# Reading roadmap: Special subvarieties of non-arithmetic ball quotients and Hodge theory

The proposed proof constructs an integral variation of Hodge structure on a ball quotient, compares geometric and Hodge-theoretic special loci through monodromy and Mumford–Tate groups, applies Hodge Ax–Schanuel to atypical intersections, and counts definable parameter sets to obtain finiteness. A separate argument concerns commensurators, and later applications concern special points and fields of definition.

Source SHA-256: `fcdb1315e2897f17da85f004208a99faac821d4dcccd74aa63ae1e72832b6261`. Indexed 195,648 characters.

Edges point from prerequisite to consumer. Solid proof edges are attributed inferences, not formal certificates; references remain candidates.

## 1. Construct the variation

Integrality, total reality and a restriction-of-scalars construction supply the proposed integral variation.

Claims: integrality, totreal, thmzvhs, strategy1

## 2. Compare special structures

Monodromy and Mumford–Tate groups are used to compare geometric subspaces and period intersections.

Claims: thmmonodromy, cor123, Zspecial, mainthmproof, mainthm

## 3. Count positive-dimensional geometric loci

Ax–Schanuel gives atypicality constraints; definable parameter sets and a descending induction are used for maximal finiteness.

Claims: zasthm, mainpropwithas, thmbader

## 4. Separate consequences

The commensurator reproof, lattice-intersection corollary and special-point assertions have separate dependencies.

Claims: marguliscomm, cor1, propspepoints

## Internal claim index

### Theorem 1.2.1 — `thmbader`

Source lines 289–292.

If $\Gamma \subset G$ is non-arithmetic, then $S_\Gamma$ contains only finitely many maximal totally geodesic subvarieties.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: Let $m$ be the maximum of the dimensions of the strict special subvarieties of $S_\Gamma$. For any $j\leq m$ consider the set $\mathcal{E}_j := \{$maximal special subvarieties of $S_\Gamma$ of dimension $j\}$. Up to conjugation by elements of $\widehat{G}$, there are only a finite number of semisimple subgroups of $\widehat{G}$. Therefore, Proposition \ref{mainpropwithas} implies that there are only a finite number of special subvarieties of $S_\Gamma$ of maximal dimension $m$. That is, the set $\mathcal{E}_m$ is finite. We argue by induction (downward) on the $j$s for which $\mathcal{E}_j\neq \emptyset$. Let $\mathbf{F}$ be a semisimple subgroup of $\mathbf{G}$ for which there exists a spec

- `mainpropwithas` (proof_dependency; curated_source_check): Therefore, Proposition \ref{mainpropwithas} implies that there are only a finite number of special subvarieties of $S_\Gamma$ of maximal dimension $m$.
- `audit:induction` (proof_dependency; curated_source_check): \Sigma_2(\mathbf{F})^j:= \{\hat{g} \widehat{F}\hat{g}^{-1}: (x,\widehat{g})\in	\Sigma(\mathbf{F})^j, S_{x,\widehat{g}}\notin \mathcal{E}_{j'}, \forall j'>j\},
- `external:bader2020arithmeticity` (background; curated_source_check): The proof has then appeared in \cite{bader2020arithmeticity}.

Candidate references: `mainpropwithas` (candidate_proof_dependency); `sigmadef` (candidate_proof_dependency).

### Thm 2 — `nonarithasconjfinal`

Source lines 300–311.

Let $W \subset X \times S_\Gamma$ be an algebraic subvariety and $\Pi \subset X \times S_\Gamma$ be the graph of $\pi : X \to S_\Gamma$. Let $U$ be an irreducible component of $W \cap \Pi$ such that \codim U < \codim W + \codim \Pi, the codimension being in $X \times S_\Gamma$ or, equivalently, \dim W < \dim U + \dim S_\Gamma. If the projection of $U$ to $S_\Gamma$ is not zero dimensional, then it is contained in a strict totally geodesic subvariety of $S_\Gamma$.

*Extractive statement; proof gist not independently reconstructed.*

### Theorem 1.3.1 — `strategy1`

Source lines 317–320.

For every $\gamma \in \Gamma$, the trace of $\Ad (\gamma)$ lies in the ring of integers of a totally real number field $K$. As a consequence, up to conjugation by $G$, $\Gamma$ lies in $\mathbf{G}(\Oo_K)$. Moreover $\mathbb{V}$ induces a $\Z$-variation of Hodge structure $\widehat{\mathbb{V}}$ on $S_\Gamma$.

*Extractive statement; proof gist not independently reconstructed.*

- `thmzvhs` (proof_dependency; curated_source_check): The proof of Theorem \ref{strategy1} is eventually completed

### Theorem 1.3.2 — `mainthm`

Source lines 342–345.

The totally geodesic subvarieties of $S_\Gamma$ are precisely the $\Z$-special ones, equivalently the $\Gamma$-special ones.

*Extractive statement; proof gist not independently reconstructed.*

- `Zspecial` (proof_dependency; curated_source_check): The following is the converse of Proposition \ref{Zspecial} and their combination gives Theorem \ref{mainthm}.
- `mainthmproof` (proof_dependency; curated_source_check): The following is the converse of Proposition \ref{Zspecial} and their combination gives Theorem \ref{mainthm}.

### Corollary 1.3.3 — `cor1`

Source lines 349–352.

Let $H$ be a subgroup of $G=\PU(1,n)$ of hermitian type. If $\Gamma \cap H$ is Zariski dense in $H$, then $\Gamma \cap H$ is a lattice in $H$.

*Extractive statement; proof gist not independently reconstructed.*

- `cdk` (proof_dependency; curated_source_check): Theorem \ref{cdk} implies that the right hand side is algebraic.
- `finallemma` (proof_dependency; curated_source_check): Eventually Lemma \ref{finallemma} applies and shows that $\Gamma_F$ is a lattice in $F$, as desired.

### Cor 6 — `cor2`

Source lines 353–360.

Let $\Gamma$ be a non-arithmetic lattice in $G=\PU(1,n)$, $n>1$. Let $W$ be an irreducible subvariety of $S_\Gamma$ which is not contained in any of the finitely many maximal totally geodesic subvarieties of $S_\Gamma$. The image of \pi_1(W)\to \pi_1(S_\Gamma) \cong \Gamma gives rise to a thin subgroup of $\widehat{\mathbf{G}}(\Z)$, where $\widehat{\mathbf{G}}$ denotes the Weil restriction from $K$, the (adjoint) trace field of $\Gamma$, to $\Q$ of $\mathbf{G}$ (the $K$-form of $G$ determined by $\Gamma$).

*Extractive statement; proof gist not independently reconstructed.*

### Thm 7 — `thmfinal`

Source lines 363–366.

If $\Gamma \subset G= \PU(1,2)$ is non-arithmetic, then $S_\Gamma= \Gamma \backslash \mathbb{B}^2$ contains only finitely many one dimensional (complex) totally geodesic subvarieties.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: If $S_\Gamma$ contains no totally geodesic subvarieties, there is nothing to prove. Assume there is a totally geodesic subvariety $W$ of $S_\Gamma$, say associated to a triplet $(H=\PU(1,1),X_H=\mathbb{B}^1, \Gamma_H=\Gamma \cap H)$, such that $\Gamma_H$ is a lattice in $H$. Let $K$ be the field generated by the traces of $\Gamma$, under the adjoint representation. Thanks to Theorem \ref{strategy1}, it is a totally real number field and we assume, for simplicity, that $[K:\Q]=2$. Let $\mathbf{G}$ be the $K$-form of $G$ determined by $\Gamma$ (resp. $\mathbf{H}$ for the $K$-form of $H$), and write $\widehat{\mathbf{G}}$ for the Weil restriction from $K$ to $\Q$ of $\mathbf{G}$ (resp. $\wideha


Candidate references: `strategy1` (candidate_proof_dependency); `mainthm` (candidate_proof_dependency); `arithcrit` (candidate_proof_dependency); `mainpropsec` (candidate_proof_dependency); `zasthm` (candidate_proof_dependency).

### Defi 8 — `defi:line:444`

Source lines 444–450.

We define the (adjoint) trace field of $\Gamma$ as the field generated over $\Q$ by the set \{\tr \Ad (\gamma) : \gamma \in \Gamma\}.

*Extractive statement; proof gist not independently reconstructed.*

### Defi 9 — `defi:line:453`

Source lines 453–456.

An irreducible lattice $ \Gamma \subset G$ is locally rigid if there exists a neighbourhood $U$ of the inclusion $i: \Gamma \hookrightarrow G$ in $\Hom (\Gamma , G)$, such that every element of $U$ is conjugated to $i$.

*Extractive statement; proof gist not independently reconstructed.*

### Thm 10 — `coeffinnumbers`

Source lines 460–463.

{\cite[Theorem 0.11]{MR0267041}}

If $G$ is not locally isomorphic to $\Sl_2(\R)$, then, for every irreducible lattice $\Gamma$ in $G$ there exists $g\in G$ and a subfield $K \subset \R$ of finite degree over $\Q$, such that $g\Gamma g^{-1}$ is contained in the set of $K$-rational points of $\mathbf{G}$.

*Extractive statement; proof gist not independently reconstructed.*


Candidate references: `external:MR0267041` (external_reference).

### Thm 11 — `arithcrit`

Source lines 469–472.

Mostow–Vinberg arithmeticity criterion

Let $\Gamma \subset \mathbf{G}(\Oo_{K'})$ be a lattice in $G$. The lattice $\Gamma$ is arithmetic if and only if for every embedding $\sigma : K' \to \R $, not inducing the identity embedding of $K$ into $\R$, the group $\mathbf{G}_{K'} \otimes_{K',\sigma} \R$ is compact.

*Extractive statement; proof gist not independently reconstructed.*

### Rmk 12 — `infiniteindex`

Source lines 474–477.

If $\Gamma \subset \mathbf{G}(\Oo_K)$ is a non-arithmetic lattice, where $K$ is the trace field of $\Gamma$, then $\mathbf{G}(\Oo_K)$ is not discrete in $G$ and $\Gamma$ has infinite index in $\mathbf{G}(\Oo_K)$.

*Extractive statement; proof gist not independently reconstructed.*

### Prop 13 — `prop:line:485`

Source lines 485–491.

{\cite[Section 5.5]{MR3307755}}

The subgroup $\mathbf{G}(\Oo_K)$ of $G$ embeds as an arithmetic subgroup of $\widehat{\mathbf{G}}$ via the natural embedding \mathbf{G}(\Oo_K) \hookrightarrow \widehat{\mathbf{G}}, \ \ g \mapsto (\sigma (g))_{\sigma \in \Omega_\infty}.

*Extractive statement; proof gist not independently reconstructed.*


Candidate references: `external:MR3307755` (external_reference).

### Thm 14 — `margulisrank`

Source lines 497–500.

Arithmeticity

Let $G$ be a semisimple group without compact factors. If $G$ has rank at least $2$ and $\Gamma\subset G$ is an irreducible lattice, then $\Gamma$ is arithmetic.

*Extractive statement; proof gist not independently reconstructed.*

### Thm 15 — `superrig`

Source lines 502–505.

Superrigidity

Let $G$ be a semisimple group of rank at least $2$, and $\Gamma \subset G$ an irreducible lattice. Let $E$ be a local field of characteristic zero, and $\mathcal{G}'$ be an adjoint, absolutely simple $E$-group. Then every homomorphism $\Gamma \to \mathcal{G}'(E)$ with Zariski-dense and unbounded image extends uniquely to a continuous homomorphism $G\to \mathcal{G}'(E)$.

*Extractive statement; proof gist not independently reconstructed.*

### Thm 16 — `commcriterion`

Source lines 513–516.

Assume $G$ is without compact factors and $\Gamma$ is an irreducible lattice. Then $\Gamma$ is arithmetic if and only if it has an infinite index in $\Comm (\Gamma)$.

*Extractive statement; proof gist not independently reconstructed.*

### Defi 17 — `defi:line:534`

Source lines 534–537.

A $\Gamma$-factor of $G$ is either a one dimensional split torus in the centre of $G$, or a normal subgroup $N$ of $G\sim G'\times N$ such that the image of $\Gamma$ along the projection from $G $ to $N$ is a lattice in $N$. We call a $\Gamma$-factor irreducible, if the image of $\Gamma$ in $N$ is an irreducible lattice.

*Extractive statement; proof gist not independently reconstructed.*

### Prop 18 — `proponproducts`

Source lines 539–541.

The group $G$ can be written as a finite product of $\Gamma$-irreducible factors.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: Given $\Gamma$ in $G$ there is a unique direct product decomposition $G=\prod _{i\in I} G_i$, where $G_i$ is normal in $G$, such that $\Gamma_ {G_i}$ is an irreducible lattice in $G_i$ and $\prod_i \Gamma_{G_i}$ has finite index in $\Gamma$. See for example \cite[Theorem 5.22, page 86]{MR0507234}.


Candidate references: `external:MR0507234` (external_reference).

### Defi 19 — `realshimuradef`

Source lines 547–556.

A real Shimura datum is a triplet $(G,X, \Gamma)$ where $G$ is a real reductive algebraic group, $\Gamma \subset G$ a discrete subgroup such that its image in $G^{\ad}$ is a lattice, and $X$ is a $G$-orbit in the set of morphisms of algebraic groups $\Hom (\DT, G)$ such that for some (all) $x\in X$ the real Shimura--Deligne axioms are satisfied: \item[RSD0.] The image of $\Gamma$ in each $\Gl_2$-irreducible factor is an arithmetic lattice; \item[RSD1.] The adjoint representation $\Lie(G)$ is of type $\{(-1,1), (0,0), (1,-1)\}$; \item[RSD2.] The

*Extractive statement; proof gist not independently reconstructed.*

### Rmk 20 — `rmkfactorisa`

Source lines 557–560.

