(Springer Monographs in Mathematics) Haruzo Hida - P-Adic Automorphic Forms On Shimura Varieties-Springer (2004)

Springer Monographs in Mathematics
Springer
New York
Berlin
Heidelberg
Barcelona
Budapest
Hong Kong
London
Milan
Paris
Singapore
Tokyo
Haruzo Hida
p-Adic Automorphic Forms

on Shimura Varieties
Springer
Haruzo Hida
Mathematics Department
UCLA
Box 951555
Los Angeles, California 90095
Mathematics Subject Classification (2000): llF33, llF41, llF46, llG18, 14G35
Library of Congress Cataloging-in-Publication Data

Hida, Haruzo.
p-adic automorphic forms on Shimura varieties / Haruzo Hida.
p. cm. - (Springer monographs in mathematics)
Includes bibliographical references and index.
ISBN 978-1-4419-1923-6 ISBN 978-1-4684-9390-0 (eBook)
DOI 10.1007/978-1-4684-9390-0
I. Shimura varieties. 2. Automorphic forms. 3. p-adic analysis. 1. Title. II. Series.
QA242.5.H53 2004
515'.9-dc22 2003070362
ISBN 978-1-4419-1923-6 Printed on acid-free paper.
© 2004 Springer-Verlag New York, LLC

Softcover repnnt of the hardcover I st editIOn 2004
All rights reserved. This work may not be translated or copied in whole or in part without the
written permission of the publisher (Springer-Verlag New York, LLC, 175 Fifth Avenue, New York,
NY 10010, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use
in connection with any form of information storage and retrieval, electronic adaptation, computer
software, or by similar or dissimilar methodology now known or hereafter developed is forbidden.
The use in this publication of trade names, trademarks, service marks, and similar terms, even if
they are not identified as such, is not to be taken as an expression of opinion as to whether or not
they are subject to proprietary rights.
9 87 6 5 4 32 I SPIN 10950739
Springer- Verlag is a part of Springer Science+Business Media
springeronline.com
Preface
In the early years of the 1980s, while I was visiting the Institute for Ad-
vanced Study (lAS) at Princeton as a postdoctoral member, I got a fascinating
view, studying congruence modulo a prime among elliptic modular forms, that
an automorphic L-function of a given algebraic group G should have a canon-
ical p-adic counterpart of several variables. I immediately decided to find out
the reason behind this phenomenon and to develop the theory of ordinary
p-adic automorphic forms, allocating 10 to 15 years from that point, putting
off the intended arithmetic study of Shimura varieties via L-functions and
Eisenstein series (for which I visited lAS). Although it took more than 15
years, we now know (at least conjecturally) the exact number of variables for
a given G, and it has been shown that this is a universal phenomenon valid
for holomorphic automorphic forms on Shimura varieties and also for more
general (nonholomorphic) cohomological automorphic forms on automorphic
manifolds (in a markedly different way).
When I was asked to give a series of lectures in the Automorphic Semester
in the year 2000 at the Emile Borel Center (Centre Emile Borel) at the
Poincare Institute in Paris, I chose to give an exposition of the theory of
p-adic (ordinary) families of such automorphic forms p-adic analytically de-
pending on their weights, and this book is the outgrowth of the lectures given
there.
In addition to an exposition of p-adic deformation theory of automor-
phic forms, I decided to give a comprehensive account of the construction
of Shimura varieties (carrying a canonical family of abelian varieties) and
Shimura's reciprocity laws characterizing them. Although this part is essen-
tially expository, some new results of the author are included:
1. a (partially) new proof of some reciprocity laws of Shimura varieties;
2. irreducibility of the Igusa tower over unitary Shimura varieties;
3. a construction of invertible sheaves on Hilbert modular varieties that gives
optimal p-integrality on Hilbert modular forms.
It is my hope that this book is useful to graduate students and researchers
entering this area of research.
The author would like to thank the organizers of the Automorphic Semester
at the Emile Borel Center for their invitation and also wishes to thank many
people including his former and present graduate students who helped him
to write correct English and mathematics. The author acknowledges partial
support from the National Science Foundation (through the research grant:
DMS 9988043 and DMS 0244401).
Los Angeles, California,

September 2003 Haruzo Hida
Contents
1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Automorphic Forms on Classical Groups. . . . . . . . . . . . . . . . . . . 5
1.2 p-Adic Interpolation of Automorphic Forms. . . . . . . . . . . . . . . . . 8
1.3 p-Adic Automorphic L-functions .......................... 12
1.4 Galois Representations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 13
1.5 Plan of the Book. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 13
1.6 Notation............................................... 15
2 Geometric Reciprocity Laws. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 17

2.1 Sketch of Classical Reciprocity Laws. . . . . . . . . . . . . . . . . . . . . .. 18
2.1.1 Quadratic Reciprocity Law ......................... 18
2.1.2 Cyclotomic Version. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 19
2.1.3 Geometric Interpretation. . . . . . . . . . . . . . . . . . . . . . . . . .. 20
2.1.4 Kronecker's Reciprocity Law ........................ 21
2.1.5 Reciprocity Law for Elliptic Curves. . . . . . . . . . . . . . . . .. 24
2.2 Cyclotomic Reciprocity Laws and Adeles . . . . . . . . . . . . . . . . . .. 24
2.2.1 Cyclotomic Fields. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 24
2.2.2 Cyclotomic Reciprocity Laws ....................... 26
2.2.3 Adelic Reformulation .......... . . . . . . . . . . . . . . . . . . .. 28
2.3 A Generalization of Galois Theory. . . . . . . . . . . . . . . . . . . . . . . .. 31
2.3.1 Infinite Galois Extensions .......................... 31
2.3.2 Automorphism Group of a Field. . . . . . . . . . . . . . . . . . . .. 35
2.4 Algebraic Curves over a Field. . . . . . . . . . . . . . . . . . . . . . . . . . . .. 36
2.4.1 Algebraic Function Fields .......................... 36
2.4.2 Zariski Topology .................................. 43
2.4.3 Divisors.......................................... 44
2.4.4 Differentials...................................... 45
2.4.5 Adele Rings of Algebraic Function Fields. . . . . . . . . . . .. 50
2.5 Elliptic Curves over a Field. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 51
2.5.1 Dimension Formulas ............................... 51
2.5.2 Weierstrass Equations of Elliptic Curves. . . . . . . . . . . . .. 52
viii Contents
2.5.3 Moduli of Weierstrass Type. . . . . . . . . . . . . . . . . . . . . . . .. 54

2.5.4 Group Structure on Elliptic Curves. . . . . . . . . . . . . . . . .. 55
2.5.5 Abel's Theorem. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 56
2.5.6 Torsion Points on Elliptic Curves. . . . . . . . . . . . . . . . . . .. 57
2.5.7 Classical Weierstrass Theory. . . . . . . . . . . . . . . . . . . . . . .. 60
2.6 Elliptic Modular Function Field . . . . . . . . . . . . . . . . . . . . . . . . . .. 62
3 Modular Curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 67
3.1 Basics of Elliptic Curves over a Scheme .................... 67
3.1.1 Definition of Elliptic Curves ............... . . . . . . . .. 68
3.1.2 Cartier Divisors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 68
3.1.3 Picard Schemes ........ . . . . . . . . . . . . . . . . . . . . . . . . . .. 69
3.1.4 Invariant Differentials. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 70
3.1.5 Classification Functors. . . . . . . . . . . . . . . . . . . . . . . . . . . .. 70
3.1.6 Cartier Duality ................................... 71
3.2 Moduli of Elliptic Curves and the Igusa Tower. . . . . . . . . . . . .. 72
3.2.1 Moduli of Levell over Z[il ......................... 72
3.2.2 Moduli of Pr,(N) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
3.2.3 Action of G m . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 75
3.2.4 Compactification.................................. 77
3.2.5 Moduli of r(N)-Level Structure. . . . . . . . . . . . . . . . . . . .. 79
3.2.6 Hasse Invariant ................................... 80
3.2.7 IgusaCurves ...................................... 81
3.2.8 Irreducibility of Igusa Curves ...... . . . . . . . . . . . . . . . .. 82
3.2.9 p-Adic Elliptic Modular Forms. . . . . . . . . . . . . . . . . . . . .. 85
3.3 p-Ordinary Elliptic Modular Forms . . . . . . . . . . . . . . . . . . . . . . .. 86
3.3.1 Axiomatic Treatment .............................. 86
3.3.2 Bounding the p-Ordinary Rank ..................... 89
3.3.3 p-Ordinary Projector .............................. 90
3.3.4 Families of p-Ordinary Modular Forms . . . . . . . . . . . . . .. 90
3.4 Elliptic A-Adic Forms and p-Adic L-functions . . . . . . . . . . . . . .. 92
3.4.1 Generality of A-Adic Forms ......................... 92
3.4.2 Some p-Adic L-Functions ........................... 94
4 Hilbert Modular Varieties ................ . . . . . . . . . . . . . . . .. 97

4.1 Hilbert-Blumenthal Moduli ............................... 98
4.1.1 Abelian Variety with Real Multiplication. . . . . . . . . . . .. 98
4.1.2 Moduli Problems with Level Structure ............... 102
4.1.3 Complex Analytic Hilbert Modular Forms ............ 104
4.1.4 Toroidal Compactification .......................... 110
4.1.5 Tate Semi-Abelian Schemes with Real Multiplication .. 115
4.1.6 Hasse Invariant and Sheaves of Cusp Forms ........... 117
4.1.7 p-Adic Hilbert Modular Forms of Level r(N) ......... 119
4.1.8 Moduli Problem of rl(IJ1)-Type ..................... 123
4.1.9 p-Adic Modular Forms on PGL(2) . .................. 125
Contents ix
4.1.10 Hecke Operators on Geometric Modular Forms ........ 127

4.2 Hilbert Modular Shimura Varieties ........................ 131
4.2.1 Abelian Varieties up to Isogenies .................... 133
4.2.2 Global Reciprocity Law ............................ 140
4.2.3 Local Reciprocity Law ............................. 154
4.2.4 Hilbert Modular Igusa Towers ....................... 158
4.2.5 Hecke Operators as Algebraic Correspondences ........ 164
4.2.6 Modular Line Bundles ............................. 165
4.2.7 Sheaves over the Shimura Variety of PGL(2) ......... 175
4.2.8 Hecke Algebra of Finite Level ....................... 177
4.2.9 Effect on q-Expansion .............................. 178
4.2.10 Adelic q-Expansion ................................ 184
4.2.11 Nearly Ordinary Hecke Algebra with Central Character 189
4.2.12 p-Adic Universal Hecke Algebra ..................... 191
4.3 Rank of p-Ordinary Cohomology Groups ................... 192
4.3.1 Archimedean Automorphic Forms ................... 192
4.3.2 Jacquet-Langlands-Shimizu Correspondence .......... 198
4.3.3 Integral Correspondence ............................ 202
4.3.4 Eichler-Shimura Isomorphisms ...................... 205
4.3.5 Constant Dimensionality ........................... 206
4.4 Appendix: Fundamental Groups ........................... 209
4.4.1 Categorical Galois Theory .......................... 209
4.4.2 Algebraic Fundamental Groups ...................... 216
4.4.3 Group-Theoretic Results ........................... 218
5 Generalized Eichler-Shimura Map ......................... 225

5.1 Semi-Simplicity of Hecke Algebras ......................... 225
5.1.1 Jacquet Modules .................................. 226
5.1.2 Double Coset Algebras ............................. 227
5.1.3 Rational Representations of G ...................... 230
5.1.4 Nearly p-Ordinary Representations .................. 232
5.1.5 Semi-Simplicity of Interior Cohomology Groups ....... 234
5.2 Explicit Symmetric Domains .............................. 236
5.2.1 Hermitian Forms over <C .........................•.. 236
5.2.2 Symmetric Spaces of Unitary Groups ................ 238
5.2.3 Invariant Measure ................................. 243
5.3 The Eichler-Shimura Map ................................ 244
5.3.1 Unitary Groups ................................... 245
5.3.2 Symplectic Groups ................................ 247
5.3.3 Hecke Equivariance ................................ 248
x Content:;
6 Moduli Schemes ........................................... 251

