abelian varieties.

Proof.
If (ii) holds, the polarization is represented by an étale cover of the theta divisor of a product of
$k$
nonzero principally polarized abelian varieties, hence (i) holds.
Assume (i). By Theorem
4 .c) and ( 2 ),
$D$
must have a component of multiplicity at least
$m$
. If
$D=mE$
, with
$E\in \ell $
prime, the pair
$(A,E)$
is canonical (Theorem 4 .b)) and by ( 2 ), this contradicts (i).
Therefore, by Corollary
10 .b), there are nonzero principally polarized abelian varieties
$({B}_{1},[{\Theta}_{1}\left]\right)$
and
$({B}_{2},[{\Theta}_{2}\left]\right)$
and an isogeny
$p:A\to {B}_{1}\times {B}_{2}$
such that
$$D=m{p}^{*}({\Theta}_{1}\times {B}_{2})+{p}^{*}({B}_{1}\times {D}_{2})$$
where
${D}_{2}\in \leftm{\Theta}_{2}\right$
.
If
$S$
is a component of
$Sin{g}_{mk}D$
of maximal dimension, there is an integer
$l\le k$
such that
$$S\subset Sin{g}_{l}{\Theta}_{1}\times Sin{g}_{m(kl)}{D}_{2}$$
From [
Ko1]
, Theorem 17.1, we get
$$codim(Sin{g}_{l}{\Theta}_{1})\ge l$$
From [
EL]
, Proposition 3.5, we get
$$codim(Sin{g}_{m(kl)}{D}_{2})\ge kl$$
Since
$S$
has codimension
$k$
, both inequalities must be equalities. By [
EL]
, Corollary 2,
$({B}_{1},[{\Theta}_{1}\left]\right)$
splits as a product of
$k$
nonzero principally polarized abelian varieties, and by [
H1]
, Corollary 2,
$({B}_{2},[{\Theta}_{2}\left]\right)$
splits as the product of at least
$kl$
principally polarized abelian varieties, so that (ii) holds. □
6 Reducible divisors in indecomposable polarizations
We gather in this last section elementary results on polarized abelian varieties that were used earlier.
Lemma 8.
Let
$(A,\ell )$
be an indecomposable polarized abelian variety. If the restriction of
$\ell $
to an abelian subvariety
${A}_{1}$
of
$A$
is principal, either
${A}_{1}=0$
or
${A}_{1}=A$
.

Proof.
Let
${A}_{2}$
be the neutral component of the kernel of
$$f:A\stackrel{{\phi}_{\ell}}{\u27f6}Pic0\left(A\right)\u27f6Pi{c}^{0}\left({A}_{1}\right)$$
The sum map
$p:{A}_{1}\times {A}_{2}\to A$
is an isogeny of polarized abelian varieties whose kernel is isomorphic to
${A}_{1}\cap {A}_{2}$
. Since
$\ell $
induces a principal polarization on
${A}_{1}$
, the restriction of
$f$
to
${A}_{1}$
is injective, i.e. ,
${A}_{1}\cap {A}_{2}=\left\{0\right\}$
and
$p$
is injective. □
We now study indecomposable polarizations that contain nonprime divisors. The situation is manageable when
$A$
is simple or the degree is
$2$
. For degrees at least
$3$
, more and more exceptional cases arise.
Proposition 9.
Let
$(A,\ell )$
be an indecomposable polarized abelian variety of degree
$d$
and dimension
$g\ge d$
such that
$\ell $
contains a nonprime divisor
$E$
.
a) The abelian variety
$A$
is not simple.
b) If
$d=2$
, there exist a decomposable principally polarized abelian variety
$(B,[\Theta \left]\right)$
and an isogeny
$p:A\to B$
of degree
$d$
such that
$E={p}^{*}\Theta $
.