Let $X$ be a $G$-orbit in the set of morphisms of algebraic groups $\Hom (\DT, G)$ satisfying RSD1, RSD2 and RSD3. Let $x \in X$ be such that $x : \DT \to G$ factorises trough $H$, for some subgroup $H$ of $G$. Then the $H$ orbit of $x$ in $X$ satisfies RSD1 and RSD2. See for example \cite[Proposition 3.2]{MR2180407}.

*Extractive statement; proof gist not independently reconstructed.*


Candidate references: `external:MR2180407` (external_reference).

### Defi 21 — `defi:line:564`

Source lines 564–571.

Let $(G_1,X_{1}, \Gamma_{1})$ and $(G_2,X_2, \Gamma_2)$ be real Shimura data. A morphism of real Shimura data (G_1,X_{1}, \Gamma_{1})\to(G_2,X_2, \Gamma_2) is a morphism of real algebraic groups $f: G_1 \to G_2$ such that for each $x\in X_1$, the composition $f\circ x : \DT \to G_2$ is in $X_2$ and $f ( \Gamma_{1}) \subset \Gamma_2$.

*Extractive statement; proof gist not independently reconstructed.*

### Defi 22 — `defsubshimura`

Source lines 573–576.

Let $(G,X, \Gamma)$ a real Shimura datum, $H$ a subgroup of $G$ and $\Gamma_H$ a lattice in $H$. A real sub-Shimura datum is a triplet $(H,X_H, \Gamma_H)$ where $X_H$ is a $H$-orbit in the set $\Hom (\DT, G)$ satisfying RSD1, RSD2 and RSD3 of Definition \ref{realshimuradef}.

*Extractive statement; proof gist not independently reconstructed.*


Candidate references: `realshimuradef` (internal_reference).

### Defi 23 — `defi:line:580`

Source lines 580–587.

Let $(G,X, \Gamma)$ be a real Shimura datum. We define its adjoint real Shimura datum, simply denoted by (G,X, \Gamma)^{\ad}, as the triplet $(G^{\ad},X^{\ad},\Gamma^{\ad})$ where $X^{\ad}$ is the $G^{\ad}$-conjugacy class of morphisms $x : \DT \to G \to G^{\ad}$ and $\Gamma^{\ad}$ is the image of $\Gamma$ in $G^{\ad}$. It is again a real Shimura datum.

*Extractive statement; proof gist not independently reconstructed.*

### Defi 24 — `defi:line:590`

Source lines 590–593.

Let $(G,X,\Gamma)$ be a real Shimura datum. We denote by $S_\Gamma $ the quotient $\Gamma \backslash X$ and we refer to it as a Shimura variety.

*Extractive statement; proof gist not independently reconstructed.*

### Prop 25 — `irred`

Source lines 598–605.

Any real Shimura datum $(G,X,\Gamma)$ can be uniquely written, up to isogeny, as a product of a finite number of real sub-Shimura data $(G_i,X_i, \Gamma_i)$ such that the $\Gamma_i$ are irreducible lattices in $G_i$ and \prod_i S_{\Gamma _i }\to S_\Gamma is a finite covering (of complex manifolds).

*Extractive statement; proof gist not independently reconstructed.*

### Thm 26 — `vhsonsgamma`

Source lines 607–611.

{\cite[Proposition 1.1.14 and Corollary 1.1.17]{delignetravaux}}

Let $(G,X,\Gamma)$ be a real Shimura datum. Then $X$ has a unique structure of a complex manifold such that, for every real representation $\rho : G \to \Gl (V_\R)$, $(V_\R, \rho \circ h)_{h \in X}$ is a holomorphic family of Hodge structures. For this complex structure, each family $(V_\R, \rho \circ h)_{h \in X}$ is a variation of Hodge structure, and $X$ is a finite disjoint union of Hermitian symmetric domains.

*Extractive statement; proof gist not independently reconstructed.*


Candidate references: `external:delignetravaux` (external_reference).

### Rmk 27 — `rmk:line:615`

Source lines 615–618.

Let $(\mathbf{G}, \mathcal{X})$ be a Shimura datum in the sense of Deligne (i.e. $\mathbf{G}$ is assumed to be a $\Q$-group). For any faithful representation $\mathbf{G}\to \Gl(V_\Z) $, the triplet $(\mathbf{G}_\R, \mathcal{X}, \mathbf{G}(\Q)\cap \Gl(V_\Z))$ defines a real Shimura datum. Given a Shimura datum $(\mathbf{G}, \mathcal{X})$ as before and $K_{\mathbb{A}_f}$ a compact open subgroup of the finite adelic points of $\mathbf{G}$, the triplet $(\mathbf{G}_\R, \mathcal{X}, \mathbf{G}(\Q)\cap K_{\mathbb{A}_f})$ is a real Shimura datum. We r

*Extractive statement; proof gist not independently reconstructed.*

### Thm 28 — `algebraic`

Source lines 619–622.

Baily-Borel, Siu-Yau, Mok

Every Shimura variety $S_\Gamma$ has a unique structure of a quasi-projective complex algebraic variety.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: If the lattice $\Gamma$ is arithmetic, it was proven by Baily and Borel \cite{MR0216035}. To prove the result, we may and do replace $\Gamma$ by a finite index subgroup. In particular, thanks to Proposition \ref{irred}, we may assume that $\Gamma$ is irreducible. Thanks to Theorem \ref{margulisrank}, if $G$ has rank bigger or equal than two, then $\Gamma$ is arithmetic. Regarding rank one groups we have that $X$ is isomorphic to the complex unit ball $\mathbb{B}^n\subset \C^n$, for some $n \geq 1$. If $n=1$ the axiom RSD0 ensures that $\Gamma$ is arithmetic. If $n>1$, Siu-Yau \cite{MR645748} and Mok \cite{MR2884042} proved that $S_\Gamma$ is biholomorphic to a quasi-projective variety. To se


Candidate references: `irred` (candidate_proof_dependency); `margulisrank` (candidate_proof_dependency); `external:MR0216035` (external_reference); `external:MR645748` (external_reference); `external:MR2884042` (external_reference); `external:deng2020big` (external_reference).

### Defi 29 — `gammaspeci`

Source lines 629–636.

Let $W\subset S_\Gamma$ be an irreducible algebraic subvariety. We say that $W$ is $\Gamma$-special if it is induced by a real sub-Shimura datum $(H,X_H, \Gamma_H)$ of $(G,X, \Gamma)$. We say that $W$ is weakly-$\Gamma$-special if there exists a real sub-Shimura datum $(H,X_H, \Gamma_H)$ of $(G,X, \Gamma)$ such that its adjoint splits as a product (H,X_H, \Gamma_H)^{\ad}=(H_1,X_1, \Gamma_1) \times (H_2,X_2, \Gamma_2), and $W$ is the image of $X_1^+\times \{x_2\}$ for some $x_2\in X_2$.

*Extractive statement; proof gist not independently reconstructed.*

### Defi 30 — `defi:line:641`

Source lines 641–644.

Given a real Shimura datum $(G,X,\Gamma)$, we say that $S_\Gamma$ is an arithmetic Shimura variety if $\Gamma$ is arithmetic. If $\Gamma$ is irreducible and non-arithmetic, we say that $S_\Gamma$ is a non-arithmetic ball quotient.

*Extractive statement; proof gist not independently reconstructed.*

### Defi 31 — `defparabolci`

Source lines 679–682.

A parabolic subgroup $P\subset G$ is called $\Gamma$-rational if its unipotent radical $N_P$ intersects $\Gamma$ as a lattice.

*Extractive statement; proof gist not independently reconstructed.*

### Thm 32 — `thm:line:685`

Source lines 685–688.

{\cite[Corollary 7.6, Chapter III]{MR2884042}}

The toroidal compactification $\overline{S_\Gamma}$ of $S_\Gamma$ is a smooth compactification with strict normal crossing divisor at infinity $B:= \overline{S_\Gamma} - S_\Gamma$.

*Extractive statement; proof gist not independently reconstructed.*


Candidate references: `external:MR2884042` (external_reference).

### Prop 33 — `quasiunipot`

Source lines 691–694.

For any $\Gamma$-rational parabolic $P$, the set $U_P \cap \Gamma$ is isomorphic to $\Z$. The image of the fundamental group of $U_\alpha \cap S_\Gamma= \Delta^{n-1}\times \Delta^*$ in the fundamental group of $\overline{S_\Gamma}$ lies in $\Gamma_{U_P}\cong \Z$.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: The first part can be found in \cite[Section 1.3]{MR2884042}. For the second part, we work with the local coordinates as in \cite[pages 255-256]{MR471627}. Let $X^\vee$ be the compact dual of $X$ and $\mathcal{X}\subset X^\vee$ be the Borel embedding. Assume that we are working with a $\Gamma$-rational boundary component $F=\{b\}$ corresponding to a $\Gamma$-rational parabolic $P$, and let $V$ be the quotient of the unipotent radical of $P$ by it centre. It is a real vector space of rank $n-1$ (where $n=\dim S_\Gamma$). Set \mathcal{X}_b := \bigcup _{g\in U_P \otimes \C} g \cdot \mathcal{X} \subset X^\vee. There exists a canonical holomorphic isomorphism j: \mathcal{X} \cong \C^{n-1}\times \


Candidate references: `external:MR2884042` (external_reference); `external:MR471627` (external_reference).

### Defi 34 — `poincaregrowthdef`

Source lines 756–762.

We say that $\omega$ is of Poincaré-growth with respect to the toroidal compactification $\overline{S_\Gamma}$ if, for any $\alpha$, $\omega$ restricted to $U_\alpha$ is bounded by C_\alpha \left( \sum_{k=1}^{n-1} |dz_k|^2 + \frac{|dz_n|^2}{|z_n|^2|\log z_n|^2} \right).

*Extractive statement; proof gist not independently reconstructed.*

### Thm 35 — `poincaregrowth`

Source lines 764–767.

{\cite[Section 1.2]{MR2884042}} and {\cite[Section 2]{MR1238407}}

The K\"{a}hler form $\omega$ on $S_\Gamma$ is of Poincaré-growth.

*Extractive statement; proof gist not independently reconstructed.*


Candidate references: `external:MR2884042` (external_reference); `external:MR1238407` (external_reference).

### Prop 36 — `siegelanddn`

Source lines 802–808.

Let $\Sigma =\Sigma _{t, \Omega}$ be a Siegel set in $X$ for the action of $\Gamma$. Then $ \Sigma$ is covered by a finite union of open subsets $\Theta$ with the following properties. For each $\Theta$ there exists $\Gamma$-rational boundary component $F=\{b\}$ corresponding to a $\Gamma$-rational parabolic $P$, a positive integer $N$ large enough, relatively compact subsets $U' \subset U(P)$, $Y'\subset \C^{n-1}$ such that $\Theta$, up to a bi-isometry, can be described as \{(z_1,\dots , z_n)\in \C^{n-1}\times U(P)_\C, \operatorname{Re}(z_n)\

*Extractive statement; proof gist not independently reconstructed.*

### Defi 37 — `defi:line:812`

Source lines 812–819.

A fundamental set for the action of $\Gamma$ on $X$ is a connected open subset $\F$ of $X$ such that $\Gamma \cdot \overline{\F}=X$ and the set \{\gamma \in \Gamma : \gamma \F \cap \F \neq \emptyset\} is finite.

*Extractive statement; proof gist not independently reconstructed.*

### Thm 38 — `fundamentalset`

Source lines 820–832.

Garland-Raghunathan

For any Siegel set $\Sigma_{t, \Omega}$ the set \{\gamma \in \Gamma : \gamma \Sigma_{t,\Omega} \cap \Sigma_{t,\Omega}\neq \emptyset \} is finite. There exists a Siegel set $\Sigma_{t_0, \Omega}$ and a finite subset $\Xi$ of $G$ such that: \item For all $b \in \Xi$, $\Gamma \cap b N_P b^{-1}$ is a lattice in $b N_P b^{-1}$; \item The set $\Xi \cdot \Sigma_{t_0, \Omega}$ is a fundamental set for the action of $\Gamma$ on $X$. Furthermore, when $\Omega$ is chosen to be semialgebraic, the associated Siegel sets $\Sigma_{t,\Omega}$ are semialgebraic

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: The first part of the statement is \cite[Theorem 4.10]{MR0267041}. It is formulated for admissible discrete subgroups of $G$ (see Definition 0.4 in loc. cit.), a class that includes any lattice $\Gamma \subset G$, as proved in Theorem 0.7 of loc. cit.. When $\Omega$ is chosen to be semialgebraic, the Siegel set for $G$ \Omega \cdot A_t \cdot K_0 is semialgebraic since it is defined as product in $G$ of semialgebraic sets and therefore its image $\Sigma_{t,\Omega}$ in $X$ is again semialgebraic. Finally, since $\Xi$ is finite, $\Xi \cdot \Sigma_{t_0, \Omega}$ is semialgebraic.


Candidate references: `external:MR0267041` (external_reference).

### Prop 39 — `funct`

Source lines 844–847.

Let $(H,X_H, \Gamma_H)$ be a real sub-Shimura datum of $(G=\PU(1,n), X, \Gamma)$. Any Siegel sets of $H$ is contained in a Siegel set for $G$.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: Since we are interested only in Siegel set in $X_H$ and $X$, as recalled in Section \ref{embeddings}, we may assume that $H=\PU(1,m)$ for some $m<n$. Let $P$ be a minimal $\Gamma$-rational parabolic subgroup of $P$ and $P_H$ be $P\cap H$. As above, consider the decompositions relative to $K_0 $ and $K_0 \cap H$ P=N_PAM, \ \ P_H= N_H A_H M_H. As explained in Section \ref{embeddings} we have that $N_P\cap P_H = N_H$, $A \cap P_H = A_H$ and $M\cap P_H=M_H$. The result then follows by the construction of Siegel sets explained above.


Candidate references: `embeddings` (candidate_proof_dependency).

### Thm 40 — `defini`

Source lines 862–865.