6.1 Hilbert Schemes ......................................... 251
6.1.1 Vector Bundles .................................... 252
6.1.2 Grassmannians .................................... 253
6.1.3 Flag Varieties ..................................... 256
6.1.4 Flat Quotient Modules ............................. 258
6.1.5 Morphisms Between Schemes ....................... 262
6.1.6 Abelian Schemes .................................. 264
6.2 Quotients by PGL(n) .................................... 269
6.2.1 Line Bundles on Projective Spaces ................... 270
6.2.2 Automorphism Group of a Projective Space .......... 270
6.2.3 Quotient of a Product of Projective Spaces ........... 271
6.3 Mumford Moduli ........................................ 276
6.3.1 Dual Abelian Scheme and Polarization ............... 276
6.3.2 Moduli Problem ................................... 276
6.3.3 Abelian Scheme with Linear Rigidification ............ 278
6.3.4 Embedding into the Hilbert Scheme ................. 279
6.3.5 Conclusion ....................................... 280
6.3.6 Smooth Toroidal Compactification ................... 282
6.4 Siegel Modular Variety ................................... 290
6.4.1 Moduli Functors .................................. 291
6.4.2 Siegel Modular Reciprocity Law ..................... 293
6.4.3 Siegel Modular Igusa Tower ......................... 296
7 Shimura Varieties .......................................... 303

7.1 PEL Moduli Varieties .................................... 304
7.1.1 Polarization, Endomorphism, and Lattice ............. 304
7.1.2 Construction of the Moduli ......................... 309
7.1.3 Moduli Variety for Similitude Groups ................ 314
7.1.4 Classification of G ................................. 316
7.1.5 Generic Fiber of Sh~) ............................. 317
7.2 General Shimura Varieties ................................ 321
7.2.1 Axioms Defining Shimura Varieties .................. 321
7.2.2 Reciprocity Law at Special Points ................... 324
7.2.3 Shimura's Reciprocity Law ......................... 326
8 Ordinary p-Adic Automorphic Forms ...................... 329

8.1 True and False Automorphic Forms ........................ 329
8.1.1 An Axiomatic Igusa Tower ......................... 330
8.1.2 Rational Representation and Vector Bundles .......... 331
8.1.3 Weight of Automorphic Forms and Representations .... 333
8.1.4 Density Theorems ................................. 335
8.1.5 p-Ordinary Automorphic Forms ..................... 339
8.1.6 Construction of the Projector eCL . . . . . . . . . . . . . . . . . . . 341
8.1. 7 Axiomatic Control Result .......................... 344
Contents Xl
8.2 Deformation Theory of Serre and Tate ..................... 345

8.2.1 A Theorem of Drinfeld ............................. 346
8.2.2 A Theorem of Serre-Tate ........................... 348
8.2.3 Deformation of an Ordinary Abelian Variety .......... 349
8.2.4 Symplectic Case ................................... 350
8.2.5 Unitary Case ..................................... 351
8.3 Vertical Control Theorem ................................ 355
8.3.1 Hecke Operators on Deformation Space .............. 356
8.3.2 Statements and Proof .............................. 360
8.4 Irreducibility of Igusa Towers ............................. 363
8.4.1 Irreducibility and p-Decomposition Groups ........... 364
8.4.2 Closed Immersion into the Siegel Modular Variety ..... 364
8.4.3 Description of a p-Decomposition Group ............. 367
8.4.4 Irreducibility Theorem in Cases A and C ............. 369
References . .................................................... 375
Symbol Index . ................................................. 383
Statement Index . .............................................. 385
Subject Index ................................................. 387

1
Introduction
The two major trends in number theory, automorphic and arithmetic, were
initiated hand in hand in the mid nineteenth century by Kronecker and Kum-
mer. One of Kronecker's legacies is the theory of arithmetic elliptic modular
functions (and modular forms) having well-determined algebraic values (up
to a specific transcendental factor: "period") at special points on modular
curves. Shimura varieties (including modular curves) were principally studied
by Shimura and Deligne in the later part of the last century. In particular, the
theory gives a foundation of the rationality of automorphic L-values, because
Shimura varieties often supply enough rationality for us to be able to pinpoint
the transcendental factors of specific L-values (the automorphic periods).
On the other hand, the arithmetic side of number theory heavily influ-
enced by Kummer's work is represented by Iwasawa's theory which interprets
the algebraic part of the L-value as the size of a naturally given group (or
module) defined by purely arithmetic means. The origin of this thought goes
back to the class number formula of Dirichlet, Kummer, and Dedekind, which
gives the size of the ideal class group of a number field as a special value
of an abelian L-function (up to a power of the period 27fi of the exponen-
tial function). The main conjecture in Iwasawa's theory predicts precisely the
same effect, replacing the class group by the so-called "Selmer group" and
its "Iwasawa module" (attached to automorphic Galois representations) and
abelian L-values by automorphic p-adic L-functions (see, for example, [MFG]
Chapter 5).
Although these two trends seemed rather disparate until recently, the strik-
ing idea of Wiles (which has proven two of the most fundamental conjectures
in number theory) has spanned a bridge between the two concepts. Indeed, it
gives a way to find an automorphic form from a purely arithmetically given ob-
ject, for example, a rational elliptic curve whose L-function is directly related
to an elliptic modular Hecke L-function via the Shimura-Taniyama conjecture
(now proven by Wiles and others; see [Wi2] and [BrCDT]).
This type of intricate relation between arithmetic and geometry can also
be found in classical reciprocity laws. There are two aspects of the Artin reci-
H. Hida, p -Adic Automorphic Forms on Shimura Varieties

© Springer-Verlag New York, LLC 2004
2 1 Introduction
pro city law in class field theory. One is representation theoretic, for example,
for the ring A(oo) of finite adeles,
via the identity of L-functions. Here Qab is the maximal abelian extension
of Q (so, it is a composite of all Galois extensions of Q inside C with an
abelian Galois group) and Q~ is the group of positive rational numbers (see
Theorem 2.12). Another geometric reciprocity can be formulated as
Since this is a description of the field automorphism group Gal(Qab /Q) by

an adelic Lie group: GL 1 (A(oo)), we call it a geometric reciprocity law. They
are equivalent by Pontryagin duality (which characterizes a locally compact
abelian group by its characters), and the first is generalized (conjecturally) by
Langlands in a non-abelian setting. Geometric reciprocity in a non-abelian set-
ting would be via Tannakian duality. Tannakian duality characterizes a given
(algebraic) group by the category of its rational representations with some ad-
ditional data ([HMS] Chapter II and [Br]). A simple class of Shimura varieties
(so-called, of PEL type) is given as a moduli variety of abelian schemes with
a given endomorphism algebra (or equivalently, a moduli of motives of the
first-degree cohomology of such abelian schemes). Such a motive H generates
a Tannakian category whose dual group MTH/IQ (the so-called motivic Galois
group of the motive) could be related in possibly different ways to the auto-
morphism group of the Shimura variety. If the motive is sufficiently generic,
the adele points of the group MTH would give almost all automorphisms of
the Shimura variety (cf. [D3] 1.7). Thus Shimura variety appears as a natural
object of study if one wants general geometric reciprocity laws.
Iwasawa's theory is built upon the geometric reciprocity law. The cy-
clotomic field Q(flpoo) is the maximal p-ramified extension of Q fixed by
Z(p) c A x /Qx ~~ removing the p-inertia multiplicative factor Z;. We then
try to study arithmetically constructed modules X (Iwasawa modules) from
Q(flpoo) C Qab; for example, Iwasawa himself studied the projective limit X
(under the norm map) of the p-primary part of the class groups of number
fields inside Q(flpoo). A principal idea is to regard X as a module over the
. . . GL (A(oo») )
Iwasawa algebra ( which is a completed Hecke algebra relative to GLl~i(p»)IQ~ ,
and ring-theoretic techniques are used to determine X (see [ICF] Chapter 13).
General Tannakian duality produces from the category of all (pure) mo-
tives an enormous group, the so-called motivic Galois group (see [HMS] Chap-
ters II through IV and [MTV]). However, for Iwasawa-theoretic applications,
the full motivic Galois group appears to be hopelessly large to adapt. If one
wants to create something like Iwasawa's theory in a non-abelian setting, we
therefore need a scheme whose automorphism group has an identification with
G(A(oo))/Z(Q) for a reductive algebraic group G. If G = GL(2)/IQ, the tower
1 Introduction 3
V/lQiabof modular curves has Aut(V/IQi) identified with GL 2 (A( CXJ l)/Z(Q) as
Shimura proved (see Theorem 2.43). The (geometric) decomposition group of
(p) is given by B(Qp) x SL 2 (AJp CXJ l)/{±1} for a Borel subgroup B (see Corol-
lary 4.22). We have been studying various arithmetically constructed modules
over the Hecke algebra of ~!;2;~(7)\
GL2 Z U
p zp
x, relative to the unipotent subgroup
1Qi+
U(71 p ) c B(71 p ) (removing the multiplicative inertia factor from the decom-
position group). Often these are modules given by cohomology groups (either
algebrogeometric or topological). Such study has yielded a p-adic deforma-
tion theory of p-ordinary automorphic forms (see Chapter 8), and it would
be important to study the decomposition group at p of a given Shimura vari-
ety, which is close to the automorphism group of the mod p Shimura variety.
We describe quite a few automorphism groups related to Shimura varieties of
PEL type coming from symplectic and unitary groups (following Shimura's
method and also using slightly more algebraic methods). In particular, the
study of the p-decomposition subgroup provides us with a simple proof of the
irreducibility of the Igusa tower over the mod p Shimura variety (see Sec-
tion 8.4), which supplies us with one of the most useful tools in the p-adic
study of automorphic forms: the q-expansion principle.
Besides determination of the automorphism groups of Shimura varieties,
we present a concrete construction of p-integral models of Shimura varieties
(following works of Shimura, Crothendieck, Mumford, Deligne, and Kottwitz)
and the p-adic deformation theory of p-ordinary automorphic forms, in partic-
ular, the theory of analytic families of p-ordinary (cohomological) automorphic
forms on reductive groups.
We are particularly interested in Shimura varieties that are canonically
moduli varieties of abelian schemes with extra structures. Since abelian
schemes are very specific projective subschemes of a projective space, we can
study abelian schemes through classification theory of projective sub schemes
in a given projective space. The classification was initiated by W.-L. Chow and
was completed by Crothendieck as the theory of Hilbert schemes. In the late
1950s (e.g., [59b] in [CPS] I), Shimura used the Chow coordinate associated
with each projective variety in his determination of the field of moduli of a
given family of abelian varieties (which was later incorporated as an essential
ingredient in his construction of canonical models of automorphic varieties;
see [AAF] 2.8 and Chapter II), and in [CIT] Mumford used Hilbert schemes to
study the integral moduli of abelian schemes, which can be used to construct
Shimura varieties. We give a down-to-earth exposition of the construction of
the Hilbert schemes and the Mumford integral moduli in Chapter 6.
Assuming basic knowledge of abelian schemes (which can be found in
[ABV]) , toroidal compactifications of the moduli (in [DAV]) , and class field
theory, we then give an exposition of the construction of the Shimura varieties
and their models over p-adic integer rings. The Shimura variety is character-
ized by its local and/or global recipmcity laws. The global reciprocity law gives
a geometric expression of the automorphism group of the function field of the
4 1 Introduction
Shimura variety and essentially determines the variety, because the Shimura
variety is often a minimal model of the function field. The local reciprocity
law determines via class field theory the move of special points, called CM
points, under Galois action. The CM points are those points carrying abelian
varieties with complex multiplication (also often called abelian varieties of
CM type) and are Zariski dense in the variety, thereby providing rationality of
the variety. An abelian variety has complex multiplication if its endomorphism
algebra contains a semi-simple commutative algebra of degree over Ql twice
the dimension of the variety (see [ACM] 5.2 and [ABV] Section 19). The first
principal purpose of this book is to supply the base of the construction of
the (p-integral) Shimura variety in as elementary a manner as possible (in-
cluding an exposition of the local and global reciprocity laws), because it is a
pity that no other books (so far published) give a detailed scheme-theoretic
construction of the variety now so important in every aspect of algebraic and
geometric number theory. Shimura's books [AAF] and [EPE] give an elemen-
tary exposition on rationality of unitary and Siegel Shimura varieties and of
automorphic forms and L-functions on them and will be indispensable for
graduate students and researchers who want to enter this area of research.
After going through our exposition of Shimura varieties and p-adic auto-
morphic forms on them, natural subjects of our exposition are
1. p-adic deformation theory of p-ordinary automorphic forms;
2. p-adic automorphic L-functions;
3. p-adic Galois representations;
4. the Iwasawa-theoretic significance of the p-adic L-functions.
In this book, we mainly touch upon the first topic with some examples of
p-adic L-functions, leaving other topics to future treatments. Thus the second
purpose of this book is to introduce integrality of automorphic forms on the
varieties and hence to supply a foundation in order for researchers to pursue
p-adic analytic study of automorphic forms and automorphic L-values.
Here is a brief history of p-adic modular and automorphic forms. The the-
ory of p-adic elliptic modular forms was initiated by Serre in [Se3] taking the
p-adic limit of q-expansions of elliptic modular forms. The geometric signif-
icance of Serre's definition was revealed by Deligne and Katz (e.g., see [K],
[Kl], and [K3]). Serre observed that the Eisenstein series of a given level Np
has Fourier coefficients p-adically analytically depending on its weight k, and
through this observation, a construction of p-adic zeta functions of totally
real fields was given in [Se3], which was later brought to completion in [DR]
through a more geometric formulation. A systematic study of elliptic cuspi-
dal families of Hecke eigenforms p-adic analytically depending on weight was
started by the author of this book in [H86a] and [H86b] and has been devel-
oped into a p-adic deformation theory of p-ordinary automorphic forms on
Shimura varieties and classical groups (e.g., [H02], [H98], and [TiU]). More
recently, Mazur, Gouvea, and Coleman (e.g., [GM], [Co], and [Wa]) extended
the theory to non-p-ordinary elliptic modular forms with positive slope. In par-
1.1 Automorphic Forms on Classical Groups 5
ticular, Mazur and Gouvea conjectured, and Coleman has proven, that there
always exists a p-adic analytic family of elliptic modular forms containing a
given modular form with a positive slope, although only p-ordinary families
seem to have a precise relation to classical objects (typical in Iwasawa's the-
ory; see (3) in Section 1.2), because only for p-ordinary automorphic forms
is p-adic analyticity of their coefficients described by Iwasawa's functions be-
longing to the Iwasawa algebra (and its finite flat extensions). So p-ordinary
automorphic forms enjoy stronger p-adic analyticity than nonordinary forms.
The deformation theory has found numerous profound applications by many
different researchers. For example, it was used as an essential ingredient of a
construction of two-dimensional Galois representations from Hilbert modular
forms [Wi], of a proof of Iwasawa's conjecture in [Wil], of the proof of Fermat's
last theorem ([Wi2 ], [TaW], and see also [MFG] Chapter 3), of the proof of
the Shimura-Taniyama conjecture ([Wi2], [BrCDT] and see also [GME] Chap-
ter 5), and of the proof of the Artin conjecture for many nonsoluble icosahedral
Galois representations (e.g., [BuT] and [BuDST]). More recently, the linear
growth (relative to the degree of the base totally real fields) of the number
of variables of the p-ordinary family of Hilbert modular forms was used by
Skinner and Wiles ([SkW]) to prove modularity of residually reducible (but
irreducible) two-dimensional p-adic Galois representations.
Since our second principal goal is to give an introductory exposition of p-
adic automorphic forms, in the following sections of this introductory chapter,
we give an overview of such automorphic forms (covering all four topics),
putting off to the following chapter an introductory discussion on reciprocity
laws (including a brief history).
1.1 Automorphic Forms on Classical Groups
Let G m be a classical (reductive) group and G1/rQ be the derived group of G.