Proof.
Write
$E={E}_{1}+{E}_{2}$
, with
${E}_{1}$
and
${E}_{2}$
effective and nonzero, and let, for
$j\in \{1,2\}$
,
${A}_{j}=K({E}_{j}{)}^{0}$
,
${B}_{j}=A/{A}_{j}$
, and
${g}_{j}=dim{B}_{j}>0$
.
Since
${A}_{1}\cap {A}_{2}$
is contained in
$K(\ell )$
, it is finite, hence
${g}_{1}+{g}_{2}\ge g$
. There is an ample divisor
${D}_{j}$
on
${B}_{j}$
which pulls back to
${E}_{j}$
, and
$$d=\frac{1}{g!}({E}_{1}+{E}_{2}{)}^{g}=\frac{1}{{g}_{1}!(g{g}_{1})!}{E}_{1}^{{g}_{1}}{E}_{2}^{g{g}_{1}}+\cdots +\frac{1}{(g{g}_{2})!{g}_{2}!}{E}_{1}^{g{g}_{2}}{E}_{2}^{{g}_{2}}$$
The first term of this sum is
$$\begin{array}{ccc}\frac{1}{(g{g}_{1})!}deg\left({D}_{1}\right)\left[{A}_{1}\right]\cdot {E}_{2}^{g{g}_{1}}& =& deg\left({D}_{1}\right)deg\left({E}_{2}{}_{{A}_{1}}\right)\end{array}$$  
$$\begin{array}{ccc}& =& deg\left({D}_{1}\right)deg(\ell {}_{{A}_{1}})>0\end{array}$$  
and similarly for the last term, hence all terms are positive^{2 }
integers.
When
$A$
is simple, we have
${g}_{1}={g}_{2}=g$
, hence
$d>g$
. This proves a).
We now assume
$d=2$
and prove b). By Lemma 8 ,
$\ell $
does not restrict to a principal polarization on
${A}_{j}$
unless
${A}_{j}=0$
, i.e. ,
${g}_{j}=g$
. The only possibility is
${g}_{1}+{g}_{2}=g$
, the polarization
$\left[{D}_{j}\right]$
on
${B}_{j}$
is principal, the map
$p:A\to {B}_{1}\times {B}_{2}$
is an isogeny, and
$E={p}^{*}({D}_{1}\times {B}_{2})+{p}^{*}({B}_{1}\times {D}_{2})$
. □
We use these results to bound the multiplicities of the components of elements of
$m\ell $
.
Corollary 10.
Let
$(A,\ell )$
be an indecomposable polarized abelian variety of degree
$d$
and dimension
$g\ge d$
and let
$D\in m\ell $
, with
$m>0$
.

a)
If
$A$
is simple, we have
$\lfloor \frac{1}{m+1}D\rfloor =0$
. Moreover,
$\lfloor \frac{1}{m}D\rfloor =0$
unless
$D=mE$
, with
$E\in \ell $
.

b)
If
$d=2$
, we have
$\lfloor \frac{1}{m+1}D\rfloor =0$
. Moreover,
$\lfloor \frac{1}{m}D\rfloor =0$
unless
$D=mE$
, with
$E\in \ell $
prime, or there are nonzero principally polarized abelian varieties
$({B}_{1},[{\Theta}_{1}\left]\right)$
and
$({B}_{2},[{\Theta}_{2}\left]\right)$
, and an isogeny
$p:A\to {B}_{1}\times {B}_{2}$
of degree
$2$
, such that
$$D=m{p}^{*}({\Theta}_{1}\times {B}_{2})+{p}^{*}({B}_{1}\times {D}_{2})$$
with
${D}_{2}\in \leftm{\Theta}_{2}\right$
.

Proof.
The arguments of the proofs of Lemmas 2.2 and 2.3 of [H1] yield:

$\bullet $
$\lfloor \frac{1}{m+1}D\rfloor =0$
, unless
$\ell $
contains a reducible divisor of the form
${E}_{1}+{E}_{2}$
, with
${E}_{2}$
ample and
${E}_{1}$
,
${E}_{2}$
, and
$D(m+1){E}_{1}$
effective nonzero;