Let $(G,X,\Gamma)$ be a real Shimura datum. There exists a semi--algebraic fundamental set $\F$ for the action of $\Gamma$ in $X$ such that the restriction $\pi: \F \to S_\Gamma$ of the uniformising map $\pi : X \to S_\Gamma$ is definable in $\R_{\operatorname{an, exp}}$.

*Extractive statement; proof gist not independently reconstructed.*

### Cor 41 — `algmorphism`

Source lines 868–871.

Let $(H,X_H, \Gamma_H)$ be a real sub-Shimura datum of $(G=\PU(1,n), X, \Gamma)$. The induced map $S_{\Gamma_{H}}\to S_{\Gamma}$ is algebraic.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: By definition $(H,X_H, \Gamma_H) \to (G, X, \Gamma)$ induces a commutative diagram [scale=2] \node (A) at (-1,1) {$X_H$}; \node (B) at (1,1) {$X$}; \node (C) at (-1,0) {$S_{\Gamma_H}$}; \node (D) at (1,0) {$S_{\Gamma}$}; \path[->,font=\scriptsize,>=angle 90] (A) edge node[above]{$\tilde{i}$} (B) (C) edge node[above]{$i$} (D) (A) edge node[right]{} (C) (B) edge node[right]{} (D); . For some $x \in X_H$, we can write the graph of $\tilde{i}$ as \{h.(x, \tilde{i}(x)), \ h \in H\}. Let $\F_H$ be a fundamental set for the action of $\Gamma_H$ in $X_H$. By Proposition \ref{funct}, we have that the graph of $\tilde{i}$ restricted to $\F_H$ is a definable set. Let $\F$ be a fundamental set for $\Gam


Candidate references: `funct` (candidate_proof_dependency); `defini` (candidate_proof_dependency).

### Thm 42 — `ogaga`

Source lines 897–898.

{\cite[Theorem 4.4 and Corollary 4.5]{MR2492989}}

Let $S$ be a smooth complex algebraic variety. Let $Y \subset S$ be a closed complex analytic subset which is also a definable subset (for some o-minimal structure expanding $\R_{\an}$). Then $Y$ is an algebraic subvariety of $S$.

*Extractive statement; proof gist not independently reconstructed.*


Candidate references: `external:MR2492989` (external_reference).

### Theorem 4.1.1 — `integrality`

Source lines 925–928.

Let $\Gamma$ be a lattice in $G=\PU(1,n)$ for some $n>1$, $K$ its trace field and $\mathbf{G}$ the $K$-form of $G$ determined by $\Gamma$. There exists a finite index subgroup $\Gamma ' \subset \Gamma $ with integral traces. Equivalently, up to conjugation by $G$, $\Gamma'$ lies in $\mathbf{G}(\Oo_K)$, where $\Oo_K$ is the ring of integers $K$.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: In the proof we are free to replace $\Gamma$ by a finite index subgroup, and so may and do assume that, as in Section \ref{toroidalsubse}, $\Gamma$ is torsion free and that the parabolic subgroups of $\Gamma$ are unipotent. Moreover, as explained before, we assume also that $\Gamma$ lifts to $\SU(1,n)$. Let $S_\Gamma$ be the ball quotient $\Gamma \backslash X$. Let $\mathbb{V}$ be the local system on $S_\Gamma $ associated to the standard representation of $\SU(1,n)$, as in (\ref{locsystem}). Thanks to Bass--Serre theory \cite{MR586867} (or \cite[Theorem 2]{MR0279206}), $\mathbb{V}$ is integral in the sense of Theorem \ref{esgro} if and only if the image of every $\gamma \in \Gamma$ in $\Gl(


Candidate references: `toroidalsubse` (candidate_proof_dependency); `locsystem` (candidate_proof_dependency); `esgro` (candidate_proof_dependency); `compac` (candidate_proof_dependency); `adjointrep` (candidate_proof_dependency); `external:MR586867` (external_reference); `external:MR0279206` (external_reference); `external:MR2457528` (external_reference).

### Thm 44 — `garalndragvanishing`

Source lines 931–937.

{\cite[Theorem 1.10]{MR0267041}}

Let $G$ be a semisimple Lie group, not locally isomorphic to $\Sl_2$, nor to $\Sl_2(\C)$. For any lattice $\Gamma$ in $G$, the first Eilenberg--MacLane cohomology group of $\Gamma$ with respect to the adjoint representation is zero. In symbols: H^1(\Gamma, \Ad)=0.

*Extractive statement; proof gist not independently reconstructed.*


Candidate references: `external:MR0267041` (external_reference).

### Thm 45 — `esgro`

Source lines 947–956.

{\cite[Theorem 1.1]{MR3874695}}

Let $S$ be a smooth connected quasi-projective complex variety. Then a complex local system $\mathbb{V}$ on $S$ is integral, i.e. it comes as extension of scalars from a local system of projective $\Oo_L$-modules of finite type (for some number field $L\subset \C$), whenever it is: \item Irreducible; \item Quasi-unipotent local monodromies around the components at infinity of a compactification with normal crossings divisor $i:S \hookrightarrow \overline{S}$; \item Cohomologically rigid, that is $\Hh^1(\overline{S}, i_{!* } \End^0(\mathbb{V}))$

*Extractive statement; proof gist not independently reconstructed.*


Candidate references: `external:MR3874695` (external_reference).

### Thm 46 — `deformation`

Source lines 1013–1016.

{\cite[Theorem 8.1]{MR2457528}}

Suppose $\mathbb{V}$ is local system with quasi-unipotent monodromy at infinity. If $\mathbb{V}$ is rigid, then it underlines a $\C$-VHS.

*Extractive statement; proof gist not independently reconstructed.*


Candidate references: `external:MR2457528` (external_reference).

### Theorem 4.2.2 — `totreal`

Source lines 1019–1022.

Let $\Gamma$ be a lattice in $G=\PU(1,n)$, $n>1$. Then the trace field $K$ of $\Gamma$ is totally real.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: Possibly after conjugating, the local system \eqref{locsystem} $\rho: \pi_1(S_\Gamma)\to \Gl(V_\C)$ takes values in some field $K'$. For each $\sigma : K' \to \C$, consider \rho_\sigma = \sigma \rho : \pi_1(S_\Gamma)\to \Gl(V_\C). The first step is to prove that the local system $\mathbb{V}^\sigma$ corresponding to $\rho_\sigma$ is in fact a VHS (either a $K'$ or a $\C$-VHS). Since H^1(\Gamma , \Ad \circ \rho_\sigma)=H^1(\Gamma, \Ad) \otimes_{K',\sigma} \C=0, the twisted representation $\rho_\sigma$ is again cohomologically rigid (see also \cite[Lemma 6.6]{MR2457528}). Moreover $\mathbb{V}^\sigma$ has also quasi-unipotent monodromy at infinity by construction. Indeed, let $P_i$ be a $\Gamma$

- `audit:self-dual` (proof_dependency; curated_source_check): As $\rho$ is a unitary representation, and therefore self dual

Candidate references: `locsystem` (candidate_proof_dependency); `deformation` (candidate_proof_dependency); `localrig` (candidate_proof_dependency); `integrality` (candidate_proof_dependency); `adjointrep` (candidate_proof_dependency); `external:MR2457528` (external_reference); `external:MR1179076` (external_reference); `external:MR0279206` (external_reference).

### Rmk 48 — `rmk:line:1045`

Source lines 1045–1048.

For each embedding $\sigma : K' \to \R$, let $G_\sigma$ be the group group $\mathbf{G} \times_{K', \sigma}\R$. Since the groups $G_\sigma$ are all isomorphic over $\C$, if $G=\PU(1,n)$, the proof of Theorem \ref{totreal} shows that we can write $G_\sigma$ as $\PU(p_\sigma, q_\sigma)$ for some positive non zero integers $p_\sigma, q_\sigma$ such that $p_\sigma+q_\sigma =n+1$. See also \cite[Lemma 4.5]{MR1179076}.

*Extractive statement; proof gist not independently reconstructed.*


Candidate references: `totreal` (internal_reference); `external:MR1179076` (external_reference).

### Theorem 4.2.4 — `thmzvhs`

Source lines 1049–1061.

Let $(G,X,\Gamma)$ be a real Shimura datum. There exists a totally real number field $K'$ such that: \item $K'$ contains $K$ the (adjoint) trace field of $\Gamma$; \item $\Gamma \subset \mathbf{G}(\Oo_{K'})$; \item For each embedding $\sigma : K' \to \R$, if we let $\mathbb{V}^\sigma$ be the $K'$-VHS constructed above, then \widehat{\mathbb{V}} := \bigoplus_{\sigma : K'\to \R} \mathbb{V}^\sigma has a natural structure of (polarized) $\Q$-VHS. Moreover, by Theorem \ref{integrality} and Bass--Serre theory, it has a natural structure of a (polaris

*Extractive statement; proof gist not independently reconstructed.*

- `integrality` (proof_dependency; curated_source_check): Moreover, by Theorem \ref{integrality} and Bass--Serre theory
- `totreal` (proof_dependency; curated_source_check): After having established Theorem \ref{totreal}

Candidate references: `integrality` (internal_reference).

### Defi 50 — `defi:line:1135`

Source lines 1135–1138.

A Mumford--Tate sub-domain $D_0$ of $D'$ is an orbit $M.x$ where $x\in D'$ and $M$ is the real group associated to $\mathbf{M}$ a normal algebraic $\Q$-subgroup of $\MT(x)$. In fact $D_0$ is a complex sub-manifold of $D'$ and $\pi(D_0)=\mathbf{M}(\Z)\backslash M.x$ is a complex analytic subvariety of $\widehat{\mathbf{G}}'(\Z)\backslash D'$, which we call a Mumford--Tate subvariety.

*Extractive statement; proof gist not independently reconstructed.*

### Theorem 4.3.2 — `cdk`

Source lines 1140–1143.

{\cite[Theorem 1.1]{MR1273413}}

Let $D_0 \subset D'$ be a Mumford--Tate sub-domain. Then the set $\psi ^{-1}(\pi (D_0))$ is an algebraic subvariety of $S_\Gamma$.

*Extractive statement; proof gist not independently reconstructed.*


Candidate references: `external:MR1273413` (external_reference).

### Defi 52 — `defi:line:1144`

Source lines 1144–1147.

Subvarieties of $S_\Gamma$ of the form $\psi ^{-1}(\pi (D_0))^0$, for some Mumford--Tate sub-domain $D_0\subset D'$ are called $\Z$-special (here $(-)^0$ denotes some irreducible component). An irreducible subvariety $W$ of $S_\Gamma$ is $\Z$-Hodge generic if it is not contained in any strict $\Z$-special subvariety.

*Extractive statement; proof gist not independently reconstructed.*

### Lemma 53 — `lemmanoncomp`

Source lines 1176–1179.

The $K$-algebraic group $\leftidx{^\Gamma} {\mathbf{M}}_W$ is non-trivial and its real points are not compact.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: Heading for a contradiction suppose that the intersection between $\Gamma$ and $\leftidx{^\Gamma} {M}_W$ is finite. By Riemann existence theorem, this implies that $\widetilde{W}$ is algebraic and therefore does not admit bounded holomorphic functions. This, however, contradicts the fact that $X=\mathbb{B}^n$ admits a bounded realisation. Finally, since $\Gamma \cap \leftidx{^\Gamma} {M}_W $ is a discrete and infinite subgroup of $\leftidx{^\Gamma} {M}_W$, $\leftidx{^\Gamma} {M}_W$ can not be compact.

### Defi 54 — `deftracefield`

Source lines 1183–1186.

The field generated by the traces of the elements in $\Gamma \cap \leftidx{^\Gamma} {M}_W $, under the adjoint representation (of $M_W$), is called the trace field of $W$ and denoted by $K_W$.

*Extractive statement; proof gist not independently reconstructed.*

### Rmk 55 — `rmknew`

Source lines 1186–1193.

The field $K_W$ is naturally a sub-field of $K$. Indeed, by construction, the field L:= \Q\{\tr \ad_{|{M_W}}(\gamma): \gamma \in \Gamma \cap \leftidx{^\Gamma} {M}_W \} is a sub-field of $K$, where $\ad_{|{M_W}}$ denotes the restriction of the adjoint representation of $G$ to $M_W$. By Vinberg's theorem \cite{MR0279206}, as explained in Section \ref{localrig}, we have that the adjoint trace field of $\Gamma \cap \leftidx{^\Gamma} {M}_W$ is a sub-field of $L$, and therefore of $K$.

*Extractive statement; proof gist not independently reconstructed.*


Candidate references: `localrig` (internal_reference); `external:MR0279206` (external_reference).

### Theorem 5.1.4 — `thmmonodromy`

Source lines 1211–1217.

The $\Z$-monodromy of $W$ is the Weil restriction from $K_W$ to $\Q$ of the $\Gamma$-monodromy of $W$. In symbols we write: \widehat{\leftidx{^\Gamma} {\mathbf{M}}_W}=\leftidx{^\Z} {\mathbf{M}}^{\ad}_W.

*Extractive statement; proof gist not independently reconstructed.*

- `audit:embedding-images` (proof_dependency; curated_source_check): we obtain that $\sigma_i(K_W)=\sigma_j(K_W)$. This is impossible

### Lemma 57 — `goursat`

Source lines 1258–1261.

Goursat's Lemma

Let $G_1, G_2$ be algebraic groups and $H$ be a closed subgroup of $G_1 \times G_2$ such that the two projections from $H$ to $G_1$ and $G_2$ are surjective. For $i=1,2$, let $K_i$ be the kernel of $H\to G_i$. The image of $H$ in $G_1/K_2 \times G_2/K_1$ is the graph of an isomorphism $G_1/K_2 \to G_2/K_1$.

*Extractive statement; proof gist not independently reconstructed.*

### Thm 58 — `deligneandre`

Source lines 1311–1314.