Take a Borel subgroup Bm of G 1 and its torus T/rQ c G 1. When G is split
over CQl, we may embed G into G L( n) /rQ so that we may take B to be the
group of upper triangular matrices in G 1 and T to be the group of diagonal
matrices. We have B = T ~ U for the unipotent radical U of B. On the
quotient variety Gr/U (see Section 6.1.3), T acts by gUt = gtU, and hence T
acts on the structure sheaf OCl/u by t¢(gU) = ¢(gtU). This action gives rise
to an order on X(T) = Homal g . gp.(T/QJl CG m ) so that the positive cone in X(T)
is made of r;, such that the r;,-eigenspace L(r;,) on the global sections of OCl/u
is nontrivial. A character r;, E X (T) is called the weight of T, and a weight
is called dominant if it is in the positive cone. We then have a representation
L(r;,; R) = Lc(r;,; R) on L(r;,) given by ¢(gU) >-1 ¢(h-lgU) for h E G 1(R), as
long as T is split over a ring R. Often we can naturally extend the action of
G 1 on L(r;,; R) to G (and we proceed hereafter in the introduction regarding
Lc(r;,; R) as a G-module). When G = GL(2), T ~ CG m , X(T) ~ Z by r;, H n if
r;,( x) = x n , and L( r;,; R) is the r;,th symmetric tensor representation of G £(2),
6 1 Introduction
which can be realized on the space of homogeneous polynomials of degree n so

that a E GL(2) acts on a polynomial P(X, Y) by a· P(X, Y) = P((X, y)ta')
with a' = det(a)a- 1 .
Naively, an automorphic form on G is an L(;;,; C)-valued analytic function
on the symmetric space of the Lie group G1(~) with a prescribed move (or
functional equation) under the action of a congruence subgroup r of G(Q).
A congruence subgroup of G(Q) is a subgroup given by S n G(Q) for an open
compact subgroup S of G(A,(oo») for the ring A,(oo) of finite adeles (see Sec-
tion 2.2.3 in the text and [MFG] Chapter 3 for adeles). A congruence subgroup
is a discrete subgroup of G 1 (~) with finite covolume (i.e., Ic! (1R)/ r! dg < 00 for
the Haar measure dg of G 1 (~) and r 1 = r n G 1 (Q)). A congruence subgroup
r is called cocompact if G 1 (~) / r1 is compact.
There are two ways of associating a weight with automorphic forms on
G: one is to consider the cohomology group Hd(r, L(K; R)) of an appropriate
degree d for a given congruence subgroup r c G(Q) and call harmonic au-
tomorphic forms spanning Hd(r, L(K; C)) automorphic forms of (topological)
weight K. This works well for any classical (or more general reductive) groups.
When the symmetric space of G(~) is isomorphic to a (bounded) Her-
mitian domain 1{ with origin 0, like (the restriction of scalars to Q of) F-
forms of GSp(2n) or GU(m, n) over totally real fields F, we have another
way to associate a weight with holomorphic automorphic forms. We have
7-l ~ Gd~)/Co = G(~)/CoZ(~) for the center Z of G and the stabilizer Co
in G 1 (~) of 0, which is a maximal compact subgroup of G 1 (~). In the case of
GL(2)/Q = GSp(2)/Q, Co = S02(~) and 7-l = Sj := {z E q Im(z) > O} with
G1(~)/S02(~) ~ Sj by g H g(A) (so, = A ° E Sj). As is well known, Sj
is holomorphically equivalent to the open unit disk in C by z H :~~ (so it
is equivalent to a bounded domain).
The group Co can be regarded as a group of real points with respect to
a twisted complex conjugation in the complexification C of Co. In the case
of GL(2)/Q, S02(~) can be considered as Sl in T(C) = Gm(C) by (~:1) H
c A + d E Sl, and Sl is the set of fixed points of the twisted "complex
conjugation" x H x- 1 in T(C) = C X • Generalizing this example, we see that
the compact group U(n) is the subgroup of GLn(C) fixed by the complex
conjugation x H t x -1. Any holomorphic representation p : C ---+ GL(V(p))
gives rise to a holomorphic complex vector bundle V = (G1(~) x V)/Co by
the actio~ (g, v) H (gu, u- 1 v) for u E Co. Since 7-l is simply connected, we
can split V ~ 1{ x V as holomorphic vector bundles; so, we have a linear map
Jp(g, z) : Vz ---+ Vg(z) for each given g E G 1 (~) which identifies the fibers Vz
and Vg(z) of 17. Thus we have a function Jp : G1(~) x 7-l---+ GL(V) satisfying
1. (Cocycle Relation) Jp(gh, z) = Jp(g, h(z))Jp(h, z) for g, h E G1(~);
2. (Holomorphy) Jp(g, z) is holomorphic in z and nonvanishing.
A function Jp as above is called an automorphicjactorofG 1 (associated with
the representation p). Automorphic factors of G are defined similarly.
1.1 Automorphic Forms on Classical Groups 7
When G 1 = SL(2) /Q' then Co = S02(lR) c C = T(C) whose irreducible

complex representation is given by
( c~s(e)
sm(e)
- sin(e)) =
cos(e)
(*cd*) H (* *) = (ci + d)k.
P cd
In this case, Jp(g, z) = (cz + d)k. This goes as follows. Split GL 2 (lR) = PCo
for P made of upper triangular matrices with 1 at the lower right corner. For
z = x + iy, define pz = (6 ~). Then for 9 E SL 2 (lR), write gpz = Pg(z)u with
u E Co, and we have p(u) = p(p;(~)gpz) = (cz + d)k by computation. Indeed,
Jp(g, z) sends (v,Pz) to (uv,Pg(z)) ~ (v, gpz) = (v,Pg(z)u).
One can view the complexification C as a real algebraic group; let Te be
a maximal real torus of C. With any character K of Te, we can associate a
rational representation p = p~ : C H GL(N) of C with representation space
Lc(K; C) of dimension N. Let V(K) = Lc(K; C). For a congruence discrete
subgroup
r c G(Q), a holomorphic automorphic form of (coherent) weight K is a
holomorphic function J : H H V(K) satisfying J([(z)) = J p ([, z)J(z) for all
'Y E r (with some additional growth condition if r\ H is not compact). Again
the space of holomorphic automorphic forms is trivial unless the weight K is
dominant (with respect to B chosen appropriately).
Often the complex manifold r\ H is canonically algebraizable, yielding
an algebraic variety (or a scheme) X r , called canonical models or Shimura
vaT'ieties, defined over a valuation ring W in a number field with residual
characteristic p. At the same time, we can algebraize the vector bundle V(K)
associated wit~ V(K). Thus we can often have a coherent sheaf i±JK on Xr
giving rise to V(K) after extending the scalar to C. The global sections of
HO(Xr,i±JKR) for a W-algebra R are called R-integral automorphic forms of
weight K. Note that Te is isomorphic to T, because they are maximal tori in
the same group G 1 . Thus we identify T and Te (with a compatible choice of
Borel subgroups B and Be = B n C). On X r , we may regard the r-module
LC(K; R) as a locally constant sheaf associating, with an open subset U c X r ,
sections over U of the covering space L(K; R) = r\(H x LC(K; R)) over X r .
Here the quotient r\(H x LC(K; R)) is taken through the diagonal action.
Thus each dominant weight K E X(T) gives two spaces of automorphic forms:
Hd(Xr, LC(K; R)), GK(r; R) := HO(X r , i±J/R)'

There is (at least conjecturally) a correspondence K H K* such that
HO(Xr,i±JK) Y Hd(Xr,Lc(K*;C))
by a generalized "Eichler-Shimura map" which is supposed to be equivari-
ant under Hecke operators. If such equivariance holds, we say that the two
modules, the source and the image, are equivalent as Heeke modules. In the
example of GL(2)/Q, we have r c SL 2(71), K E X(T) = X(CG m ) = 7l, and
K* = K - 2 with
8 1 Introduction
and the original Eichler-Shimura map is given by sending J to the cohomology

class [J(z)(X - zY)K- 2 dz] of the holomorphic L(r;, - 2; C)-valued differential
J(z)(X - zY)K- 2 dz ([59c] and [62a] in [CPS], [IAT] Chapter 8, and see also
[LFE] Chapter 6). This is valid if r;, 2:: 2.
1.2 p-Adic Interpolation of Automorphic Forms
We now interpolate these two families of cohomology groups
when the weights r;, vary continuously in the p-adic analytic group D =
Homtop_gp(T(Zp), Q;) (fixing an algebraic closure Qp of Qp). Here to describe
our idea of how to interpolate automorphic forms, we write W for the p-adic
completion of W.
On these two families of spaces, there is a natural action of Hecke op-
erators; so, we also want this interpolation to take into account the Hecke
operators. To carry out the p-adic interpolation, we need a good p-integral
structure on the group G; so, we hereafter assume that G is defined over
Z(p) = Q n Zp as an affine group scheme and that G 1 is isomorphic over Zp to
the restriction of the scalar to Zp of a classical Chevalley group over the p-adic
integer ring Op of an unramified finite semi-simple extension of Qp; so, the
Borel subgroup B and its maximal torus T is well defined over Z(p), and T is
split over Op. Thus, if Op = Zp, T(Zp) ~ (Z;r for the rank r of Glover Qp,
and D is (isomorphic to) a disjoint union of finitely many copies of the p-adic
(open) unit disk. In general, T(Zp) is a product of a torsion-free multiplicative
group rT isomorphic to the additive group Z; and a finite group. Then the
Iwasawa algebra W[[T(Zp)]] (which is the projective limit of the group alge-
bras {W[T(Z/pmZ)]}m) is a free module of finite rank over W[[rT]], which is
in turn (noncanonically) isomorphic to the power series ring W[[T1 , ..• , Tn]]
of n variables. What we now do is:
(1) (Universality) Construct a (big) compact module V over W[[T(Zp)]]
such that the r;,-eigenspace V[r;,] contains canonically, as W[[T(Zp)]]-
submodules, Hd(Xr, L(r;,; W)) in the topological case, and HO(Xr / w , ~K)
in the coherent case, respectively;
(2) (Hecke operators) Establish a natural action of Hecke operators on V, and
show the inclusion in (1) is Hecke equivariant;
(3) (Precise control) Find an appropriate W[[T(Zp)]]-submodule X c V of
cofinite type ({o} W -dual is of finite type) such that X is stable under
Hecke operators and X[r;,] (if r;, » 0) is canonically isomorphic to a well-
described subspace of automorphic forms of weight r;, as Hecke modules.
1.2 p-Adic Interpolation of Automorphic Forms 9
Item (3) is often called a vertical control theorem (abbreviated as VCT) of