$\bullet $
$\lfloor \frac{1}{m}D\rfloor =0$
, unless
$\ell $
contains a divisor of the form
${E}_{1}+{E}_{2}$
, with
${E}_{1}$
,
${E}_{2}$
, and
$Dm{E}_{1}$
effective.
The corollary therefore follows from the proposition. □
Remark 11.
Polarized abelian varieties that satisfy the condition in Proposition 9 .b) are all obtained as follows. Let
$d$
be any positive integer and, for
$j\in \{1,2\}$
, let
$({B}_{j},[{\Theta}_{j}\left]\right)$
be a nonzero principally polarized abelian variety of dimension
${g}_{j}$
.
Endow
$B={B}_{1}\times {B}_{2}$
with the product polarization
$[\Theta ]$
. Choose a point
${\beta}_{j}$
in
${B}_{j}$
of order
$d$
and consider the cyclic isogeny
$p:A\to B$
of degree
$d$
associated with the point
${\phi}_{\Theta}({\beta}_{1},{\beta}_{2})$
of
$Pi{c}^{0}\left(B\right)$
. The divisor
${p}^{*}\Theta $
is reducible and defines a polarization
$\ell $
of type
$\left(d\right)$
on
$A$
.
Let
${p}_{j}:{A}_{j}\to {B}_{j}$
be the degree
$d$
cyclic isogeny associated with
${\phi}_{{\Theta}_{j}}\left({\beta}_{j}\right)$
and let
${\ell}_{j}$
be the polarization
$\left[{p}_{j}^{*}{\Theta}_{j}\right]$
(of type
$\left(d\right)$
) induced on
${A}_{j}$
. There is a factorization
$${p}_{1}\times {p}_{2}:{A}_{1}\times {A}_{2}\stackrel{\pi}{\u27f6}A\stackrel{p}{\u27f6}B$$
Another way to construct
$(A,\ell )$
is to start from nonzero polarized abelian varieties
$({A}_{1},{\ell}_{1})$
and
$({A}_{2},{\ell}_{2})$
of type
$\left(d\right)$
, to choose elements
${\alpha}_{1}\in K\left({\ell}_{1}\right)$
and
${\alpha}_{2}\in K\left({\ell}_{2}\right)$
or order
$d$
, and to take the quotient of
${A}_{1}\times {A}_{2}$
by the subgroup generated by
$({\alpha}_{1},{\alpha}_{2})$
. For more details, see [
D1]
, Proposition 9.1.
Assume
$d$
is prime. A polarized abelian variety of degree
$d$
is decomposable if and only if it has a nonzero principally polarized abelian factor. It follows that the polarized abelian variety
$(A,\ell )$
obtained by the above construction is indecomposable if and only if both polarized abelian varieties
$({A}_{1},{\ell}_{1})$
and
$({A}_{2},{\ell}_{2})$
are indecomposable.
References

Debarre, O., Sur les variétés abéliennes dont le diviseur thêta est singulier en codimension
$3$
, Duke Math. J. 57 (1988), 221–273.

Debarre, O., Théorèmes de connexité et variétés abéliennes, Amer. J. Math. 117 (1995), 787–805.

Ein, L., Multiplier ideals, vanishing theorems and applications, Algebraic geometry—Santa Cruz 1995, 203–219, Proc. Sympos. Pure Math. 62, Part 1, Amer. Math. Soc., Providence, RI, 1997.

Ein, L., Lazarsfeld, R., Singularities of theta divisors and the birational geometry of irregular varieties, J. Amer. Math. Soc. 10 (1997), 243–258.

Esnault, H., Viehweg, E., Lectures on vanishing theorems, DMV Seminar 20, Birkhäuser Verlag, Basel, 1992.

Hacon, C., Divisors on principally polarized abelian varieties, Comp. Math. 119 (1999), 321–329.

Hacon, C., Fourier transforms, generic vanishing theorems and polarizations of abelian varieties, Math. Zeit. 235 (2000), 717–726.

Hacon, C., Pardini, R., Birational characterization of products of curves of genus
$2$
, Math. Research Letters 12 (2005), 129–140.

Kempf, G., appendix to: Mumford, D., Varieties defined by quadratic equations.1970 Questions on Algebraic Varieties (C.I.M.E., III Ciclo, Varenna, 1969), 29–100, Edizioni Cremonese, Rome.

Kollár, J., Shafarevich maps and automorphic forms, M. B. Porter Lectures. Princeton University Press, Princeton, NJ, 1995.

Kollár, J., Singularities of pairs, Algebraic geometry—Santa Cruz 1995, 221–287, Proc. Sympos. Pure Math. 62, Part 1, Amer. Math. Soc., Providence, RI, 1997.

Kollár, J., Higher direct images of dualizing sheaves II, Ann. of Math. 124 (1986), 171–202.

Mukai, S., Duality between
$D\left(X\right)$
and
$D\left(\hat{X}\right)$
with its application to Picard sheaves, Nagoya Math. J. 81 (1981), 153–175.
Mathematique – IRMA – UMR 7501, Universite Louis Pasteur, 7, rue Rene Descartes, 67084 Strasbourg Cedex, France Email address : debarre@math.ustrasbg.fr URL: http://wwwirma.ustrasbg.fr/
$\sim $
debarre Department of Mathematics, University of Utah, 155 South 1400 East, Room 233, Salt Lake City, UT 84112, USA Email address : hacon@math.utah.edu URL: http://www.math.utah.edu/
$\sim $
hacon