Deligne, André

Let $W$ a smooth quasi-projective variety supporting a $\Z$-VHS and $s \in W$ a Hodge generic point. Let $\mathbf{M}_s$ be the algebraic monodromy of $W$ (at $s$). Then $\mathbf{M}_s$ is a normal subgroup of the derived subgroup $\MT(s)^{\der}$ of the Mumford--Tate group at $s$. In symbols we simply write $\mathbf{M}_s \triangleleft \MT(s)^{\der}$.

*Extractive statement; proof gist not independently reconstructed.*

### Corollary 5.2.2 — `gammadense`

Source lines 1317–1320.

The $\Z$-monodromy group of $S_\Gamma$ is $\widehat{\mathbf{G}}$, i.e. $\Gamma$ has Zariski dense image in $\widehat{G}$. Moreover the derived subgroup of the generic Mumford--Tate group of $\widehat{\mathbb{V}}$ is $\widehat{\mathbf{G}}$.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: By definition, the $\Gamma$-monodromy of $S_\Gamma$ is given by $\mathbf{G}$. Theorem \ref{thmmonodromy}, applied to $W=S_\Gamma$, shows that the $\Z$-monodromy of $S_\Gamma$ is $\widehat{\mathbf{G}}$, since $K_W$ is just $K$, the trace field of $\Gamma$. By Theorem \ref{deligneandre}, we conclude that generic Mumford--Tate group of $\widehat{\mathbb{V}}$ is $\widehat{\mathbf{G}}$.

- `thmmonodromy` (proof_dependency; curated_source_check): Theorem \ref{thmmonodromy}, applied to $W=S_\Gamma$

Candidate references: `thmmonodromy` (candidate_proof_dependency); `deligneandre` (candidate_proof_dependency).

### Rmk 60 — `remaknew`

Source lines 1324–1327.

As a direct consequence of Corollary \ref{gammadense}, we have the following. Let $S_{\Gamma_H}$ be a $\Gamma$-special subvariety of $S_\Gamma$ associated to a triplet $(H,X_H,\Gamma_H\subset \mathbf{H}(\Oo_K))\subset (G,X,\Gamma\subset \mathbf{G}(\Oo_K))$. The Mumford--Tate group of $\widehat{\mathbb{V}}$ restricted to the (smooth locus of) $S_{\Gamma_H}$ is the Weil restriction from $K_H$ to $\Q$ of $\mathbf{H}$, where $K_H$ is the subfield of $K$ given by the trace field of the lattice $\Gamma_H$.

*Extractive statement; proof gist not independently reconstructed.*


Candidate references: `gammadense` (internal_reference).

### Rmk 61 — `rmk:line:1375`

Source lines 1375–1381.

We notice here that, by Theorem \ref{arithcrit}, the group $\prod_{i=2,\dots, r}G_{\sigma_i} $ is compact if and only if $\Gamma$ is an arithmetic lattice. This will be a fundamental step in Section \ref{sectionfinit}, whose starting point is the following: $\Gamma$ is non-arithmetic if and only if $\codim_{\widehat{\mathbf{G}}(\Z)\backslash D} ({\psi({S_\Gamma}^{\an})})>0$.

*Extractive statement; proof gist not independently reconstructed.*


Candidate references: `arithcrit` (internal_reference); `sectionfinit` (internal_reference).

### Corollary 5.2.5 — `cor123`

Source lines 1382–1385.

Let $\mathbf{H}/\Q$ be the the generic Mumford--Tate group of a subvariety (of positive dimension) $W\subset S_\Gamma$. Then $\mathbf{H}/\Q$ is a Weil restriction from $K_W$ to $\Q$ of some subgroup $\mathbf{F}$ of $\mathbf{G}$.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: Let $W$ be a (smooth, irreducible, closed) subvariety of $S_\Gamma$. Applying Theorem \ref{deligneandre} to $W$, we see that \leftidx{^\Z} {\mathbf{M}}_W \triangleleft \MT(w)^{\der} \subseteq \widehat{\mathbf{G}}, where $w\in W$ is a point whose Mumford--Tate group is $\mathbf{H}$, the generic Mumford--Tate of $W$. That is a Hodge generic point, in the sense of Section \ref{recapperioddomains}. Theorem \ref{thmmonodromy} shows that the left hand side is the Weil restriction, from $K_W$ to $\Q$, of some subgroup $\mathbf{F} \subset \mathbf{G}$. Since the centraliser of $\mathbf{F}_\R$ in $\mathbf{G}_\R$ is finite (being $G$ of rank one), we conclude that \widehat{\mathbf{F}}= \leftidx{^\Z} {\

- `thmmonodromy` (proof_dependency; curated_source_check): Theorem \ref{thmmonodromy} shows that the left hand side is the Weil restriction
- `audit:centralizer` (proof_dependency; curated_source_check): Since the centraliser of $\mathbf{F}_\R$ in $\mathbf{G}_\R$ is finite (being $G$ of rank one)

Candidate references: `deligneandre` (candidate_proof_dependency); `recapperioddomains` (candidate_proof_dependency); `thmmonodromy` (candidate_proof_dependency).

### Proposition 5.3.1 — `Zspecial`

Source lines 1398–1405.

$\Gamma$-special $\Rightarrow$ $\Z$-special

Let $(H,X_H,\Gamma_H\subset \mathbf{H}(\Oo_K))$ be a real sub-Shimura datum of $(G,X,\Gamma\subset \mathbf{G}(\Oo_K))$. Denote by $\widehat{\mathbf{H}}$ the Weil restriction from $K$ to $\Q$ of $\mathbf{H}$. The subvariety $S_{\Gamma_H}$ of $S_\Gamma$ can be written an irreducible component of the preimage along the period map \psi : {S_\Gamma }^{\an}\to \widehat{\mathbf{G}}(\Z) \backslash D_{\widehat{G}} of the Mumford-Tate sub-domain $\widehat{\mathbf{H}}(\Z) \backslash D_{\widehat{H}} \subset \widehat{\mathbf{G}}(\Z) \backslash D_{\widehat{G

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: As explained in Theorem \ref{algebraic} and Corollary \ref{algmorphism}, $S_{\Gamma_H}$ is an irreducible algebraic subvariety of $S_\Gamma$. We can restrict the $\Z$-VHS $\widehat{\mathbb{V}}$ from $S_\Gamma$ to (the smooth locus of\footnote{The variety $\Gamma_H \backslash X_H$ is smooth (at least when $\Gamma$ is torsion free), but it may be that its image in $S_\Gamma$ is not smooth.}) $S_{\Gamma_H}$. From Corollary \ref{gammadense} (see indeed Remark \ref{remaknew}), we know that the \MT(\widehat{\mathbb{V}}_{| S_{\Gamma_H}})^{\der}= \widehat{\mathbf{H}}^{\der}. Here $\widehat{\mathbf{H}}$ denotes the Weil restriction from $K$ to $\Q$ of $\mathbf{H} \subset \mathbf{G}/K$. To be more pre


Candidate references: `algebraic` (candidate_proof_dependency); `algmorphism` (candidate_proof_dependency); `gammadense` (candidate_proof_dependency); `remaknew` (candidate_proof_dependency); `twofields` (candidate_proof_dependency); `eqmtnew` (candidate_proof_dependency); `section46` (candidate_proof_dependency).

### Rmk 64 — `twofields`

Source lines 1439–1446.

It could happen that $\mathbf{H}$ is defined over a smaller field (but not over a bigger one, as explained in Remark \ref{rmknew}). The trace field of $\Gamma_H$, which we denote by $K_H=K_{S_{\Gamma_H}}$, may be strictly contained in $K$ (it could even happen that $K_H=\Q$). In this case we can consider the Weil restriction $\Res ^{K_H}_\Q \mathbf{H}$ and the associated period domain \tilde{D}:= D_{\Res ^{K_H}_\Q \mathbf{H}(\R)} \subset D_{\widehat{G}}. We remark here that $S_{\Gamma_H}$ can also be written as $\psi^{-1}\pi(\tilde{D})^0$. Inde

*Extractive statement; proof gist not independently reconstructed.*


Candidate references: `rmknew` (internal_reference); `Zspecial` (internal_reference).

### Proposition 5.3.3 — `mainthmproof`

Source lines 1453–1456.

$\Z$-special $\Rightarrow$ $\Gamma$-special

Let $(G=\PU(1,n),X,\Gamma)$ be a real Shimura datum and let $S_\Gamma$ be the associated Shimura variety. Any $\Z$-special subvariety $W$ of $S_\Gamma$ is $\Gamma$-special.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: Let $W \subset S_\Gamma$ be a $\Z$-special subvariety, as usual assumed to be of positive dimension. Let $w \in \widetilde{W}$ be such that $\pi(w)$ is smooth and $\MT(w)=\MT(W)$. By Corollary \ref{cor123}, we can assume that $\MT(w)$ is the Weil restriction of some $K$-subgroup of $\mathbf{G}$. To keep the notation more compact, we write $\mathbf{M}$ for the associated $K$-subgroup of $\mathbf{G}$, $M$ for its real points, and $\widehat{\mathbf{M} }$ for its Weil restriction to $\Q$. Being $W$ a $\Z$-special subvariety with generic Mumford--Tate group $\widehat{\mathbf{M} }$ means that W=\psi^{-1}(\widehat{\mathbf{M} }(\Z)\backslash D_{\widehat{M}})^0. Here $\widehat{M}$ denotes the associa

- `cor123` (proof_dependency; curated_source_check): By Corollary \ref{cor123}, we can assume that
- `finallemma` (proof_dependency; curated_source_check): we can apply Lemma \ref{finallemma} below to conclude that

Candidate references: `cor123` (candidate_proof_dependency); `eqW` (candidate_proof_dependency); `cdk` (candidate_proof_dependency); `finallemma` (candidate_proof_dependency).

### Lemma 66 — `finallemma`

Source lines 1474–1477.

Let $(G=\PU(1,n),X, \Gamma)$ be a real Shimura datum. Let $X_H$ be a Hermitian symmetric subspace of $X$, associated to a subgroup $H\subset G$, such that $\pi(X_H)$ is an algebraic subvariety of $S_\Gamma$. Then $(H,X_H, \Gamma_H)$ is a real sub-Shimura datum of $(G,X, \Gamma)$. That is $\Gamma_H$ is a lattice in $H$.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: Let $\Omega \subset \C^m$ be a bounded domain such that $\Gamma_\Omega\backslash \Omega$ is a quasi-projective variety form some discrete subgroup $\Gamma_\Omega$ of the automorphisms of $\Omega$ acting freely on $\Omega$. Recall that Griffiths \cite[Proposition 8.12]{MR310284} proved that $\Gamma_\Omega\backslash \Omega$ has finite Kobayashi-Eisenman volume \cite{MR0259165} (see also discussion after \cite[Question 8.13]{MR310284}). Applying such a result to $\Omega=X_H=\mathbb{B}^m$ (for some $m<n$) and $\Gamma_\Omega=\Gamma_H$ we see that $\Gamma_H\backslash X_H$ has finite Kobayashi-Eisenman volume. Thanks to \cite[Proposition 2.4 (page 57)]{MR0259165}, the Kobayashi-Eisenman volume form


Candidate references: `external:MR310284` (external_reference); `external:MR0259165` (external_reference).

### Defi 67 — `defispecial`

Source lines 1481–1484.

A smooth irreducible closed algebraic subvariety $W\subset S_\Gamma$ (of dimension $>0$) is special if it is $\Gamma$-special or equivalently $\Z$-special or equivalently a totally geodesic subvariety. We define the special closure of $W$ as the smallest special subvariety of $S_\Gamma$ containing $W$. We say that $W$ is Hodge generic if it is not contained in any strict special subvariety of $S_\Gamma$.

*Extractive statement; proof gist not independently reconstructed.*

### Theorem 5.4.1 — `nonarithas7`

Source lines 1488–1499.

Non-arithmetic Ax-Schanuel Conjecture

Let $(G,X,\Gamma)$ be an irreducible Shimura datum. Let $W \subset X \times S_\Gamma$ be an algebraic subvariety and $\Pi \subset X \times S_\Gamma$ be the graph of $\pi : X \to S_\Gamma$. Let $U$ be an irreducible component of $W \cap \Pi$ such that \codim U < \codim W + \codim \Pi, the codimension being in $X \times S_\Gamma$ or, equivalently, \dim W < \dim U + \dim S_\Gamma. If the projection of $U$ to $S_\Gamma$ is not zero dimensional, then it is contained in a strict special subvariety of $S_\Gamma$.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: If $\Gamma$ is arithmetic, the result is a special case of \cite[Theorem 1.1]{as}, so we may and do assume that $\Gamma$ is non-arithmetic. Thanks to Theorem \ref{strategy1} and the discussion after Corollary \ref{gammadense}, we have the following commutative diagram [scale=2] \node (A) at (-1,1) {$X$}; \node (B) at (1,1) {$D=D_{\widehat{G}}$}; \node (C) at (-1,0) {${S_\Gamma}^{\an}$}; \node (D) at (1,0) {$\widehat{\mathbf{G}}(\Z)\backslash D$}; \path[right hook->,font=\scriptsize] (A) edge node[above]{$\tilde{\psi}$} (B); \path[->,font=\scriptsize,>=angle 90] (C) edge node[above]{$\psi$} (D) (A) edge node[right]{$\pi_\Gamma$} (C) (B) edge node[right]{$\pi_\Z$} (D); . In the proof we actual

- `audit:cartesian` (proof_dependency; curated_source_check): In the proof we actually use that the above diagram is cartesian.

Candidate references: `strategy1` (candidate_proof_dependency); `gammadense` (candidate_proof_dependency); `identification` (candidate_proof_dependency); `atypeq` (candidate_proof_dependency); `zasthm` (candidate_proof_dependency); `external:as` (external_reference).

### Theorem 5.4.2 — `zasthm`

Source lines 1503–1510.