the subspace X. Contrary to the case of complex-valued automorphic forms,
p-adic automorphic forms, as a whole, regarded as a representation space of
G(A(poo)), have little discrete spectrum, and they tend to involve big contin-
uous spectrums. A vertical control theorem cuts out from the entire space
of p-adic automorphic forms a continuous spectrum well behaving under the
toric action. Examples of VCT are given as Theorem 3.7 for elliptic modular
forms, Theorem 3.8 for a p-adic family of elliptic modular forms, Theorem 4.9
for Hilbert modular forms, Corollary 4.32 for Hilbert modular Hecke algebras
and Theorem 8.13 for automorphic forms on unitary groups. A more general
result on VCT can be found in [H02].
We mainly deal with the coherent case where G admits Shimura varieties
that are given as moduli of abelian varieties with level structure (the so-called
PEL structure: a polarization and points of finite order, a specified endo-
morphism algebra, and a specific lattice giving integral homology of abelian
varieties). We give an exposition of the construction of the p-integral models
of the variety in Chapters 6 and 7. However, we need to use some results
obtained from the topological consideration; so, some sections in Chapter 5
are also devoted to describing the situation in topological cases.
In the coherent case, we define V to be the space of p-adic automorphic
forms that classify abelian schemes with "ordinary" level poo structure in
addition to a PEL structure outside p. In other words, V is the space of formal
functions on a formal (pro- )scheme, called the Igusa tower, which is the formal
modulus classifying abelian schemes over p-adically complete W -algebras with
"ordinary" level pOO-structure. We prove a vertical control theorem for the
space X = v ord of nearly p-ordinary automorphic forms and prove that its
W-dual Homw(V Ord , W) is W[[T(Zp)]]-projective of finite type.
Actually, we can give for any classical group a good definition of nearly
p-ordinary cusp forms; that is, a cusp form is called nearly p-ordinary if the
Newton polygon of its Hecke polynomial at p is equal to the hypothetical
Hodge polygon mechanically constructed from the weight K, (of the motive
attached to the cusp form). We prove that the Newton polygon is always
on or above the Hodge polygon (without recourse to hypothetical motives);
so, a nearly p-ordinary form has the lowest Newton polygon identical to the
Hodge polygon (see Chapter 5). For a Hecke eigenform of high weight (i.e., the
weight is inside the positive cone of dominant weights far from its boundaries),
whether it is nearly p-ordinary for almost all primes p is a difficult question to
answer (cf. [Gu]). If the eigenform has low weight (conjectured to be associated
with Artin Galois representations or abelian varieties), the density 1 property
of ordinary primes is expected, and it is known for elliptic modular forms of
weight 2 (see [Se4] Section 7) and is obvious for Artin representations. Thus
lifting such a low weight p-ordinary form to a family of nearly p-ordinary
automorphic forms is at present the only theoretical way to produce high-
weight nearly p-ordinary forms.
10 1 Introduction
One would expect that vo rd := Homw(Vo rd , W) is W[[T(Zp)]]-projective

of finite rank if G admits Shimura varieties. Contrary to this, when we deal
with a group such as GL(n) (n > 2) without having Shimura variety, the
module vo rd is of finite type over W[[T(Zp)]], but we can prove that it is of
torsion (although there is a conjectural formula of its Krull dimension; see
[H95] and [H98]). Natural questions are:
(QI) When can one expect that the space vo rd is W[[T(Zp)]]-projective?
What is the (expected) minimal value of Ii at which the vertical control
holds? What happens if one specializes at very low weight? If vo rd is
torsion, what is the Krull dimension of the W[[T(Zp)]]-module vo rd ?
What is its characteristic power series if the codimension of vo rd
equals I in Spec(W[[T(Zp)]])?
It turns out that all these questions are quite arithmetic, as we show in the
principal text. In the elliptic modular case, the lowest weight where VCT
holds is 2. However, as Buzzard and Taylor studied in [BuT], there is a good
criterion via Galois representations to guarantee the limit at weight I to be
a true modular form (not just p-adic), which played an important role in the
proof of the Artin conjecture for some icosahedral cases by the school of R.
Taylor ([SBT], [Tal], and [Ta2]).
Write X[pn] for the kernel of multiplication by pn of an abelian scheme
X (as a locally free group scheme over the base), and define its Barsotti-
Tate group (p-divisible group) by X[POO] = ~nX[pn]. Our description of
automorphic forms is geometric, whose origin goes back to some work of Katz
and Deligne-Ribet (cf. [K3] and [DR]). In the example of G = GL(2)jQ,
we take an arbitrary p-adically complete W -algebra R = ~ nRj pn R, and
consider a test object (E,cPp,cPN)jR made of an elliptic curve E, a level poo_
structure cPp : /Lp= '-+ E[POO] (i.e., a closed immersion of ind-group schemes)
and a level N-structure cPN, like a point of order N (an inclusion of Zj NZ into
the set of N-torsion elements E[N] in E), all these data being defined over
R. A p-adic elliptic modular form f is a functorial rule associating an element
of R to a test object (E,cPp,cPN)jR. Thus we have f(E,cPp,cPN) E R, and for
each p-adically continuous W -algebra homomorphism R -4 R', we have
Though f is a functorial morphism (as we said a "rule"), hereafter abusing

the language, we pretend as if f were a function of test objects (obeying the
specific rules we have described). A modular form which is an eigenform of
T(p) (or U(p)) is p-ordinary if its eigenvalue at p is a p-adic unit. In gen-
eral, p-ordinary modular forms are linear combinations of such (we give a
more thorough definition in the text; see Section 3.3). The evaluation of f
at the Tate curve Tate(q)jZ((q)) (see Section 3.2.4) at the cusp 00 yields the
q-expansion:
1.2 p-Adic: Interpolation of Automorphic: Forms 11
L
00
f(q) = f(Tate(q)oo, ¢~an, ¢J!r) = a(n, f)qn.

n=O
We can deduce from the irreducibility of the Igusa tower that V is the p-adic
completion of
We may identify the Fourier expansion of f E Gk(Tl (N); q with the q-

expansion f(q) E C[[q]], writing q = exp(27riz).
In the topological case of G = GL(2)/Q)) V is given by the cohomology
group H1(T',C(f'jU(Zp), W)), where C(f'jU(Zp), W) is the space of COn-
tinuous functions on the p-adic analytic space f' jU(Zp), T' = Tl (N) n
To(p) and f' is the closure of T' in GL2(Zp). Then f' = rJi;), and
we may regard P E L(r;,; W) as a polynomial function on fo(p)jU(Zp) C
(SL(2)jU)(Zp), and this inclusion, L(r;,; W) '--+ C(fo(p)jU(Zp), W), induces
the map Hl(T', L(r;,; W)) '--+ V.
For any W[[T(Zp)]]-module X satisfying VeT, the eigenvalue A(t) of each
Heeke operator t acting on X is algebraic over W[[T(Zp)]]. Indeed, the Heeke
algebra h in Endw[[T(Zp)ll(X) generated by (appropriate) Heeke operators is
an algebra over W[[T(Zp)]] of finite (generic) rank (or even of torsion). Take
an irreducible component Spf(lI) of the formal spectrum Spf(h) of h. The
operator t projected to II, written as A(t) (i.e., A : h ---» II is the projection), can
be considered to be a formal function on the formal scheme Spf(II). The p-adic
space Spf(II)(W) of W-points of Spf(lI) is given by the set of continuous W-
algebra homomorphisms of II into W. In particular, if P E Spf(II)(W) satisfies
PIW[[T(Zp)ll = r;, for r;, » 0, A(t)(P) = P(A(t)) is the eigenvalue of t occurring
in either HO(X r , (,,2.") or Hd(Xr, L(r;,; W)).
When G = GL(2)/IQ, T(Zp) = Z; = uZp xL1 for a finite group ,1. Thus
W[[T(Zp)]] = A[L1] for A = W[[u Zp ]] ~ W[[U]] (the classical Iwasawa algebra
and a formal power series ring) via US >--+ (1 + U)' = L~=o (~) un. Note that
The algebra homomorphism r;, : A ---+ W is the "evaluation" at U = UK - I!

Thus if II = A, Spf(A)(rIJ. p) is the open unit disk DO in ijp by r;, >--+ r;,(u) - 1,
and A(T(n))(r;,) = A(T(n))(u" -1) (viewing A(T(n)) as a power series) gives a
p-adic analytic interpolation of Heeke eigenvalues over DO. This type of p-adic
analytic function s >--+ <J5( US - 1, ... , US - 1) for a power series <J5(Tl' ... ,Tn) E
W[[T1, ... , Tn]] is called an Iwasawa function (which gives a special subclass
of p-adic analytic functions on the product of copies of the open unit disk DO).
In general, we get the p-adic interpolation of Heeke operators parameterized
by Spf(II)(W).
12 1 Introduction
1.3 p-Adic Automorphic L-functions
Fix embeddings ioo : Q -+ C and ip : Q '-+ Qp- Since the specialization

)"(T(n))(uK. - 1) (in the simplest case of GL(2)) is the Hecke eigenvalue oc-
curring in the space of cusp forms, it can be considered as a complex number
uniquely. Thus we can think of automorphic L-functions L(s, ).,(11:)) made from
such eigenvalues; for example, the modular Hecke L-function
L i().,(T(n))(uK. -
00
L(s, ).,(11:)) = l))n- S ,

n=l
writing i = ioo 0 i; 1.
In general, for sufficiently positive weight II: of T, we have Hecke eigenvalues
).,(t)(II:) occurring on the coherent cohomology with coefficients in '<d..K. on the
Shimura variety. We thus have a family {).,( II:)} K. of systems of eigenvalues
(or equivalently, a family of holomorphic automorphic representations). We
give ourselves a rational representation r of the Langlands dual LG of G
and consider the family of automorphic L-functions {L(s, ).,(11:), r)}K.. Suppose
that L(m, ).,(11:), r) for (sufficiently many) integers m (depending on 11:) has
rationality up to a transcendental factor or a period D(m, 11:, r). Then natural
questions we would ask are:
(Q2) Is it possible to interpolate p-adically the value Lh(~,~~~)) ? Is it pos-
sible to find L)..,r(U) E lI[[U]] such that the p-adically analytic function
Zp x SpJ(lI)(W) 3 (s, P) t-+ Lp(s, P, r) = P(L)..,r(u S - 1))
satisfies the identity: Lp(m, 11:, r) = Lh(~,~~~')) (up to a standard modi-

ficationfactor) for points (m, 11:) E ZxSpJ(lI)(W) with II:IT(Zp) E X(T)
densely populated in Zp x Spf(lI)(W)?
This problem of course involves a subtle integrality question of how to nor-
malize the factor {D(m,lI:,r)}K.,m in the aggregate (varying II: E X(T) and
integers m) to obtain an "optimal" integrality; so, it is more involved than
proving rationality (see Section 3.4 for some examples and [SGL] for a general
theory). Once we are successful in constructing canonical p-adic L-functions,
we could ask more specifically,
(Q3) When is the p-adic L-function entire? Where could it have singulari-
ties? If there is a singularity, what is the order of the zero or the pole?
If we have a pole, what is the residue?
See [SGL] for particular G and r for which we have a partial answer to these
questions and for some conjectural discussion on these questions.
1.5 Plan of the Book 13
1.4 Galois Representations

Once an irreducible component IT of the Hecke algebra is given, one would
expect to have a Galois representation p>, : Gal(Q/ F) --+ LG(IT) such that the
Hecke polynomial at e i- p gives rise to the characteristic polynomial of the
Frobenius element. Here abusing our language, LG is the p-integral Langlands
dual defined in an appropriate manner (see [DGH]). We could heuristically (or
conjecturally) answer the question (Q3) that the associated p-adic L-function
Lp (s, P, r) has singularity at P if the Galois representation r 0 p>, specialized
at P; that is, (r 0 p>,) mod P contains the trivial representation (a p-adic
Artin conjecture; see [SGL]). We then further ask
(Q4) For a pair of representations (P>" r) as above, is there any good way to
associate a Selmer group Sel(r 0 p>,) so that the characteristic element
in IT of the Selmer group should be equal to the p-adic L-function or
its numerator?
See [MFG] Chapter 5 for a general description of Selmer groups. If this is
affirmative, then this gives a description of the zero-set of the p-adic L-function
in arithmetic terms. Related to this, VCT is often useful to identify the nearly
p-ordinary Hecke algebra with the nearly p-ordinary universal deformation
ring if at one weight the deformation ring with the fixed weight is identified
with the Hecke algebra of the specific weight (see [Wi2] Theorem 3.3, [MFG]
5.3.5, and [HM] Section 4.3 for this type of argument in the case of GL(2)/Q).
The argument proving VCT often yields another type of control theorem: the
so-called horizontal control theorem (HCT; e.g., [GME] 3.1.2 and 3.2.5), giving
a precise description of the behavior of a Hecke algebra if one adds primes
outside p to the level of the Hecke algebra. This horizontal control (HCT) is
used in the case of G L (2) to construct the Tay lor-Wiles systems which was
invented by Wiles and Taylor in the celebrated papers [Wi2] and [TaW]. With
the help of the system, Wiles obtained the identification of the Hecke algebra
of a specific weight with the Galois deformation ring (an essential ingredient of
Wiles' proof of the Shimura-Taniyama conjecture and Fermat's last theorem).
See [Wi2J, [MFG] Chapter 3, and [GME] Chapter 3 for these topics.
1.5 Plan of the Book

We answer some of the above questions in this book in some specific cases
in a concrete way and in some other cases conjecturally. Since this book is
an outgrowth of a series of lectures given at the Emile Borel Center in its
Automorphic Semester of the year 2000, it would be helpful to include here
the original plan of the lectures to give a brief outline of the structure of the
book. Here is the plan.
1. In the first two chapters (Chapters 2 and 3), we recall the classical reci-
procity law and its generalization in the theory of the elliptic modular
14 1 Introduction
forms. Chapter 2 is from the notes of a graduate course I gave at UCLA

in the academic year 1999-2000, and from Chapter 3 on, the book is an
expanded version of the lectures at the Emile Borel Center (see [H03a] for
the original lecture notes). We also give a brief proof of the VCT in the
elliptic modular case as a prototype of the general theory.
2. Chapter 4 is devoted to another example: Hilbert modular forms. There
we prove diverse vertical control theorems as well as the freeness of the uni-
versal nearly ordinary Hecke algebra over the Iwasawa algebra of T(Zp).
We also give fairly detailed proofs of global and local geometric reciprocity
laws for Hilbert modular functions, which are slightly more algebraic than
Shimura's treatment (although the basic ideas behind the proof are the
same).
3. In Chapter 5, we define near-ordinarity of co homological automorphic
forms by comparing the p-adic Newton polygon and the Hodge polygon
of Hecke eigenforms. Furthermore we prove semi-simplicity of the Hecke
operator action on topological nearly ordinary cohomology groups (for
general reductive groups).
4. In Chapter 6, we recall the basic techniques (due to Grothendieck and
Mumford) of constructing algebrogeometric moduli schemes. We state and
prove, at the end, Shimura's global reciprocity law for Siegel modular
functions.
5. In Chapter 7, we first construct the Shimura varieties via the integral
moduli theory of abelian schemes. Thereafter we recall briefly the global
reciprocity law for general Shimura varieties not necessarily of PEL type
(without any detailed proof).
6. In Chapter 8, we prove the VCT for unitary and symplectic groups. A
key ingredient is the deformation theory (due to Serre-Tate) of p-ordinary
abelian schemes. At the end, combining the global reciprocity law and
the deformation theory, we prove the irreducibility of the Igusa tower for
unitary and symplectic groups. This supplies us with the q-expansion prin-
ciple for Shimura varieties of symplectic and unitary groups. We should
point out that the computation of cuspidal monodromy which success-
fully proved the irreducibility of the Igusa tower in the Siegel modular
case ([DAV] V.7) does not yield the irreducibility for groups of the form
ResF/IQIG o for unitary and symplectic groups Go over a totally real field
F bigger than Q by a reason explained in [DT] Section 12 (see also a
remark after Theorem 4.21). This point is subtly related to the Leopoldt
conjecture for F and p.
The author hopes to write another book (expanding Chapter 4) to cover the
theory of automorphic p-adic L-functions, horizontal control theorems, and
the construction of Galois representations in the Hilbert modular case, which
have been very useful in proving some main conjectures in the (generalized)
Iwasawa theory and in identifying Hecke algebras treated in this book with
1.6 Notation 15
Galois deformation rings (with a specific universal property; see [FI], [SkW],
and [MFG]).
Some of the papers and preprints of mine related to these subjects can be
downloaded from my web site: www.math.ucla.edu{hida.
1.6 Notation
In this book, we continue to use the notation utilized in this introduction. Here
we describe some of the notation which is used without much explanation. The
symbol Zp denotes the p-adic integer ring inside Qp, and the symbol Z(p) is
used to indicate the valuation ring Zp n Q. A sub field E of ij is called a
number field (often assuming dimlQl E < 00 tacitly). For a number field E,
DE denotes the integer ring of E, DE,p = DE 0z Zp c Ep = E 01Ql Qp and
DE,(p) = DE 0z Z(p) c E. The symbol W is exclusively used to indicate a
valuation ring inside ij with residual characteristic p, and often W is supposed
to be unramified over Z(p). The p-adic completion ~ n W /pnW is denoted by
W, and we write Wm = W/pmw = W/pmw.
The symbol A is the adele ring of Q, and for a vector space of a number
field E, we write VA = V(A) for V 01Ql A. For a subset E of the set of rational
places, we set A(E) = {x E Alx v = 0 for VEE} and ZE = TIpEEZp, where p
runs over finite places in E. We then define Z(E) = ZE n Q. If E = {oo}, {p},
or {p, oo} for a prime p, we write A (00), A (p) or A (poo) for A (E) accordingly. If
E is a set of finite places, we write A(Eoo) for A(EU{oo}). We identify A(E)X
with {x E A x Ixv = 1 for vEE} in an obvious way. The maximal compact
subring of A(oo) is denoted by Z, which is identified with the profinite ring
c
TIp Zp = ~ NZ/NZ. For any Z-module c of finite type, we write for c 0z Z.
An algebraic group T (defined over a subring R of ij) is called a torus
if its scalar extension T jij = T 0R ij is isomorphic to a product G~
of copies of the multiplicative group G m . The character group X*(T) =
HOmalg_gp(Tjij, G mjij ) is simply denoted by X(T), and elements of X(T) are
often called weights of T.
The symbol G is exclusively used for a reductive linear algebraic group
defined over Q. Then G 1 is its derived group of G (as an algebraic group
defined over Q), and Z is the center of G. We use the symbol PG = Gad for
G/Z as an algebraic group over Q. For the real points G(JR) of G, we regard
it as a Lie group, and G(JR)+ is the identity-connected component of the
Lie group G(JR). The symbol B often indicates a standard (upper triangular)
Borel subgroup inside G, and the letters U and N are used to indicate the
unipotent radical of B. The letter U is used when N is used to indicate an
integer (and N is used when U is used to denote a unitary group). The torus
B/U is denoted by Te, and the maximal torus of G 1 induced by Te is denoted
by T.
2
Geometric Reciprocity Laws
Since reciprocity laws play an important role in this book, we now describe
several reciprocity laws appearing in number theory in terms of automorphism
groups of a field fixing a given prime or a given point. Only this chapter
contains (elementary) exercises.
To fix our idea, we describe here briefly a model reciprocity law. For a
given field ji, we consider the field automorphism group Aut(ji) equipped with
the Krull topology (which is described in Section 2.3). As is clear from the
introduction, a geometric global reciprocity law gives a canonical description
of Aut(ji) by the adelic points of an algebraic (reductive) group G/iQ (modulo
rational center Z(Ql)): Aut(ji) '---+ G~~») typically. If a local ring V with
quotient field ji is given, we could define a decomposition subgroup D of
Aut(ji) (equipped with the Krull topology) by
D = {O' E Aut(ji) IO'(V) = V} ,
and the local reciprocity law for V describes D via a well-defined algebraic
subgroup H of G. Often D is given by the image of adele points, H(A(oo)) (or
Qlp-points, H(Qlp)), of H in G~~»). This description may involve a subtle use
of class field theory, and class field theory supplies us with one of the simplest
examples of the reciprocity laws. The global reciprocity law is a particular
case of the local one (taking V = ji). To have such reciprocity laws, the field
ji cannot be finitely generated over its prime subfield except for the case where
G is finite. So for an algebraic group G of positive dimension, the reciprocity
laws get more geometric flavor, and ji is then realized as the function field of
an infinite tower V of (geometrically irreducible) algebraic varieties defined
over a tower of finite extensions over the prime field. Then if we have local
reciprocity laws covering sufficiently many points of the tower V, the laws
characterize a global model V of ji (a model of a field K is an algebraic variety
whose function field is isomorphic to K in a canonical way). Thus to have a
coherent system of local reciprocity laws is almost equivalent to having a well-
defined global model of a given function field (of infinite type). Similarly, if V

18 2 Geometric Reciprocity Laws
is minimal, the algebrogeometric automorphism group Aut(V) of the variety

V coincides with Aut(J:t), and hence the tower is realized as a collection of
algebraic varieties {VslVjS}s for S running through open compact subgroups
S of Aut(J:t), where VjS is the quotient variety of V by the action of S. Thus
by the global reciprocity law, we can recover each member of the tower. This
type of dichotomy appears in the theory of Shimura varieties.
In the later part of the book, we study more modern reciprocity laws (in-
volving reductive groups of positive dimension) principally created by Shimura
as an explicit description of automorphisms of Shimura varieties. In this be-
ginning part of the book, we describe some reciprocity laws (including those
relating finite groups G with the field automorphism groups) that only re-
quire minimal knowledge of algebraic geometry. Our requirement is reasonable
knowledge of the theory of algebraic curves over number fields in this chapter.
Some of the reciprocity laws are stated here without proof, though in the later
chapters proofs are given in a more general setting. It would be a good exercise
for the readers to deduce the results described in this chapter from the more
elaborate versions in the later chapters. In the following chapter, we extend
our scope in order to incorporate integral models of modular curves, and in
the later chapters, we study Shimura varieties and their integral models via
the language of schemes.
2.1 Sketch of Classical Reciprocity Laws
In this section, we sketch the classical reciprocity laws starting with the his-
toric quadratic reciprocity invented by Euler (and proved by Gauss), and
ending with a reciprocity law for a single rational elliptic curve. Here, primes
p and q are always distinct odd primes.
2.1.1 Quadratic Reciprocity Law
For an integer n (p t n), the Legendre symbol (~) is defined by
if x 2 == n mod p has a solution,

otherwise.
Since ::z:2 == n mod p has a solution if and only if n E (IF;)2 for lFp = ZjpZ,
n H (~) gives an identification of IF; j(lF;)2 with {±1}; in particular, n H
(~) is a character of the finite multiplicative group (ZjpZV.