Let $\widehat{W} \subset {D_{\widehat{G}}} \times S_\Gamma$ be an algebraic subvariety. Let $\widehat{U}$ be an irreducible component of $\left(\widehat{W} \cap D_{\widehat{G}} \times_{ \widehat{\mathbf{G}}(\Z)\backslash D} S_\Gamma \right)\subset D_{\widehat{G}} \times S_\Gamma $ such that \codim \widehat{U} < \codim \widehat{W} + \codim D_{\widehat{G}} \times_{ \widehat{\mathbf{G}}(\Z)\backslash D} S_\Gamma, the codimension being in $ D_{\widehat{G}} \times S_\Gamma $. Then the projection of $\widehat{U}$ to $S_\Gamma$ is contained in a stric

*Extractive statement; proof gist not independently reconstructed.*

- `mainthm` (proof_dependency; curated_source_check): The next theorem follows from the work of Bakker and Tsimerman \cite[Theorem 1.1]{MR3958791}, Theorem \ref{mainthm} and the fact that $G$ has rank one.
- `external:MR3958791` (proof_dependency; curated_source_check): The next theorem follows from the work of Bakker and Tsimerman \cite[Theorem 1.1]{MR3958791}

### Cor 70 — `gammaalw`

Source lines 1568–1571.

Non-arithmetic Ax-Lindemann-Weierstrass

Let $(G=\PU(1,n),X,\Gamma)$ be a real Shimura datum and $Y\subset X$ be an algebraic subvariety. Then any irreducible component of the Zariski closure of $\pi(Y)$ in $S_\Gamma$ is special.

*Extractive statement; proof gist not independently reconstructed.*

### Cor 71 — `maincor`

Source lines 1572–1581.

Let $W\subset S_\Gamma$ be an irreducible algebraic subvariety (of dimension $>0$). The following are equivalent: \item $W$ is totally geodesic; \item $W$ is bi-algebraic, i.e. some (equivalently any) analytic component of the preimage of $W$ along $\pi : X \to S_\Gamma$ is algebraic; \item $W$ is a special subvariety (in the sense of Definition \ref{defispecial}); \item $W=\psi^{-1}(\pi_{\Z} (Y))^0$ for some algebraic subvariety $Y$ of $D_{\widehat{G}}$.

*Extractive statement; proof gist not independently reconstructed.*


Candidate references: `defispecial` (internal_reference).

### Cor 72 — `cor:line:1586`

Source lines 1586–1593.

Let $(G=\PU(1,n),X,\Gamma\subset \mathbf{G}(\Oo_K))$ be a real Shimura datum. Let $W$ be a Hodge generic subvariety of $S_\Gamma$ (in the sense of Definition \ref{defispecial}). If $\Gamma$ is non-arithmetic, the image of \pi_1(W)\to \Gamma gives rise to a thin subgroup of $\widehat{\mathbf{G}}(\Z)$.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: As in Remark \ref{infiniteindex}, since $\Gamma$ is non-arithmetic, $\Gamma$ has infinite index in $\mathbf{G}(\Oo_K)$ and so the same holds for its image in $\widehat{\mathbf{G}}(\Z)$. Because of the Hodge genericity assumption, the André--Deligne monodromy theorem (Theorem \ref{deligneandre}) implies that the image of $\pi_1(W)\to \Gamma \to \widehat{\mathbf{G}}(\Z)$ is Zariski dense in $\widehat{\mathbf{G}}$ and therefore a thin subgroup of $\widehat{\mathbf{G}}(\Z)$.


Candidate references: `defispecial` (internal_reference); `infiniteindex` (candidate_proof_dependency); `deligneandre` (candidate_proof_dependency).

### Cor 73 — `cor:line:1598`

Source lines 1598–1601.

Let $(F,X_F)$ be a real sub-Shimura couple of $(G=\PU(1,n),X)$ (in the sense of Section \ref{embeddings}). If $\Gamma_F$ is Zariski dense in $F$, then $\Gamma_{F}$ is a lattice in $F$.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: The fact that $(F,X_F)$ is a real sub-Shimura couple of $(G=\PU(1,n),X)$ implies that there exists some $x\in X$ such that the map x: \DT \to G factorises trough $F$. Since $\Gamma_F \subset F $ is Zariski dense in $F$, it determines a the $K$-form of $F$, which we denote by $\mathbf{F}$. Let $\widehat{\mathbf{F}}$ be the Weil restriction from $K$ to $\Q$ of $\mathbf{F}$. Consider the sub-pair (\widehat{\mathbf{F}}, D_{\widehat{F}}=\widehat{\mathbf{F}}(\R). \tilde{\psi}(x)) \subset (\widehat{\mathbf{G}}, D=D_{\widehat{G}}). Because \eqref{eq1234} factorises trough $F$, $(\widehat{\mathbf{F}}, D_{\widehat{F}})$ is a sub-Mumford-Tate domain of $(\widehat{\mathbf{G}}, D=D_{\widehat{G}})$. As us


Candidate references: `embeddings` (internal_reference); `eq1234` (candidate_proof_dependency); `strategy1` (candidate_proof_dependency); `gammadense` (candidate_proof_dependency); `cdk` (candidate_proof_dependency); `finallemma` (candidate_proof_dependency).

### Theorem 6.0.1 — `thm123`

Source lines 1646–1649.

If $S_\Gamma$ is non-arithmetic, then it contains only finitely many special subvarieties.

*Extractive statement; proof gist not independently reconstructed.*

### Thm 75 — `thm1234`

Source lines 1651–1658.

Let $(Y_i)_{i\in \N}$ be a family of special subvarieties of $S_\Gamma$. Then every irreducible component of the Zariski closure of the union \bigcup _{i \in \N} Y_i is either one of the $Y_i$s, or it is a special and arithmetic subvariety of $S_\Gamma$.

*Extractive statement; proof gist not independently reconstructed.*

### Lemma 76 — `lemma:line:1681`

Source lines 1681–1684.

Let $(G=\PU(1,n),X,\Gamma)$ be a real Shimura datum. There exists a compact subset $C$ of $S_\Gamma$ such that every special subvariety $S' \subset S_\Gamma$ has non-empty intersection with $C$.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: If $\Gamma$ is arithmetic this appears as \cite[Lemma 4.5]{MR2180407}, for the general case see \cite[Proposition 3.2]{MR3936536}. Here is important that $G$ has rank one, and therefore that there is no difference between special (or totally geodesic) subvarieties and what is called in \cite{MR2180407} strongly special subvarieties (see Section 4.1 in loc. cit.).


Candidate references: `external:MR2180407` (external_reference); `external:MR3936536` (external_reference).

### Lemma 77 — `623`

Source lines 1692–1699.

The set \Pi_0(\mathbf{H}):=\{(x,\hat{g}) \in \mathcal{C} \times \widehat{G}: \im (\tilde{\psi}(x): \DT \to \widehat{G})\subset \hat{g} \widehat{H} \hat{g}^{-1} \} is definable in an o-minimal structure.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: Thanks to the description of Siegel sets of Section \ref{fundamentalset}, we can assume that $\mathcal{C}$ is the closure of an open semialgebraic subset of $\F$. Note that the restriction of $\tilde{\psi}: X \to D_{\widehat{G}}$ to $\mathcal{C}$ is holomorphic and therefore definable in $\R_{\operatorname{an}}$. Eventually the condition on the image of $\tilde{\psi}(x)$ is a definable condition.


Candidate references: `fundamentalset` (candidate_proof_dependency).

### Rmk 78 — `rmk:line:1711`

Source lines 1711–1714.

The reason why we work with the compact $\mathcal{C}$, rather than with the fundamental set $\F$, is to avoid discussing the definability of (opportune restrictions of) $\tilde{\psi}$. However it should be possible to apply the arguments of \cite[Theorem 1.5]{bakker2018tame} to prove that $\tilde{\psi}_{| \F}$ is definable. It would be a consequence of the fact that $\tilde{\psi}$ restricted to ${D^{(N)}}$, where $D^{(N)}\subset X$ was defined in \ref{dn}, is definable, at least for $N$ large enough, and that the complement of a compact subset 

*Extractive statement; proof gist not independently reconstructed.*


Candidate references: `dn` (internal_reference); `external:bakker2018tame` (external_reference).

### Proposition 6.1.4 — `mainpropwithas`

Source lines 1737–1740.

Let $m= \dim W$ be the maximum of the dimensions of the strict special subvarieties of $S_\Gamma$. The set $\Sigma^m=\Sigma(\mathbf{H})^m$ is finite.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: It is enough to prove that $\Sigma^m$ is definable and countable. First observe that, since the local dimension of a definable set at a point is a definable function, $\Pi^m_1(\mathbf{H})$ is a definable set. Consider the map \Pi^m_1(\mathbf{H}) \to \widehat{G}/N_{\widehat{G}}(\widehat{H}), \ \ \ (x,\widehat{g}) \mapsto \widehat{g} \cdot N_{\widehat{G}}(\widehat{H}), where $N_{\widehat{G}}(\widehat{H})$ denotes the normaliser of $\widehat{H}$ in $\widehat{G}$. Such a map is definable, and therefore its image, which is in bijection with $\Sigma^m$, is definable. To prove that $\Sigma^m$ is countable it is enough to show that each $\hat{g} \widehat{H}\hat{g}^{-1}$ is defined over the rational 

- `zasthm` (proof_dependency; curated_source_check): Assuming the claim, the $\Z$-Ax-Schanuel, which was introduced before as Theorem \ref{zasthm}

Candidate references: `zasthm` (candidate_proof_dependency); `equ` (candidate_proof_dependency); `arithcrit` (candidate_proof_dependency).

### Rmk 80 — `klingrmk`

Source lines 1814–1817.

Recently Klingler \cite{klingler2017hodge} has proposed a generalisation of the Zilber--Pink Conjecture for arbitrary irreducible smooth quasi-projective complex varieties supporting a $\Z$-VHS. It is interesting to notice that his conjecture for the pair $(S_{\Gamma}, \widehat{\mathbb{V}})$ implies Theorem \ref{thmbader}. Given an irreducible subvariety $W\subset S_\Gamma$, Klingler introduced a notion of Hodge codimension\footnote{Namely the rank of the horizontal part of the Lie algebra of the generic Mumford--Tate group of $W$ minus the dim

*Extractive statement; proof gist not independently reconstructed.*


Candidate references: `thmbader` (internal_reference); `equ` (internal_reference); `mainpropsec` (internal_reference); `external:klingler2017hodge` (external_reference).

### Defi 81 — `defi:line:1823`

Source lines 1823–1826.

A $\Gamma$-special correspondence for $S_\Gamma$ is a $\Gamma$-special subvariety of $S_\Gamma \times S_{\Gamma}$ whose projections $\pi_1, \pi_2 : S_\Gamma \times S_\Gamma\to S_\Gamma$ are finite and surjective.

*Extractive statement; proof gist not independently reconstructed.*

### Theorem 6.2.2 — `marguliscomm`

Source lines 1827–1830.

Let $(G=\PU(1,n),X,\Gamma)$ be a real Shimura datum. The lattice $\Gamma$ is arithmetic if and only if $S_\Gamma$ admits infinitely many $\Gamma$-special correspondences.

*Extractive statement; proof gist not independently reconstructed.*

- `audit:countability` (proof_dependency; curated_source_check): We proved that $\Sigma$ is definable and countable, therefore finite.

### Rmk 83 — `rmk:line:1830`

Source lines 1830–1833.

If $n=1$, the axiom RSD0 of real Shimura datum implies that $\Gamma$ is arithmetic and therefore that $S_\Gamma$ admits infinitely many $\Gamma$-special correspondences. The proof of Theorem \ref{marguliscomm} carries on assuming only that $\Gamma$ is (cohomologically\footnote{Over a one dimensional base, all rigid systems are known to be cohomologically rigid \cite[Theorem 1.1.2, Corollary 1.2.4]{MR1366651}.}) rigid. However we are not able to recover Theorem \ref{commcriterion} for all lattices in $\PU(1,1)$.

*Extractive statement; proof gist not independently reconstructed.*


Candidate references: `marguliscomm` (internal_reference); `commcriterion` (internal_reference); `external:MR1366651` (external_reference).

### Defi 84 — `defi:line:1859`

Source lines 1859–1862.

A $\Z$-special correspondence for $S_\Gamma$ is a $\Z$-special subvariety of $S_\Gamma \times S_{\Gamma}$ whose projections $\pi_i : S_\Gamma \times S_\Gamma\to S_\Gamma$ are finite and surjective.

*Extractive statement; proof gist not independently reconstructed.*

### Prop 85 — `prop:line:1865`

Source lines 1865–1868.

The $\Gamma$-special correspondences are precisely the $\Z$-special ones.

*Extractive statement; proof gist not independently reconstructed.*

### Defi 86 — `gammasppoint`

Source lines 1920–1927.

A point $x \in X$ is pre-$\Gamma$-special if the smallest $K$-subgroup of $\mathbf{G}$ whose extension to the real numbers contains the image of x: \DT \to \mathbf{G}_{\R} is commutative. A point $s\in S_\Gamma$ is $\Gamma$-special if it is the image along $\pi: X \to S_\Gamma$ of a pre-$\Gamma$-special point $x\in X$.

*Extractive statement; proof gist not independently reconstructed.*

### Prop 87 — `prop:line:1928`

Source lines 1928–1931.

Every $\Gamma$-special subvariety contains an analytically dense set of $\Gamma$-special points. Pre-$\Gamma$-special point are defined over $\Qbar$, with respect to the $\Qbar$-structure given by looking at $X$ inside its associated flag variety $X^\vee$.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: Let $S_{\Gamma_H}$ be a $\Gamma$-special subvariety associated to $(H,X_H, \Gamma_H)$. First notice that if one $\Gamma$-special point exists in $S_{\Gamma_H}$, then there is a dense set of $\Gamma$-special points. Indeed each point in $\pi(\mathbf{H}(K).x) \subset S_{\Gamma_H}$ is again $\Gamma$-special and the set $\mathbf{H}(K).x$ is dense by the usual approximation property. For the existence of a $\Gamma$-special point we argue as in \cite[Section 5]{delignetravaux}. Let $x\in X_H$ and $T\subset G$ be a maximal torus containing the image of x: \DT \to H. Write $T$ as the centraliser of some regular element $\lambda$ of the Lie algebra of $T$. Choose $\lambda' \in \mathbf{H}(K)$ sufficie


Candidate references: `external:delignetravaux` (external_reference); `external:MR2825237` (external_reference).

### Defi 89 — `defi:line:1956`

Source lines 1956–1959.

A point $s\in S_\Gamma$ is $\Z$-special if it is a zero dimensional $\Z$-special subvariety.

*Extractive statement; proof gist not independently reconstructed.*

### Prop 90 — `propspepoints`

Source lines 1964–1967.

Let $s=\pi(x)\in S_\Gamma$ be a $\Z$-special point, then $s$ is $\Gamma$-special (in the sense of Definition \ref{gammasppoint}).

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: Let $\mathbf{R}$ be a $\Q$-subgroup of $\widehat{\mathbf{G}}$ inducing a Mumford--Tate sub-domain $D_{R}$ of $D_{\widehat{G}}$. If $\mathbf{R}(\Z)\backslash D_R$ intersects $\psi ({S_\Gamma}^{\an})$ in a finite number of points, then $R \cap G_{\sigma_1}$ is a compact subgroup of $\widehat{\mathbf{G}}(\R)$. Let $x\in \tilde{\psi}^{-1}(D_{R})\subset X$. It corresponds to a map x: \DT \to G, and, by construction $x(\DT)$ is contained in $R \cap G_{\sigma_1}$. Since $G$ is of adjoint type, $x (\DT)$ is contained in the centre of $R \cap G_{\sigma_1}$. However $R \cap G_{\sigma_1}$ has a natural structure of $K$-subgroup of $\mathbf{G}$, which we denote by $\mathbf{R}_{\sigma}$. It follows that 

- `audit:weak-definition` (proof_dependency; curated_source_check): $\mathbf{M}$ a normal algebraic $\Q$-subgroup of $\MT(x)$

Candidate references: `gammasppoint` (internal_reference).

### Prop 91 — `prop:line:1978`

Source lines 1978–1981.

If $\Gamma$ is non-arithmetic, $\Z$-special points of $S_\Gamma$ are atypical intersections in $\widehat{\mathbf{G}} (\Z) \backslash D_{\widehat{G}}$ between $\psi ({S_\Gamma}^{\an})$ and Mumford--Tate sub-domains of $D_{\widehat{G}}$.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: Let $\mathbf{R}\subset \widehat{\mathbf{G}}$ be as in the proof of Proposition \ref{propspepoints}. The intersection between $\mathbf{R}(\Z)\backslash D_R$ and $\psi ({S_\Gamma}^{\an})$ is typical when \codim_{\widehat{\mathbf{G}} (\Z) \backslash D_{\widehat{G}}} \{s\} = \codim _{\widehat{\mathbf{G}} (\Z) \backslash D_{\widehat{G}}}S_{\Gamma} + \codim_{\widehat{\mathbf{G}} (\Z) \backslash D_{\widehat{G}}} \mathbf{R}(\Z)\backslash D_R, where $s$ is one of the finite points lying in $\mathbf{R}(\Z)\backslash D_R \cap \psi ({S_\Gamma}^{\an})$. Let $r=[K:\Q]$. Equivalently \sum_{i=1}^r d_{i}=\sum_{i=2}^r d_{i}+ \left( d_1 + \sum_{i=2}^r d_{i} - d_{R_i}\right), where $d_{i}$ and $d_{R_i}$ were de


Candidate references: `propspepoints` (candidate_proof_dependency); `mainpropwithas` (candidate_proof_dependency).

### Prop 93 — `prop:line:2002`

Source lines 2002–2005.

Let $S_{\Gamma_1}$, $S_{\Gamma_2}$ be $\Gamma$-special subvarieties of positive dimension in $S_\Gamma$ intersecting in a finite number of points. If $s \in S_{\Gamma_1}\cap S_{\Gamma_2}$, then $s$ is $\Z$-special.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: Since $S_{\Gamma_1}$ and $S_{\Gamma_2}$ are also $\Z$-special subvariety, every point in the intersection can be written as the the preimage along the period map of the intersection of two Mumford--Tate sub-domains. The point $s$ is therefore $\Z$-special.

### Defi 94 — `defcm`

Source lines 2013–2016.

A point $s\in S_\Gamma$ is called a CM-point if the Mumford--Tate group of $\widehat{\mathbb{V}}$ at $s$ is commutative.

*Extractive statement; proof gist not independently reconstructed.*

### Thm 96 — `thm:line:2033`

Source lines 2033–2039.

Shimura, Calabi-Vesentini

Let $(G=\PU(1,n),X,\Gamma)$ be a real Shimura datum such that $S_\Gamma$ is projective. Let $\Theta$ be the sheaf of germs of holomorphic sections of the tangent bundle of $S_\Gamma$. If the dimension of $S_\Gamma$ is greater or equal to two, then, for $i=0,1$ H^i(S_\Gamma, \Theta)=0. It follows that $S_\Gamma$ has a unique $\Qbar$-structure.

*Extractive statement; proof gist not independently reconstructed.*

### Thm 97 — `actionofc`

Source lines 2044–2055.

Let $(G,X,\Gamma)$ be a real Shimura datum and $\sigma \in \Aut(\C)$. There exists a real Shimura datum $(\leftidx{^\sigma}{G},\leftidx{^\sigma}{X},\leftidx{^\sigma}{\Gamma})$ such that \leftidx{^\sigma}{S}{_\Gamma} \cong S_{^\sigma \Gamma}. Moreover: \item $G$ has rank one if and only if $\leftidx{^\sigma}{G}$ has rank one; \item $\Gamma$ is arithmetic if and only if $\leftidx{^\sigma}{\Gamma}$ is arithmetic.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: In the arithmetic case, the result follows from the work of Borovoi and Kazdhan \cite{MR693357}. To prove the result, we may and do replace $\Gamma$ by a finite index subgroup. In the non-arithmetic case, thanks to the work of Margulis (Theorem \ref{margulisrank}), the first part of the above statement reduces to the case of ball quotients (of dimension strictly bigger than one). The result follows from fact that the Chern numbers of $S_\Gamma$ satisfy the equality from Yau's solution of the Calabi conjecture (in the cocompact case, and by Tsuji \cite{zbMATH04097530} in the general case), which characterise ball quotients by their Chern numbers. See for example \cite[Proof of the reduction, 


Candidate references: `margulisrank` (candidate_proof_dependency); `compac` (candidate_proof_dependency); `eqqqqq123` (candidate_proof_dependency); `sectionmarguliscomm` (candidate_proof_dependency); `commcriterion` (candidate_proof_dependency); `external:MR693357` (external_reference); `external:zbMATH04097530` (external_reference); `external:MR1211885` (external_reference); `external:MR3424477` (external_reference).

### Rmk 98 — `propmod`

Source lines 2079–2082.

To complete our discussion of the action of $\Aut(\C)$, we should explain what happens to special subvarieties of positive dimension of $S_\Gamma$ (i.e. in the sense of Definition \ref{defispecial}). Indeed it is a well known fact that, in the theory of arithmetic Shimura varieties, $\Aut(\C)$ maps special subvarieties of $S_\Gamma$ to special subvarieties of $S_{^\sigma \Gamma}$. Thanks to the description of special subvarieties of $S_\Gamma$ as the Hodge locus of a $\Z$-VHS, as we established in Corollary \ref{maincor}, we can see that $\sigm

*Extractive statement; proof gist not independently reconstructed.*


Candidate references: `defispecial` (internal_reference); `maincor` (internal_reference); `external:2020arXiv201003359K` (external_reference).

### Thm 99 — `thm:line:2086`

Source lines 2086–2089.

Let $(G,X,\Gamma)$ be a real Shimura datum. Then $S_\Gamma$ admits a model over a number field.

*Extractive statement; proof gist not independently reconstructed.*

Proof opening: If $\Gamma$ is arithmetic, the result follows for example from Faltings \cite{MR791585}. Therefore we may and do assume that $G=\PU(1,n)$, for some $n>1$. Let ${S_\Gamma}^{BB}$ be Baily-Borel compactification of $S_\Gamma$, as discussed in Section \ref{compac}. From \cite[Criterion 1 (page 3)]{MR2253591}, the following are known to be equivalent: \item[(a)] ${S_\Gamma}^{BB}$ can be defined over $\Qbar$; \item[(b)] The set $\{\leftidx{^\sigma}{{S_\Gamma}}^{BB}: \sigma \in \Aut(\mathbb{C} / \Qbar)\}$ contains only finitely many isomorphism classes of complex projective varieties; \item[(c)] The set $\{\leftidx{^\sigma}{{S_\Gamma}}^{BB}: \sigma \in \Aut(\mathbb{C} / \Qbar)\}$ contains only coun


Candidate references: `compac` (candidate_proof_dependency); `actionofc` (candidate_proof_dependency); `localrig` (candidate_proof_dependency); `external:MR791585` (external_reference); `external:MR2253591` (external_reference).

### Auxiliary self-duality assertion — `audit:self-dual`

Source lines 1038–1038.

Auxiliary self-duality assertion

Source assertion indexed separately from its containing theorem.

### Auxiliary finite-centralizer assertion — `audit:centralizer`

Source lines 1390–1390.

Auxiliary finite-centralizer assertion

Source assertion indexed separately from its containing theorem.

### Image-field contradiction — `audit:embedding-images`

Source lines 1284–1284.

Image-field contradiction

Source assertion indexed separately from its containing theorem.

### Countability of subgroup Sigma — `audit:countability`

Source lines 1910–1910.

Countability of subgroup Sigma

Source assertion indexed separately from its containing theorem.

### Global Cartesian assertion — `audit:cartesian`

Source lines 1527–1527.

Global Cartesian assertion

Source assertion indexed separately from its containing theorem.

### Weak-domain normal-subgroup definition — `audit:weak-definition`

Source lines 1136–1136.

Weak-domain normal-subgroup definition

Source assertion indexed separately from its containing theorem.

### Downward-induction exclusion — `audit:induction`

Source lines 1809–1809.

Downward-induction exclusion

Source assertion indexed separately from its containing theorem.

## External inputs and references

### [1] MR1154159 — `external:MR1154159`

Source lines 2117–2123.

Y.~Andr\'{e}. \newblock Mumford-{T}ate groups of mixed {H}odge structures and the theorem of the fixed part. \newblock {\em Compositio Math.}, 82(1):1--24, 1992.


Candidate references: `comparisonsection` (external_reference).

### [2] MR2590897 — `external:MR2590897`

Source lines 2123–2130.

A.~Ash, D.~Mumford, M.~Rapoport, and Y.-S. Tai. \newblock {\em Smooth compactifications of locally symmetric varieties}. \newblock Cambridge Mathematical Library. Cambridge University Press, Cambridge, second edition, 2010. \newblock With the collaboration of Peter Scholze.


Candidate references: `sectiongeneral` (external_reference).

### [3] bader2020arithmeticity — `external:bader2020arithmeticity`

Source lines 2130–2136.

U.~{Bader}, D.~{Fisher}, N.~{Miller}, and M.~{Stover}. \newblock {Arithmeticity, superrigidity and totally geodesic submanifolds of complex hyperbolic manifolds}. \newblock {\em arXiv e-prints}, page arXiv:2006.03008, June 2020.

- `thmbader` (background; curated_source_check): The proof has then appeared in \cite{bader2020arithmeticity}.

Candidate references: `section:1` (external_reference).

### [4] 2019arXiv190308467B — `external:2019arXiv190308467B`

Source lines 2136–2141.

U.~{Bader}, D.~{Fisher}, N.~{Miller}, and M.~{Stover}. \newblock {Arithmeticity, Superrigidity, and Totally Geodesic Submanifolds}. \newblock {\em Ann. of Math.}, 193(3):837--861, 2021.


Candidate references: `section:1` (external_reference).

### [5] MR0216035 — `external:MR0216035`

Source lines 2141–2147.

W.~L. Baily, Jr. and A.~Borel. \newblock Compactification of arithmetic quotients of bounded symmetric domains. \newblock {\em Ann. of Math. (2)}, 84:442--528, 1966.


Candidate references: `section:1` (external_reference); `algebraic` (external_reference).

### [6] bakker2018tame — `external:bakker2018tame`

Source lines 2147–2153.

B.~{Bakker}, B.~{Klingler}, and J.~{Tsimerman}. \newblock {Tame topology of arithmetic quotients and algebraicity of Hodge loci}. \newblock {\em {J. Am. Math. Soc.}}, 33(4):917--939, 2020.


Candidate references: `section:1` (external_reference); `zhodgesection` (external_reference); `rmk:line:1711` (external_reference).

### [7] MR3958791 — `external:MR3958791`

Source lines 2153–2158.

B.~Bakker and J.~Tsimerman. \newblock The {A}x--{S}chanuel conjecture for variations of {H}odge structures. \newblock {\em Invent. Math.}, 217(1):77--94, 2019.

- `zasthm` (proof_dependency; curated_source_check): The next theorem follows from the work of Bakker and Tsimerman \cite[Theorem 1.1]{MR3958791}

Candidate references: `section:1` (external_reference); `zhodgesection` (external_reference); `comparisonsection` (external_reference); `sectionfinit` (external_reference).

### [8] MR586867 — `external:MR586867`

Source lines 2158–2163.

H.~Bass. \newblock Groups of integral representation type. \newblock {\em Pacific J. Math.}, 86(1):15--51, 1980.


Candidate references: `integrality` (external_reference).

### [9] MR0244260 — `external:MR0244260`

Source lines 2163–2170.