The quadratic reciprocity law guessed by Euler and proven by Gauss,
(~) = (_1)(p-l)(q-l)/4 (~) (Legendre, 1785),

2.1 Sketch of Classical Reciprocity Laws 19
has an equivalent formulation due to Euler
(~) = (p;) (Euler, 1744),
where p* = (_1)(p-l)/2 p . For a general odd positive integer n prime to p,

define (p~) = ITq (Pq*) e(q) if the prime factorization of n is given by ITq qe(q).
Thus n f--t (If) is equal to the character: n f--t (~).
This character decides how a prime decomposes in it = Q[ v'P*). We look
into the ring R = Z[v'P*) = Z[X]/(X 2 _pO). In R, a prime ideal (q) of Z could
remain prime or become a product of two prime ideals; that is, (q) = qq' or
(q) = q for prime ideals q, q' in R. Note that R/(q) = lFq[X]/(X2 - p*). The
polynomial X 2 - p* is reducible in lFq[X] if and only if it has a solution in lFq:
(q)=qq' {==} R/(q)~lFqEBlFq {==} (~*) =1,
where q is the kernel of the projection of R onto the first factor IFq of IF q EB IFq.
The nontrivial automorphism a ofQ[v'P*) interchanges the roots of X 2 _p*
and interchanges q and q'; so, q' = a(q). Identifying {±1} with Aut(it) =
Gal(Q[v'P*)/Q), (z;i)
gives the generator of the subgroup of Aut(it) fixing a
prime factor of (q).
If (q) = q remains prime, dimlFq R/(q) = 2, and a(q) = q. The stabilizer of
q is the entire Galois group, and (( p;)) = {±1} ~ Aut(it) gives the stabilizer.
In summary, identifying Aut(it) with {±1}, the subgroup generated by (z;i)
gives the stabilizer of a prime ideal qlq in R (in other words, the stabilizer of
the q-adic valuation ring V of it). The stabilizer is called the decomposition
subgroup of q. The information of the decomposition group of q is equivalent
to knowing how the prime (q) splits in R.
2.1.2 Cyclotomic Version
In the mid nineteenth century, Kummer extended the quadratic reciprocity

law to cyclotomic fields (in his study of decomposition of prime numbers into
a product of his "ideal" prime numbers). Let f.lp be the group of all pth roots
of unity, and consider the extension it = Q[f.lp] generated by pth roots of
unity. Fixing one primitive root of unity, say, ( = (p = exp e7r~), f.lp is a
cyclic group of order p generated by (p. Each automorphism a of f.lp takes (
to another primitive root of unity (m. Since (m is primitive, p f m and hence,
we have an identification Aut(f.lp) ~ IF; by Xp : a f--t m. Actually the finite
fiat group scheme f.lp over Q is the model of its function field it = Q[f.lp],
and Aut(it) = Gal(Q[f.lp]/Q) = Aut(f.lp). Since a E Gal(Q[f.lp]/Q) induces an
element in Aut(lLp) (and [Q[lLp] : Q] = IlF; I), we see that Gal(Q[lLpl/Q) ~ IF;
by XP' We write ¢q E Gal(Q[lLp]/Q) with Xp(¢q) = q; so, ¢q(() = (q.
Again we ask how a prime (q) decomposes in the ring R = Z[lLp], Pick a
prime ideal qlq in R; we find that R/q is a finite extension of lFq; so, it is of the
form lFqf for f = dimlFq R/q. The Galois group Gal(lFqf /lFq) is a cyclic group
of order f generated by a canonical generator F taking x E IF qf to x q E IF qf ;
thus, the automorphism F of lFqf is induced by ¢q. In other words,
• The decomposition group of qlq is given by (¢q) ~ (q) elF;, and
• q is of order f in IF; {=} (q) = qa(q)··· a 9 - 1 (q) in R for the integer
g = [Q[lLp] : Ql/ f,
where a is the generator of Gal(Q[lLpl/Q)/(¢q)' Thus one feature of the reci-

procity law is the determination of the decomposition group of a given prime
q in a given Galois extension K /Q.
2.1.3 Geometric Interpretation
To further generalize the law, we need to ponder a philosophical reason why

we have such an arithmetic way of describing the decomposition group. One
feature of the cyclotomic version is the existence of a canonical generator
¢q (the Frobenius element at q) of the decomposition group, and another is
the appearance of the exponential function exp(z), which has the following
fundamental identity,
So, roughly, the complex analytic function exp : C ---+ C x contains all the
information of the reciprocity law, and exp(2ni;;) gives a canonical generator
( of the field Q[lLp] and behaves nicely under its Galois conjugation.
The function exp gives rise to the following exact sequence,
and thus we evaluated the function exp at the fraction 2ni;; E ;;2niZ/2niZ.
The period 2ni of the exponential function "exp" is important to ensure that
the value of "exp" on 2niQ is algebraic, and we have 2ni = I, flt
for the unit
circle 'Y which generates the fundamental group nl(C X ) = Hl(CX,Z),
Hilbert asked in his twelfth problem (of his famous lecture in 1900 at
the International Congress of Mathematicians held in Paris; [HI]), for a given
Galois extension (actually an abelian extension in his original setting),
Is there any complex analytic function which describes fully the reciprocity law
of the extension?
2.1.4 Kronecker's Reciprocity Law
In the language of Poincare, the fundamental group of C x is given by Z,

and the universal covering of CX is given by C exp ) C x . We can create such
a situation starting with an imaginary quadratic field M = Q[J-D] with
discriminant -D < 0 in place of Q. For any given ideal a c R (for the integer
ring ReM) and a period [2 E C x, we consider an exact sequence:
(PP')
o -+ [2 a -+ C --'---'----'+ E (C) -+ O.
For the moment E(C) is a quotient space Cia, which is a Riemann surface of
genus 1. The period [2 of the elliptic curve E is inserted here to ensure that
the value on [2M of the Weierstrass functions P and pi (as defined below)
are algebraic. Writing w for the (translation-) invariant differential on E(C)
7fl(E(C)) = H 1 (E(C),Z) = a,
induced by du for the variable u of C, we have [2 = J'Y w for the generator, of
just as before. Indeed, for an lR-base (Wl,W2)
of C, we can think of EL(C) = CI L for L = ZWI + ZW2 replacing a by L.
Weierstrass studied analysis and geometry of Riemann surfaces (trans-
forming Riemann's marvelous but rather intuitive ideas into a rigorous math-
ematics). In particular, for the Riemann surface E(C) of genus 1, he created
the following function well-defined over E(C),
x(u) = P(u) = u 2 +
1 L (1 1)
(u _ £)2 - £2 '
£EL-{O}
averaging the translations of u\ over L, which is a two-dimensional analogue

of the partial fraction expansion of the cotangent function (see [LFE] 2.1):
7fcot(7fz) 1 {II}
= - +L
Z
-- + -- .
00
z+n z-n
n=1
This function converges absolutely over C - L and gives a meromorphic

function on E(C). The Laurent expansion of P and its derivative can be easily
computed, and we have
1 g2 2 g3 4 2 '"""'
x(u)=P(u)=u2+20u +2S u + ... , I
y(u)=P(u)=-u3+~anun,
n::::l
where g2 = g2(L) = 60 LNO £-4 and g3 = g3(L) = 140L£;",0£-6. The con-
stants g2 and g3 are actually complex analytic functions of W = (WI, W2).
Canceling the poles, we consider 'P = y2 - 4x 3 + g2X + g3 which is holo-
morphic everywhere on a compact Riemann surface E(C); so, it has to be
constant. The function 'P has to be identically 0, because 'P has no constant
term. Thus u r-+ E(u) = (u 3x(u), u 3y(u), u 3) E p2 embeds the Riemann
surface into the projective space of dimension 2, whose image is an alge-
braic curve (called an elliptic curve) defined by the homogeneous equation:
y 2Z = 4X3_g2XZ2_g3Z3 (x = XIZ and y = YIZ). We write .,1 = g§-27g~

for the discriminant of 4x 3 - g2X - g3.
If we pick >. E C x, we have a biholomorphic isomorphism of Riemann
surfaces: u f-t >.u of Ede) ~ EAde). By definition, g,(>'L) = >.-2'g,(L) and
.,1(>.L) = >.- 12 .,1(L). We have j(>.L) = j(L) for the j-invariant defined by
3
j (w) = j (L) = ~. The following facts are known from the general theory of
algebraic curves and are proven in this chapter.
• If ¢ : EL ---+ Eu is a holomorphic homomorphism with ¢(O) = 0, then there
exists>. E C X such that L' ::) >'L and ¢(u+L) = (>,u+L') (Theorem 2.39);
• Write a morphism ¢ : EL ---+ Eu as ¢(u) = (¢x(u), ¢y(u), 1) using the co-
ordinates of the projective space p2. Then ¢x and ¢y are rational functions
of XL and YL' that is ¢x = A(X,y,g2,g3) and ¢ = a(x,y,g2,g3) for polynomi-
, , B(X,y,g2,g3) y {3(X,y,g2,g3)
als A, B, 01, f3 with coefficients in Q independent of L (Corollary 2.26);
• Every genus 1 Riemann surface is obtained as EL for an L (Theorem 2.39);
• E LjIC ~ E ujlC <¢=} j(L) = j(L') (Corollary 2.35).
The function j is an example of a modular function and g2 and g3 are examples
of modular forms. Modular forms are a special kind of automorphic forms
defined on a more general complex domain.
For a given number field F, the maximal everywhere unramified abelian
extension HI F is called the Hilbert class field. Kronecker studied the Hilbert
class field H of the imaginary quadratic field M. Take a prime p of the integer
ring R of M and its prime factor >,p in the integer ring OH of H. The decompo-
sition group D~{ = {a E Gal(HI M) IO'(>,p') = >,p'} for any other prime factor
>,p'lp is a conjugate of D~ (as is well known). Since Gal(HIM) is abelian,
the group D~ is uniquely determined independently of the factor >,p of p; so,
we write Dp for D~. Since HIM is unramified (at p), Dp ~ Gal(Klk) for
the residue fields K = OHI>,p and k = Rip = Fq for a prime power q. The
Frobenius automorphism ¢p E Dp sending x E K to x q E K gives a canonical
generator of the Galois group Gal(Klk). What Kronecker found is
Theorem (Kronecker-Weber) Let HIM be the Hilbert class field of M.
Then H = M[j(a)] for any ideal 0 =1= a c Rand j(a)<Pp = j(p- 1a). In
particular, Gal( HIM) is isomorphic to the ideal class group of M.
Proof. We give a sketch of a proof. We first explain why j(a) is an algebraic

number. If End( E L) contains the integer ring R of an imaginary quadratic field
M, then RL c L; so, L c nM for 0 =1= n E L. Via multiplication by n, Cia ~
CI L for a = n- 1L. Thus we may assume that L is an ideal a of M. Since
the isomorphism class of E" only depends on a up to scalar multiplication,
there are only countably many isomorphism classes of EL with End(EL) ::) R.
Regard E" as a curve defined by the equation y2 = 4x 3 - g2x - g3, and
consider its conjugate E~ defined by y2 = 4x 3 - O'(g2)x - O'(g3) for any field
automorphism a of C. Then End(E~) contains R because all endomorphisms
are rational functions of (x, y, g2, g3); so, applying a to their coefficients, we
get an endomorphism ¢" E End(E~) from ¢ E End(Ea). The morphism

¢ f--t ¢" is an isomorphism of rings. Thus E~ = Eb for an ideal b in M and
j(a)" = j(b); so, j(a) has only count ably many conjugates. This implies that
they are finitely many, because if x E C is transcendental over Q, one can
embed Q(x) into C in continuously many different ways (C has continuously
many transcendental numbers). Thus the number of conjugates of j (a) is finite,
and hence j(a) is algebraic. Since j(b) = j(a) if and only if b = aa for some
a E M X , the fractional ideals of M modulo scalar multiplication are finitely
many. For a given j, we can create a Weierstrass equation y2 = 4x 3 - 92X - 93
in Q(j)[x, y] with invariant j, for example, if j rf. {O, I}, we put 92 = 93 = J~JI
(see just above Corollary 2.35 in the text or [IAT] 4.1). In other words, by
choosing the period Q as Q = J'Ydx =
Y
J'Y
w E C x, we have
Q-4 92 (a) = 92(Qa) = .27j = 93(Qa) E Q.