A.~Borel. \newblock {\em Introduction aux groupes arithm\'{e}tiques}. \newblock Publications de l'Institut de Math\'{e}matique de l'Universit\'{e} de Strasbourg, XV. Actualit\'{e}s Scientifiques et Industrielles, No. 1341. Hermann, Paris, 1969.


Candidate references: `sectiongeneral` (external_reference).

### [10] MR0111058 — `external:MR0111058`

Source lines 2170–2175.

E.~Calabi and E.~Vesentini. \newblock On compact, locally symmetric {K}\"{a}hler manifolds. \newblock {\em Ann. of Math. (2)}, 71:472--507, 1960.


Candidate references: `section:1` (external_reference); `pointsection` (external_reference).

### [11] MR1273413 — `external:MR1273413`

Source lines 2175–2180.

E.~Cattani, P.~Deligne, and A.~Kaplan. \newblock On the locus of {H}odge classes. \newblock {\em J. Amer. Math. Soc.}, 8(2):483--506, 1995.


Candidate references: `section:1` (external_reference); `cdk` (external_reference).

### [12] chan2018asymptotic — `external:chan2018asymptotic`

Source lines 2180–2187.

S.~T. Chan and N.~Mok. \newblock Asymptotic total geodesy of local holomorphic curves exiting a bounded symmetric domain and applications to a uniformization problem for algebraic subsets. \newblock {\em Accepted for publication in J. Differential Geom.}, 2018.


Candidate references: `section:1` (external_reference).

### [13] MR2180407 — `external:MR2180407`

Source lines 2187–2192.

L.~Clozel and E.~Ullmo. \newblock \'{E}quidistribution de sous-vari\'{e}t\'{e}s sp\'{e}ciales. \newblock {\em Ann. of Math. (2)}, 161(3):1571--1588, 2005.


Candidate references: `rmkfactorisa` (external_reference); `lemma:line:1681` (external_reference).

### [14] modularemb1 — `external:modularemb1`

Source lines 2192–2197.

P.~Cohen and J.~Wolfart. \newblock Modular embeddings for some nonarithmetic {F}uchsian groups. \newblock {\em Acta Arith.}, 56(2):93--110, 1990.


Candidate references: `section:1` (external_reference).

### [15] modularemb2 — `external:modularemb2`

Source lines 2197–2204.

P.~Cohen and J.~Wolfart. \newblock Monodromie des fonctions d'{A}ppell, vari\'{e}t\'{e}s ab\'{e}liennes et plongement modulaire. \newblock {\em Ast\'{e}risque}, (198-200):6, 97--105 (1992), 1991. \newblock Journ\'{e}es Arithm\'{e}tiques, 1989 (Luminy, 1989).


Candidate references: `section:1` (external_reference).

### [16] MR1147961 — `external:MR1147961`

Source lines 2204–2209.

K.~Corlette. \newblock Archimedean superrigidity and hyperbolic geometry. \newblock {\em Ann. of Math. (2)}, 135(1):165--182, 1992.


Candidate references: `section:1` (external_reference).

### [17] MR2457528 — `external:MR2457528`

Source lines 2209–2215.

K.~Corlette and C.~Simpson. \newblock On the classification of rank-two representations of quasiprojective fundamental groups. \newblock {\em Compos. Math.}, 144(5):1271--1331, 2008.


Candidate references: `integrality` (external_reference); `zhodgesection` (external_reference); `deformation` (external_reference); `totreal` (external_reference).

### [18] MR3867286 — `external:MR3867286`

Source lines 2215–2220.

C.~Daw and J.~Ren. \newblock Applications of the hyperbolic {A}x-{S}chanuel conjecture. \newblock {\em Compos. Math.}, 154(9):1843--1888, 2018.


Candidate references: `sectionfinit` (external_reference).

### [19] MR0498551 — `external:MR0498551`

Source lines 2220–2225.

P.~Deligne. \newblock Th\'{e}orie de {H}odge. {II}. \newblock {\em Inst. Hautes \'{E}tudes Sci. Publ. Math.}, (40):5--57, 1971.


Candidate references: `zhodgesection` (external_reference); `comparisonsection` (external_reference).

### [20] delignetravaux — `external:delignetravaux`

Source lines 2225–2230.

P.~Deligne. \newblock Travaux de {S}himura. \newblock pages 123--165. Lecture Notes in Math., Vol. 244, 1971.


Candidate references: `sectiongeneral` (external_reference); `vhsonsgamma` (external_reference); `zhodgesection` (external_reference); `prop:line:1928` (external_reference).

### [21] deligneshimura — `external:deligneshimura`

Source lines 2230–2239.

P.~Deligne. \newblock Vari\'et\'es de {S}himura: interpr\'etation modulaire, et techniques de construction de mod\`eles canoniques. \newblock In {\em Automorphic forms, representations and {$L$}-functions ({P}roc. {S}ympos. {P}ure {M}ath., {O}regon {S}tate {U}niv., {C}orvallis, {O}re., 1977), {P}art 2}, Proc. Sympos. Pure Math., XXXIII, pages 247--289. Amer. Math. Soc., Providence, R.I., 1979.


Candidate references: `sectiongeneral` (external_reference).

### [22] MR849651 — `external:MR849651`

Source lines 2239–2245.

P.~Deligne and G.~D. Mostow. \newblock Monodromy of hypergeometric functions and nonlattice integral monodromy. \newblock {\em Inst. Hautes \'{E}tudes Sci. Publ. Math.}, (63):5--89, 1986.


Candidate references: `section:1` (external_reference); `section:2` (external_reference).

### [23] delp2015problems — `external:delp2015problems`

Source lines 2245–2249.

K.~{Delp}, D.~{Hoffoss}, and J.~{Fox Manning}. \newblock {Problems In Groups, Geometry, and Three-Manifolds}, Dec. 2015.


Candidate references: `section:1` (external_reference).

### [24] deng2020big — `external:deng2020big`

Source lines 2249–2254.

Y.~{Deng}. \newblock {Big Picard theorem and algebraic hyperbolicity for varieties admitting a variation of Hodge structures}, Jan. 2020.


Candidate references: `algebraic` (external_reference).

### [25] zbMATH07195380 — `external:zbMATH07195380`

Source lines 2254–2259.

M.~{Deraux}. \newblock {A new nonarithmetic lattice in \(\mathrm{PU}(3,1)\)}. \newblock {\em {Algebr. Geom. Topol.}}, 20(2):925--963, 2020.


Candidate references: `section:2` (external_reference).

### [26] MR3461365 — `external:MR3461365`

Source lines 2259–2264.

M.~Deraux, J.~R. Parker, and J.~Paupert. \newblock New non-arithmetic complex hyperbolic lattices. \newblock {\em Invent. Math.}, 203(3):681--771, 2016.


Candidate references: `section:2` (external_reference); `sectiongeneral` (external_reference).

### [27] zbMATH07390741 — `external:zbMATH07390741`

Source lines 2264–2269.

M.~{Deraux}, J.~R. {Parker}, and J.~{Paupert}. \newblock {New nonarithmetic complex hyperbolic lattices. II}. \newblock {\em {Mich. Math. J.}}, 70(1):133--205, 2021.


Candidate references: `section:2` (external_reference).

### [28] MR3424477 — `external:MR3424477`

Source lines 2269–2274.

M.~Dimitrov and D.~Ramakrishnan. \newblock Arithmetic quotients of the complex ball and a conjecture of {L}ang. \newblock {\em Doc. Math.}, 20:1185--1205, 2015.


Candidate references: `actionofc` (external_reference).

### [29] MR1973057 — `external:MR1973057`

Source lines 2274–2279.

B.~Edixhoven and A.~Yafaev. \newblock Subvarieties of {S}himura varieties. \newblock {\em Ann. of Math. (2)}, 157(2):621--645, 2003.


Candidate references: `section:1` (external_reference).

### [30] MR0259165 — `external:MR0259165`

Source lines 2279–2286.

D.~A. Eisenman. \newblock {\em Intrinsic measures on complex manifolds and holomorphic mappings}. \newblock Memoirs of the American Mathematical Society, No. 96. American Mathematical Society, Providence, R.I., 1970.


Candidate references: `finallemma` (external_reference).

### [31] MR3874695 — `external:MR3874695`

Source lines 2286–2291.

H.~Esnault and M.~Groechenig. \newblock Cohomologically rigid local systems and integrality. \newblock {\em Selecta Math. (N.S.)}, 24(5):4279--4292, 2018.


Candidate references: `section:1` (external_reference); `zhodgesection` (external_reference); `esgro` (external_reference).

### [32] MR791585 — `external:MR791585`

Source lines 2291–2298.

G.~Faltings. \newblock Arithmetic varieties and rigidity. \newblock In {\em Seminar on number theory, {P}aris 1982--83 ({P}aris, 1982/1983)}, volume~51 of {\em Progr. Math.}, pages 63--77. Birkh\"{a}user Boston, Boston, MA, 1984.


Candidate references: `pointsection` (external_reference); `thm:line:2086` (external_reference).

### [33] MR0267041 — `external:MR0267041`

Source lines 2298–2304.

H.~Garland and M.~S. Raghunathan. \newblock Fundamental domains for lattices in ({$\mathbb{R}$}-)rank {$1$} semisimple {L}ie groups. \newblock {\em Ann. of Math. (2)}, 92:279--326, 1970.


Candidate references: `section:1` (external_reference); `coeffinnumbers` (external_reference); `sectiongeneral` (external_reference); `fundamentalset` (external_reference); `garalndragvanishing` (external_reference).

### [34] zbMATH01270600 — `external:zbMATH01270600`

Source lines 2304–2309.

W.~M. {Goldman}. \newblock {\em {Complex hyperbolic geometry}}. \newblock Oxford: Clarendon Press, 1999.


Candidate references: `sectiongeneral` (external_reference).

### [35] MR2253591 — `external:MR2253591`

Source lines 2309–2314.

G.~Gonz\'{a}lez-Diez. \newblock Variations on {B}elyi's theorem. \newblock {\em Q. J. Math.}, 57(3):339--354, 2006.


Candidate references: `thm:line:2086` (external_reference).

### [36] MR2918237 — `external:MR2918237`

Source lines 2314–2321.

M.~Green, P.~Griffiths, and M.~Kerr. \newblock {\em Mumford-{T}ate groups and domains}, volume 183 of {\em Annals of Mathematics Studies}. \newblock Princeton University Press, Princeton, NJ, 2012. \newblock Their geometry and arithmetic.


Candidate references: `section:1` (external_reference); `zhodgesection` (external_reference).

### [37] MR310284 — `external:MR310284`

Source lines 2321–2327.

P.~A. Griffiths. \newblock Complex-analytic properties of certain {Z}ariski open sets on algebraic varieties. \newblock {\em Ann. of Math. (2)}, 94:21--51, 1971.


Candidate references: `finallemma` (external_reference).

### [38] MR932135 — `external:MR932135`

Source lines 2327–2332.

M.~Gromov and I.~Piatetski-Shapiro. \newblock Nonarithmetic groups in {L}obachevsky spaces. \newblock {\em Inst. Hautes \'{E}tudes Sci. Publ. Math.}, (66):93--103, 1988.


Candidate references: `section:1` (external_reference).

### [39] MR1215595 — `external:MR1215595`

Source lines 2332–2338.

M.~Gromov and R.~Schoen. \newblock Harmonic maps into singular spaces and {$p$}-adic superrigidity for lattices in groups of rank one. \newblock {\em Inst. Hautes \'{E}tudes Sci. Publ. Math.}, (76):165--246, 1992.


Candidate references: `section:1` (external_reference).

### [40] MR3552014 — `external:MR3552014`

Source lines 2338–2343.

P.~Habegger and J.~Pila. \newblock O-minimality and certain atypical intersections. \newblock {\em Ann. Sci. \'{E}c. Norm. Sup\'{e}r. (4)}, 49(4):813--858, 2016.


Candidate references: `sectionfinit` (external_reference).

### [41] MR1374201 — `external:MR1374201`

Source lines 2343–2348.

C.~Hummel and V.~Schroeder. \newblock Cusp closing in rank one symmetric spaces. \newblock {\em Invent. Math.}, 123(2):283--307, 1996.


Candidate references: `sectiongeneral` (external_reference).

### [42] MR1366651 — `external:MR1366651`

Source lines 2348–2354.

N.~M. Katz. \newblock {\em Rigid local systems}, volume 139 of {\em Annals of Mathematics Studies}. \newblock Princeton University Press, Princeton, NJ, 1996.


Candidate references: `rmk:line:1830` (external_reference).

### [43] MR693357 — `external:MR693357`

Source lines 2354–2359.

D.~Kazhdan. \newblock On arithmetic varieties. {II}. \newblock {\em Israel J. Math.}, 44(2):139--159, 1983.


Candidate references: `actionofc` (external_reference).

### [44] klingler2017hodge — `external:klingler2017hodge`

Source lines 2359–2364.

B.~Klingler. \newblock Hodge loci and atypical intersections: conjectures, 2017. \newblock To appear in the book Motives and Complex Multiplication.


Candidate references: `section:1` (external_reference); `zhodgesection` (external_reference); `klingrmk` (external_reference); `pointsection` (external_reference).

### [45] 2020arXiv201003359K — `external:2020arXiv201003359K`

Source lines 2364–2370.

B.~{Klingler}, A.~{Otwinowska}, and D.~{Urbanik}. \newblock {On the fields of definition of Hodge loci}. \newblock {\em To appear in {A}nn. {S}ci. {\'{E}}c. {N}orm. {S}up\'{e}r.}, page arXiv:2010.03359, Oct. 2020.


Candidate references: `propmod` (external_reference).

### [46] MR3502100 — `external:MR3502100`

Source lines 2370–2375.

B.~Klingler, E.~Ullmo, and A.~Yafaev. \newblock The hyperbolic {A}x-{L}indemann-{W}eierstrass conjecture. \newblock {\em Publ. Math. Inst. Hautes \'{E}tudes Sci.}, 123:333--360, 2016.