J -1
For almost all prime ideals IiJ of Q(j (a)), 92 and 93 are liJ-integral with L1( a) t=-
O mod 1iJ, and the equation modulo IiJ gives rise to an elliptic curve Ea defined
over a finite field IF. The relative Frobenius map F : (x, y) H (x P, yP) sends
Ea to another elliptic curve E~p) defined by y2 = 4x 3 - 9gx - 9~ E IF[x, y].
Suppose that (p) = IiJ n Z is split into pp in the imaginary quadratic field M.
Then the group of p-torsion points EalP] is the direct sum of the group of p-
torsion points and that ofp: EalP] = Ea[P]EBEa[P]. As is well known (cf. [GME]
2.9.1), we can choose p uniquely so that Ea[P] ~ R/p having p distinct points.
Since F : Ea --+ E~p) is a homomorphism of groups that is a zero map on
the tangent space, it induces a group isomorphism EaOF) ~ E~p\W) (though
not a scheme isomorphism), and hence E~p) ~ Ep-l a, because p : Ea --+ Ea
is an endomorphism (see [GME] 2.9.1) which factors Ea ~ E~p) ~ Ea with
Ker(V) = E~p)[p] = p-Ia/a. This shows that j(a)<Pp = j(p-1a). Since j(a)
characterizes the isomorphism class of Ea (over Q), we know that the j(a)s
indexed by a E elM for the ideal class group elM of M are all distinct. Since
in each class of elM, we can find split p outside a given finite set of primes
of M (the Chebotarev density), we find that {j(a)}aECI M are all conjugates
of each other over M; so, they span the Hilbert class field H/M. This proves
the above theorem (and finiteness of elM).
By the theorem (or its proof), Gal(H/M) ~ elM. This is the explicit class
field theory for the imaginary quadratic field M. This type of result is gener-
alized by Shimura, Taniyama, and Weil to eM fields, using a quotient Cd / L
for a lattice L of higher rank (theory of abelian varieties with complex multi-
plication; see Theorem 4.19 in the text, [IAT] Chapter 5, and [ACM]). A CM
field means a totally imaginary quadratic extension of a totally real field.
2.1.5 Reciprocity Law for Elliptic Curves
We can slightly generalize the above construction. Let

1 2
Edp] = {U E EL(C)lpU = O} = -L/L ~ lF p .
p
Note that E(O) = (0, -2,0) E P2(Q), and hence the point 0 of EL does not
move after applying a E Aut(C). Then a E Aut(C) brings p : EL -+ EL to
p: EE -+ EE because they are rational functions of (x, y, g2, g3); so, if EL is
defined over a number field M' = Q[g2(L),g3(L)], a E Aut(lC/M') induces a
linear automorphism of E dp]. Taking a base w = (u;,' , u;,2) of E L [P], we find
a(w) = wp(a) for a 2 x 2 matrix p(a) E GL 2 (lF p ), and so, we can identify
Gal(M'(Edp])/M') with a subgroup Im(p) of GL2(lFp). Here M'(EL[P]) is
the field generated by x(W,)
p
and y(W,)
p
(i = 1,2) over M'. Thus identifying
Edp] with the column vector space lF~ by t(a, b) f--7 wt(a, b) E EL[P]' we have
x(p(a)v) = x(v)O" and y(p(a)v) = y(vr.

This reciprocity law is still half baked, because we have not made explicit the
form of p( ¢q) for the canonical generator ¢q of the decomposition group of a
prime q of M'. This can be done when L is a fractional ideal of an imaginary
quadratic field, and the refined version is an example of Shimura's reciprocity
laws (although the origin of this reciprocity goes back to Kronecker). When
L is not in an imaginary quadratic field, Im(p) is almost always full (by a
result of Serre; see [ARE] and [Se2]) , and we can (conjecturally for general
M' f Q) make explicit Tr(p(¢q)) and det(p(¢q)) (see [GME] 5.2.4). However
this information is still a bit short of completely determining the splitting of a
prime q in M' (E dp]) (see [Sh2]), and the analysis of the Galois representation
p: Gal(M'(EL[P])/M') y GL2(lFp) is still a central subject today.
From what I said, it is clear that the study of modular functions and
modular forms (and their generalization, often called automorphic forms) is
natural and crucial in algebraic number theory, though they appear to be
rather analytic and geometric objects.
2.2 Cyclotomic Reciprocity Laws and Adeles

We now give a more detailed description of the cyclotomic reciprocity laws
and relate them to the idele class group of Q. This fact was found basically
by Kronecker and is one of the simplest examples of the reciprocity laws in
class field theory.
2.2.1 Cyclotomic Fields
We look into the exponential function e : lC -» lCx given by e(z) = exp(27riz).

For a positive integer N, we write ( = (N = e (1:t ). Then ( is a primitive
2.2 Cyclotomic Reciprocity Laws and Adeles 25
Nth root of unity. When N is a prime p, it satisfies the equation XP - 1 =

(X -l)Pp(X) = 0 for pp(X) = Xp-l +Xp-2 + ... + 1. Since ( =I- 1, pp(() = O.
Proposition 2.1 The cyclotomic polynomial pp(X) is irreducible in Q[X].
Proof. The irreducibility of Pp and lJi(X) = pp(X + 1) is equivalent. We see

easily that the constant term is given by lJi(O) = Pp(l) = p. Since (X + l)P -
1 = XlJi(X) == XP mod p, by the Eisenstein criterion, lJi(X) is irreducible.
o
Corollary 2.2 The equation Ppn(X) = pp(Xpn-l) for an integer n 2: 1 is
irreducible in Q[X].
The proof of this corollary is left to the reader as an exercise.
Corollary 2.3 Let N = pn for a prime p. The cyclotomic field Q[(N] is
a Galois extension of Q whose Galois group is abelian. We have a canon-
ical isomorphism of groups X = XN : Gal(Q[(N]/Q) ~ (7L/N7L)X given by
O"(e (-k)) = e (x~)) for 0" E Gal(Q[(N]/Q).
Proof. Let f.1N be the group of Nth roots of unity. Then e induces an isomor-
phism L : 7L/ N7L ~ f.1N of groups by m H e (N). Since 0"( (N) is another prim-
itive Nth root for any 0" E Gal(ij/Q) (Q is the algebraic closure of Q in C), 0"
induces an automorphism of the field Q[(N] because 0"( (N) is again a power of
(N. Thus Q[(N]/Q is a Galois extension. Since Aut(7L/N7L) = (7L/N7LV (mul-
tiplicative group) by ¢ H ¢(1), we find X = XN : Gal(Q[(N]/Q) -+ (7L/N7L) X •
If 0" E Ker(x), 0" fixes all Nth roots of unity; so, X is injective. Its p-component
Xpn has to be surjective, since
We show later the surjectivity of XN for general N (see Theorem 2.8).

Corollary 2.4 Let 7L(p) = {% la, b E 7L, b7L + p7L = 7L}. Then the integral do-
main V = 7L(p) [X]/(Ppn (X)) is a discrete valuation ring whose maximal ideal
is generated by w = (pn -1, where (pn is the image of X in V (so, it is a prim-
itive pn th root of unity which can be identified with e (p~ )). The valuation
ring V is fully ramified over 7L(p) , that is, (w)[v:z(p)l = (p).
Proof. By the fundamental theorem of arithmetic, any x = %can be written
as pv(x)y for y = ~ with p t c and p t d; so, y-l E 7L(p). Thus v: Q -+ 7Lu {oo}
gives the valuation of Q and 7L(p) = {x E Qlv(x) 2: O}; so, 7L(p) is a discrete
valuation ring of Q.
Let w = (p-1, and let V = 7L(p) [X]/(pp(X)). Since lJi(X) == Xp-l mod p,
we see that V /(p) ~ Fp[X]/(Xp-l). Then
(1) :J (X) :J (X2) :J ... :J (Xp-l) = (0)

are the only ideals of V I(p), because (aJXJ + aJ+IXJ+1 + ... + ap_lXp-1) =
(X)J if aJ Ie 0 in IFp. Since p = tli(O) = TIaEGal(IQ[(pl/lQ) cr(w), we find that
(w) ~ (p). Then by the homomorphism theorem applied to V --» V I (p), only
ideals between (p) and (w) are (w)m for m = 1,2, ... ,p - 1. By induction on
n, one can show that
are the only ideals of V I (pn). For a given ideal a C V, a n Z(p) = (pC) because
Z(p) is a discrete valuation ring (DVR). Thus a = (w)m for O:S; m:S; jl(p-1).
This shows that V is a DVR with valuation w given by w(x) = n ~ (x) =
(w)n. The same argument as above works well for V = Z(p) [X]j(Ppn (X)) for
w = (pn - 1, since Ppn(X + 1) == Xpn-l(p-l) mod p. D
Let V be a DVR. Then we can extend the valuation v of V to its field of

fractions K by v(f;) = v(a) - v(b). In other words, V = {x E Klv(x) :2: O}.
For any x E K, we have either x E V or x-I E V. In particular, if x E K is
integral over V, then x satisfies an equation
with aJ E V. If x tf. V, v(x) < 0 ~ v(x- 1) > 0, and
0= v(l) = v( -alx-1 - a2x- 2 - ... - anx- n ) :2: min(v(aJ ) + jv(x- 1)) > 0,
J
which is a contradiction; so, x E V, and V is integrally closed. In particular

Z(p) [(pn] is the integral closure of Z(p), and (p) fully ramifies in Z(p) [(pn].
2.2.2 Cyclotomic Reciprocity Laws
Let q be a prime different from p. We now look into Vq = Z(q) [X]I (Ppn (X)) =
Z(q)[(pn]. We look into <J5pn(X) = (Ppn(X) mod q) E IFq[X]. Then VqlqVq =
IFq[X]/(<J5pn(X)) and dimlFq VqlqVq = rankz(q) Vq = deg(ppn(X)). Since we
have pn-l (p - 1) distinct primitive pnth roots of unity in IFq, <J5pn (X) does not
have multiple roots, and we see Vq I qVq ~ IF qh EB IF qh EB ... EB IF qig . The image
of X in IF qIJ is a primitive pnth root a = a) of unity in IF qIJ , and IF qIJ = IF q[a].
Thus f = h = 12 = ... = f9 is the minimal exponent so that IF qI contains a
primitive pnth root of unity. Since IF qxI is made up of (q! -l)th roots of unity,
f is the minimal exponent such that pnlq! - 1. The ideal q) = Ker(7r)) for
each projection 7r) : Vq --» VqlqVq --» IFqIJ is a prime of Vq, and we have
II q)
9
(q) = and fg = [VqlqVq : IFq] = deg(ppn) =pn-l(p_l).
)=1
Summing up, we get

Proposition 2.5 Each prime q # p decomposes into a product (q) = TI~=l q]

of prime ideals q] with Vq/q] "" lFqf for the minimal exponent f with pnlqf -l.
We also have fg = pn-l(p - 1).
Let q = q]. Then Vq/qVq = lFqf. Since Gal(lFqf /lFq) is a cyclic group of
order f generated by cPq given by cPq(x) = x q, it is induced by an element cPq E
Gal(Ql[(pnJ/Ql) with Xpn(cPq) = q. In other words, the stabilizer of q is given
by Dq = (cPq) c G = Gal(Ql[(pn ]/Ql) generated by cPq. We have an injection
G/ Dq Y {ql,"" qg} by O'Dq r-t O'(q). Since IDql = IGal(lFqf /lFq) I = f, the
image has [Ql[(pn] : Ql]/ f = g elements; so, we get
We have proven:
Theorem 2.6 (Cyclotomic Reciprocity) If q # p is a prime, then for
each prime factor q in Z(q) [(pn L the decomposition subgroup of q in G =
Gal(Ql[(pn]/Ql): Dq = {O' E GIO'(q) = q} is given by the cyclic subgroup (cPq)
generated by the Frobenius element cPq with Xpn (cPq) = q. In particular, (q) =
TIcrEG/D q O'(q) and Vq/q "" lFqf for the order f of q in (Z/pnzy.
Let Vq be the localization of Vq at q. Since TI] q] = (q) in Vq, we find that

qVq = (q), which is a unique maximal ideal. Thus qm /qm+l "" Vq/q = lFqf is
a field; so, there is no ideal between qm and qm+l in V q . If C1 is an ideal of V q ,
C1 n Z(q) = qm; so, C1 = qm. In other words, all ideals of Vq are given by
Therefore Vq is a discrete valuation ring. We have Vq C nq Iq Vq. If %E Ql[(pn]

for a, b E Vq is in the intersection, then a is prime to q for all qlq; thus, a is
prime to q. This implies Vq = nqlq V q, and Vq is integrally closed (Exercise 3).
Since ( = (pn is defined by the relation
1 + (pn-l + (2 pn - + ... + ((p_l)pn-l = 0,

1
thepn-l(p-1) elements 1,(, ... ,(pn-l(p-l)-1 form a base of Z[(pn] over Z.