Candidate references: `section:1` (external_reference); `sectiongeneral` (external_reference); `sectionfinit` (external_reference).

### [47] MR3821177 — `external:MR3821177`

Source lines 2375–2382.

B.~Klingler, E.~Ullmo, and A.~Yafaev. \newblock Bi-algebraic geometry and the {A}ndr\'{e}-{O}ort conjecture. \newblock In {\em Algebraic geometry: {S}alt {L}ake {C}ity 2015}, volume~97 of {\em Proc. Sympos. Pure Math.}, pages 319--359. Amer. Math. Soc., Providence, RI, 2018.


Candidate references: `section:1` (external_reference); `comparisonsection` (external_reference); `pointsection` (external_reference).

### [48] MR3936536 — `external:MR3936536`

Source lines 2382–2388.

V.~Koziarz and J.~Maubon. \newblock Finiteness of totally geodesic exceptional divisors in {H}ermitian locally symmetric spaces. \newblock {\em Bull. Soc. Math. France}, 146(4):613--631, 2018.


Candidate references: `lemma:line:1681` (external_reference).

### [49] MR3826464 — `external:MR3826464`

Source lines 2388–2393.

A.~Langer and C.~Simpson. \newblock Rank 3 rigid representations of projective fundamental groups. \newblock {\em Compos. Math.}, 154(7):1534--1570, 2018.


Candidate references: `zhodgesection` (external_reference).

### [50] margulis2019arithmeticity — `external:margulis2019arithmeticity`

Source lines 2393–2399.

G.~Margulis and A.~Mohammadi. \newblock Arithmeticity of hyperbolic 3-manifolds containing infinitely many totally geodesic surfaces. \newblock {\em Ergodic Theory and Dynamical Systems, 1-32.}


Candidate references: `section:1` (external_reference).

### [51] margulisbook — `external:margulisbook`

Source lines 2399–2406.

G.~A. Margulis. \newblock {\em Discrete subgroups of semisimple {L}ie groups}, volume~17 of {\em Ergebnisse der Mathematik und ihrer Grenzgebiete (3) [Results in Mathematics and Related Areas (3)]}. \newblock Springer-Verlag, Berlin, 1991.


Candidate references: `section:1` (external_reference); `section:2` (external_reference).

### [52] MR3247048 — `external:MR3247048`

Source lines 2406–2411.

D.~B. McReynolds and A.~W. Reid. \newblock The genus spectrum of a hyperbolic 3-manifold. \newblock {\em Math. Res. Lett.}, 21(1):169--185, 2014.


Candidate references: `section:1` (external_reference).

### [53] zbMATH05119702 — `external:zbMATH05119702`

Source lines 2411–2417.

T.~{Mochizuki}. \newblock {\em {Asymptotic behaviour of tame harmonic bundles and an application to pure twistor \(D\)-modules. I}}, volume 869. \newblock Providence, RI: American Mathematical Society (AMS), 2007.


Candidate references: `zhodgesection` (external_reference).

### [54] MR2884042 — `external:MR2884042`

Source lines 2417–2425.

N.~Mok. \newblock Projective algebraicity of minimal compactifications of complex-hyperbolic space forms of finite volume. \newblock In {\em Perspectives in analysis, geometry, and topology}, volume 296 of {\em Progr. Math.}, pages 331--354. Birkh\"{a}user/Springer, New York, 2012.


Candidate references: `algebraic` (external_reference); `sectiongeneral` (external_reference); `thm:line:685` (external_reference); `quasiunipot` (external_reference); `poincaregrowth` (external_reference).

### [55] mokalw — `external:mokalw`

Source lines 2425–2431.

N.~{Mok}. \newblock {Zariski closures of images of algebraic subsets under the uniformization map on finite-volume quotients of the complex unit ball}. \newblock {\em {Compos. Math.}}, 155(11):2129--2149, 2019.


Candidate references: `section:1` (external_reference); `comparisonsection` (external_reference).

### [56] as — `external:as`

Source lines 2431–2436.

N.~Mok, J.~Pila, and J.~Tsimerman. \newblock Ax-{S}chanuel for {S}himura varieties. \newblock {\em Ann. of Math.}, 189(3):945--978, 2019.


Candidate references: `section:1` (external_reference); `nonarithas7` (external_reference).

### [57] MR1211885 — `external:MR1211885`

Source lines 2436–2443.

N.~Mok and S.-K. Yeung. \newblock Geometric realizations of uniformization of conjugates of {H}ermitian locally symmetric manifolds. \newblock In {\em Complex analysis and geometry}, Univ. Ser. Math., pages 253--270. Plenum, New York, 1993.


Candidate references: `actionofc` (external_reference).

### [58] MR3307755 — `external:MR3307755`

Source lines 2443–2448.

D.~W. Morris. \newblock {\em Introduction to arithmetic groups}. \newblock Deductive Press, 2015.


Candidate references: `section:2` (external_reference); `prop:line:485` (external_reference).

### [59] MR586876 — `external:MR586876`

Source lines 2448–2453.

G.~D. Mostow. \newblock On a remarkable class of polyhedra in complex hyperbolic space. \newblock {\em Pacific J. Math.}, 86(1):171--276, 1980.


Candidate references: `section:1` (external_reference); `section:2` (external_reference).

### [60] MR471627 — `external:MR471627`

Source lines 2453–2458.

D.~Mumford. \newblock Hirzebruch's proportionality theorem in the noncompact case. \newblock {\em Invent. Math.}, 42:239--272, 1977.


Candidate references: `quasiunipot` (external_reference); `pointsection` (external_reference).

### [61] MR832040 — `external:MR832040`

Source lines 2458–2463.

M.~V. Nori. \newblock A nonarithmetic monodromy group. \newblock {\em C. R. Acad. Sci. Paris S\'{e}r. I Math.}, 302(2):71--72, 1986.


Candidate references: `section:1` (external_reference).

### [62] MR3725498 — `external:MR3725498`

Source lines 2463–2468.

C.~Peters. \newblock On rigidity of locally symmetric spaces. \newblock {\em M\"{u}nster J. Math.}, 10(2):277--286, 2017.


Candidate references: `pointsection` (external_reference).

### [63] MR2492989 — `external:MR2492989`

Source lines 2468–2473.

Y.~Peterzil and S.~Starchenko. \newblock Complex analytic geometry and analytic-geometric categories. \newblock {\em J. Reine Angew. Math.}, 626:39--74, 2009.


Candidate references: `ogaga` (external_reference).

### [64] MR0507234 — `external:MR0507234`

Source lines 2473–2479.

M.~S. Raghunathan. \newblock {\em Discrete subgroups of {L}ie groups}. \newblock Springer-Verlag, New York-Heidelberg, 1972. \newblock Ergebnisse der Mathematik und ihrer Grenzgebiete, Band 68.


Candidate references: `proponproducts` (external_reference).

### [65] MR3220897 — `external:MR3220897`

Source lines 2479–2486.

P.~Sarnak. \newblock Notes on thin matrix groups. \newblock In {\em Thin groups and superstrong approximation}, volume~61 of {\em Math. Sci. Res. Inst. Publ.}, pages 343--362. Cambridge Univ. Press, Cambridge, 2014.


Candidate references: `section:1` (external_reference).

### [66] MR0223368 — `external:MR0223368`

Source lines 2486–2491.

G.~Shimura. \newblock Algebraic varieties without deformation and the {C}how variety. \newblock {\em J. Math. Soc. Japan}, 20:336--341, 1968.


Candidate references: `pointsection` (external_reference).

### [67] MR1179076 — `external:MR1179076`

Source lines 2491–2496.

C.~T. Simpson. \newblock Higgs bundles and local systems. \newblock {\em Inst. Hautes \'{E}tudes Sci. Publ. Math.}, (75):5--95, 1992.


Candidate references: `section:1` (external_reference); `zhodgesection` (external_reference); `totreal` (external_reference); `rmk:line:1045` (external_reference).

### [68] MR645748 — `external:MR645748`

Source lines 2496–2504.

Y.~T. Siu and S.~T. Yau. \newblock Compactification of negatively curved complete {K}\"{a}hler manifolds of finite volume. \newblock In {\em Seminar on {D}ifferential {G}eometry}, volume 102 of {\em Ann. of Math. Stud.}, pages 363--380. Princeton Univ. Press, Princeton, N.J., 1982.


Candidate references: `algebraic` (external_reference); `sectiongeneral` (external_reference).

### [69] MR1238407 — `external:MR1238407`

Source lines 2504–2510.

W.-K. To. \newblock Total geodesy of proper holomorphic immersions between complex hyperbolic space forms of finite volume. \newblock {\em Math. Ann.}, 297(1):59--84, 1993.


Candidate references: `sectiongeneral` (external_reference); `poincaregrowth` (external_reference).

### [70] MR3744855 — `external:MR3744855`

Source lines 2510–2515.

J.~Tsimerman. \newblock The {A}ndr\'{e}-{O}ort conjecture for {$\mathcal{A}_g$}. \newblock {\em Ann. of Math. (2)}, 187(2):379--390, 2018.


Candidate references: `pointsection` (external_reference).

### [71] zbMATH04097530 — `external:zbMATH04097530`

Source lines 2515–2520.

H.~{Tsuji}. \newblock {A characterization of ball quotients with smooth boundary}. \newblock {\em {Duke Math. J.}}, 57(2):537--553, 1988.


Candidate references: `actionofc` (external_reference).

### [72] MR3177266 — `external:MR3177266`

Source lines 2520–2525.

E.~Ullmo. \newblock Applications du th\'{e}or\`eme d'{A}x-{L}indemann hyperbolique. \newblock {\em Compos. Math.}, 150(2):175--190, 2014.


Candidate references: `sectionfinit` (external_reference).

### [73] MR3728310 — `external:MR3728310`

Source lines 2525–2532.

E.~Ullmo. \newblock Structures sp\'{e}ciales et probl\`eme de {Z}ilber-{P}ink. \newblock In {\em Around the {Z}ilber-{P}ink conjecture/{A}utour de la conjecture de {Z}ilber-{P}ink}, volume~52 of {\em Panor. Synth\`eses}, pages 1--30. Soc. Math. France, Paris, 2017.


Candidate references: `section:1` (external_reference).

### [74] MR2825237 — `external:MR2825237`

Source lines 2532–2537.

E.~Ullmo and A.~Yafaev. \newblock A characterization of special subvarieties. \newblock {\em Mathematika}, 57(2):263--273, 2011.


Candidate references: `prop:line:1928` (external_reference).

### [75] MR1633348 — `external:MR1633348`

Source lines 2537–2543.

L.~van~den Dries. \newblock {\em Tame topology and o-minimal structures}, volume 248 of {\em London Mathematical Society Lecture Note Series}. \newblock Cambridge University Press, Cambridge, 1998.


Candidate references: `sectiongeneral` (external_reference).

### [76] MR0279206 — `external:MR0279206`

Source lines 2543–2548.

E.~B. Vinberg. \newblock Rings of definition of dense subgroups of semisimple linear groups. \newblock {\em Izv. Akad. Nauk SSSR Ser. Mat.}, 35:45--55, 1971.


Candidate references: `section:2` (external_reference); `integrality` (external_reference); `totreal` (external_reference); `rmknew` (external_reference).

### [77] MR136682 — `external:MR136682`

Source lines 2548–2553.

A.~Weil. \newblock Algebras with involutions and the classical groups. \newblock {\em J. Indian Math. Soc. (N.S.)}, 24:589--623 (1961), 1960.


Candidate references: `section:2` (external_reference).

### [78] MR137792 — `external:MR137792`

Source lines 2553–2558.

A.~Weil. \newblock On discrete subgroups of {L}ie groups. \newblock {\em Ann. of Math. (2)}, 72:369--384, 1960.


Candidate references: `section:1` (external_reference).

### [79] MR137793 — `external:MR137793`

Source lines 2558–2563.

A.~Weil. \newblock On discrete subgroups of {L}ie groups. {II}. \newblock {\em Ann. of Math. (2)}, 75:578--602, 1962.


Candidate references: `section:1` (external_reference).

### [80] MR0169956 — `external:MR0169956`

Source lines 2563–2568.

A.~Weil. \newblock Remarks on the cohomology of groups. \newblock {\em Ann. of Math. (2)}, 80:149--157, 1964.


Candidate references: `section:2` (external_reference); `zhodgesection` (external_reference).

### [81] MR670072 — `external:MR670072`

Source lines 2568–2575.

A.~Weil. \newblock {\em Adeles and algebraic groups}, volume~23 of {\em Progress in Mathematics}. \newblock Birkh\"{a}user, Boston, Mass., 1982. \newblock With appendices by M. Demazure and Takashi Ono.


Candidate references: `section:2` (external_reference).

### [82] MR931211 — `external:MR931211`

Source lines 2575–2580.

J.~Wolfart. \newblock Werte hypergeometrischer {F}unktionen. \newblock {\em Invent. Math.}, 92(1):187--216, 1988.


Candidate references: `section:1` (external_reference).

## Coverage and limits

{"mode": "native_latex_index", "claims": 100, "sections": 7, "external_references": 82, "source_bibliography_entries": 82, "unresolved_bibliography": 0, "unresolved": 1, "curated_proof_edges": 22, "proof_graph_complete": false}

- Reference edges are not certified proof dependencies. Edge direction is prerequisite → consumer.
- Numbers reflect simple source counters; labels and line anchors are authoritative.
- Extractive statements and proof openings are marked; they are not reconstructed proof gists.
- Macros, included files and custom counters require an expanded source or annotated fallback.
- Curated proof-use edges cover the adjudicated bottlenecks only; the complete inventory is retained but the proof DAG is partial.
- Historical independent results and author corrections are context, not replacements for the original proof.
- Display numbers listed in the adjudication were checked in the v3 author PDF, not assumed to be the journal numbering.