Then we have
n =n
Vq
q
Vq = (nqZ(q))l + (nqz(q))( + ... + (nqz(q))(pn-l(p-l)-l = Z[(pn],
where q runs over all prime ideals of 0 = Z[(pn] and q runs over all rational
primes. This shows that Z[(pn] is the integer ring of Ql[(pn]; in other words,
Z[(pn] is the integral closure of Z in Ql[(pn]. Thus Oq = Vq for a prime ideal q
of 0, and we can restate the reciprocity law as follows:
Corollary 2.7 The integer ring 0 of Ql[(pn ] is generated by (pn over Z. For a
prime q # p and a prime factor qIq, the decomposition group D q is independent
of the choice of q and generated by the Probenius element ¢q with Xpn (¢q) =
q. We have (q) = Il:rEG/D q tT(q) and G/q = lFqJ for the order f of q in
(71., / pn 71.,) x. The prime p is totally ramified in G, and the prime factor p of p is
principal and is generated by (pn - 1. In particular, Ip = Dp = Gal(Q[(pn ]/Q)
for the inertia subgroup Ip C Dp.
2.2.3 Adelic Reformulation
Recall that p and q are distinct primes, and put N = pmqn. Then we have a
commutative diagram
where Ip denotes the inertia group at p. Since p is unramified in Q[(qn]' we

find Ip n Iq = {I}. Since Ip = Gal(Q[(pmJ/Q) and Gal(Q[(NJ/Q) is an abelian
group (Corollary 2.3), we find that Gal(Q[(N]/Q) = Ip x Iq ~ (71.,/N71.,y by
XN. Repeating this process for N = TIp pe(p), we find
Theorem 2.8 The cyclotomic character XN induces an isomorphism
of groups. The decomposition group of a prime q f N is cyclic and gener-

ated by the Frobenius element ¢q with XN (¢q) = q E (Z/ NZ) x. For a prime
piN, writing N = pe N I with p f N I, the inertia group Ip of p is given by
x"i/((71.,/p e71.,Y), identifying (71.,/N71.,y = (71.,/p e71.,)X x (Z/NI71.,) X • The decom-
position group for piN is given by Ip x (¢p), where XNI(¢p) = P E (Z/NI71.,y·
By definition, if MIN, Q[(M] C Q[(N] because (M = ({JIM. Then it is
easy to check that the following diagram commutes:
restriction
Gal(Q[(N ]/Q) )
XN 1 (2.1)
(71.,/N71.,) X ------------+) (71.,/ M71.,) x .
(x mod N)t-+(x mod M)
Thus the composite (inside ij) of Q[(N] for all positive integers N is actually
a union QCYc = UN Q[(N].
Taking the projective limit, we have
(2.2)
We compute Z= ~ N (71.,/ N71.,). We recall that the p-adic integer ring 71.,p =
~n(71.,/pn71.,) is formed by the formal series Ln>o cnpn with 0 ~ C n ~ P - 1,
which is the valuation ring (and is the completion of Z(p) with respect to the
norm Ixlp = p-v(x)). Since any integer can be expanded as above by p-adic
expansion, Z C Zp, and this inclusion is compatible with Z(p) C Zp. Writing
the prime decomposition of N as N = I1 pl N pe(p), we find by the Chinese
remainder theorem
z= ~N(Z/NZ) = ~N II(Z/pe(p)z) = II~e(Z/pez) = IIZp, (2.3)

piN p p
where p runs over all prime numbers in the last two products. We put the
p-adic topology on Zp so that a system of neighborhoods of x E Zp is given
by {x + pnZp}n::::O. This makes Zp (resp. Z;) a compact profinite ring (resp.
group). We equip Z and ZX with the product topology of each component
(Zp and Z;). Then Z (resp. ZX) is a compact profinite ring (resp. group).
We now want to prove: Gal(Qcyc/Q) ~ AX /QxJR.~ = GLl(A(oo))/Q~ for
the adele ring A using the above expression of Gal(Qcyc /Q). The adele ring A
is a rather complicated ring, but we find it very useful later, and it contains
all arithmetic information of the field of rational numbers Q.
Let us recall the definition of A. We consider the product ring JR. x I1p Qp,
where p runs over all positive prime numbers, and Qp is the field of fractions
of Zp, and Q C Qp, which is the completion of Q under the p-adic norm I Ip
as above. Each p-adic number x E Qp has an expansion Ln»-oo cnpn (en = 0
if n is very negative). Since Z; is made up of L~=o cnpn with p t Co, Qp =
Un pnz; U {O}, where n runs over all integers. Regard Q inside JR. x I1p Qp
p
by sending ~ E Q to (~, C·· ,~, ... ) E JR. x I1p Qp. Let A be the subring of
JR. x I1p Qp generated by Q, Z, and R We often write x = (x oo , ... , x p , .•• )
for an element x E JR. x I1p Qp; so, Xoo E R
Proposition 2.9 We have
and A = (JR. x Z) + Q. Here "almost all p" means "except for finitely many
p."
Proof. The ring A' is a subring of JR. x I1p Qp- For ~ = ~ E Q (a, b E Z), if p t b,
we have ~ E Zp. Thus ~ E A', and A C A'. If x E A', we expand xp = Ln cnpn,
and define [xp] = Ln<o cnpn, which is a fraction with a p-power denominator
and is called the fraction p-part of xp. Then [xp] = 0 for almost all p, and
[x] = Lp[Xp] E Q. Then we look into xp - [x] = xp - [xp] - L#p[Xq] E Qp.
Since the denominator of [Xq] is prime to p, we find [Xq] E Zp. Thus xp - [x] =
xp - [xp] - Lqh[xq] E Zp, and hence x - [x] E Z inside A'. In particular,
x = (x - [xl) + [x] E (JR. x Z) + Q c A. This shows the last identity. 0
Writing the finite part A(oo) = A n TIp Qp, we thus have A(oo) = Z 0z Q.
Proposition 2.10 We have AX = QX(~~ x ZX) by x f--i (x)(x/(x)), where

~~ = {x E ~Ix > O}.
Proof. Since Qp = Unpnz; U{O}, we can write xp E Qp as xp = pv(xp)up with
up E Z;. Then for x E AX, X-I E A, and hence IXpip = 1 ..;=} v(xp) = 0
for almost all p. In other words, (x) = I~: I TIp pv(x p) is a rational number for
x E AX, and x(x)-1 E ~~ x Zx. 0
By translation, we extend the topology on Z to A(oo) = Q + Z. Therefore

A and A (00) are locally compact rings (and A X is a locally compact group;
see [MFG] 3.1, [LFE] 8.1-3 and [EPE] Chapter II for more about the adele
topology on the adele points G(A) of an algebraic group Gover Q).
Corollary 2.11 We have A x /Qx~~ ~ Zx. This isomorphism [ can be nor-
p ~
malized so that it takes p E Qp c A to p(poo) = (p,p, ... ,p, l,p, ... ,p) E Zx.
Proof. We see that ~ E QX n (ZX x ~~) means that ~ = ~oo > 0 and the
numerator and the denominator of ~ = ~p is prime to p for all p; so, ~ = l.
Thus we conclude QX n (ZX x ~~) = {I}. By the isomorphism theorem:
¢ : AX /Qx~~ = ZXQx~~/Qx~~ ~ ZX /Qx n (ZX x ~~) = Zx.
Since p = p(poo)ppPoo = 1 in AX /Qx~~ for Pp = (1, ... , 1,P, 1, ... ,1) E AX

00
and Poo = (p, 1, ... , 1) E A x, the above isomorphism brings Pp in Q; C A x
to (p(POO))-I. Defining [(x) = ¢(X)-I, we get the desired isomorphism. 0
Combining this with Theorem 2.8, we get

Theorem 2.12 (Class Field Theory) We have a canonical (reciprocity)
isomorphism [ : Gh(A(oo))/Q~ = AX /Qx~~ ~ Gal(QCyc/Q) such that
Pp E Qp is sent to a Frobenius element at p, and the decomposition group
at p is given by the closure of the image of Q;.
Inside the decomposition
group, the inertia group is given by the isomorphic image of z; .
Class field theory says slightly more, namely that QCYc is the maximal abelian
extension Qab of Q. For more on this, see [FAN] Chapter 6. Often the above
result is expressed as an exact sequence,
and the map [(x) is called the Artin reciprocity map and sometimes written
as [(x) = [x, Q]. The reciprocity map of the class field theory induces a lo-
cally compact (actually compact profinite) topology on the Galois group. This
topology can be defined in a way intrinsic to Galois' theory, as we show in the
following section.
2.3 A Generalization of Galois Theory 31
Exercises
1. Prove that !J> pn (X) is irred uci ble in Q[X].

2. Give a detailed proof of Corollary 2.4.
3. Prove Z(q) [(pn] is integrally closed.
4. Prove that the integer ring 0 of Q[(N] is generated by (N over Z. Hint:
First show that Q[(N] = Q[(p= ]0Q1 Q[(qn] if N = pmqn for distinct primes
p and q. Then show that 0 = Z[(p=] 02 Z[(qn] using the fact that no
prime ramifies in the two fields Q[(p=] and Q[(qn] at the same time.
5. Give a detailed proof of Theorem 2.8.
6. Deduce the quadratic reciprocity law in the introduction either from The-
orem 2.6 or Theorem 2.12.
2.3 A Generalization of Galois Theory
For our later use, we gather here some results from a generalization of Galois'
theory given in [IAT] 6.3.
2.3.1 Infinite Galois Extensions
We study the Galois theory when dimK L is infinite for a Galois extension
L/ K. First we recall Galois' fundamental theorem when L/ K is finite.
Theorem 2.13 (E. Galois) Suppose that L / K is a finite Galois extension.
Taking an intermediate field L / M / K, L / M is a Galois extension and
(1) There is one- to- one onto correspondence
{M: intermediate fieldIL/M/K} +-+ {H: subgroup of Gal(L/K)}
given by M f--t Gal(L/M) and H f--t LH = {x E L!x = (J(x)\j(J E H}.

(2) For two intermediate fields M, M', we have
• M:::> M' <==} Gal(L/M) c Gal(L/M');
• Gal(M n M') is the subgroup of Gal(L/ K) generated by the two sub-
groups Gal(L/M) and Gal(L/M');
• Gal(L/M M') = Gal(L/M) n Gal(L/M');
• For (J E Gal(L/ K), Gal(L/(J(M)) = (JGal(L/M)(J-l.
(3) M/K is a Galois extension if and only if Gal(L/M) is a normal subgroup
in the Galois group Gal(L/K). In particular, by (J f--t (JIM, we have an
isomorphism: Gal(L/ K)/Gal(L/ M) So' Gal(M/ K).
We admit this theorem, just referring the reader to any book on elementary
Galois theory.
What we want to do in this subsection is to generalize the above theorem
to an infinite Galois extension. Hereafter we suppose that L/ K is an infinite
Galois extension. Since each element ~ E L satisfies a polynomial equation of
finite degree with coefficients in K, there are only finitely many conjugates of
~. Thus the Galois closure K[~]gal of K[~] over K is a finite Galois extension
of K, and we get from Theorem 2.13 (2),
L = U K[~Fal {:} nGal(L/ K[~Fal) = {id}. (2.4)

~EL ~EL
We first generalize the third assertion of the above theorem.

Lemma 2.14 Suppose M is an intermediate Galois extension of a Ga-
lois extension L / K. Then the group Gal( L / M) is a normal subgroup of
Gal(L/K), and (J r-)- (JIM induces an isomorphism Gal(L/K)/Gal(L/M) ~
Gal( M / K). In particular, if M / K is a finite extension, then we have the
equality [Gal(L/K) : Gal(L/M)] = [M: K] < 00.
Proof. If M / K is a finite extension, for any finite Galois extension L' with
L :J L' :J M, we can extend (J E Gal(M/K) to T E Gal(L'/K) (assertion
(3) of Theorem 2.13). Since L is a union of finite Galois extensions of L',
(J E Gal(L' / K) extends to T E Gal(L/ K). If M / K is an infinite extension,
writing M as a union of finite Galois extensions M' / K and applying the
above fact to M' / K, we find that any (J E Gal( M / K) can be extended to
T E Gal(L/ K). Thus (J r-)- (JIM is a surjective homomorphism of Gal(L/ K)
onto Gal(M/K). Its kernel is given by Gal(L/M) which is normal. 0
Basically the same proof as above yields the following result which looks
stronger than the lemma.
Corollary 2.15 Let M be an intermediate extension of L/ K. Then (J r-)- (JIM
induces an isomorphism Gal(L/K)/Gal(L/M) ~ HomK(M, L), where the set
HomK (M, L) is the collection of K -linear field homomorphisms of Minto L.
The proof of this corollary is left to the reader as an exercise (Exercise 1).
We now give a topology on the group Gal(L/ K) in the following way.
Define a fundamental system of open neighborhoods of the identity by the
collection of subgroups of the form Gal(L/M) for a finite extension M/ K. If
H = Gal(L/M) for a finite extension M/K, the Galois closure Mgal is still a
finite Galois extension of K; so, we find a normal open subgroup Gal(L/Mgal)
inside any open subgroup H. Thus we may define the system to be the set
U of all normal subgroups N = Gal( L / M) for finite Galois extensions M / K
inside L. For each (J. E Gal(L/ K), we define the system of neighborhoods of
(J to be (JU = tiC! = {(IN = N(JIN E U}. By this, if II satisfies the axiom of a
fundamental system of open neighborhoods of a point, G becomes a topolog-
ical group, that is, G is a group with a topology for which the multiplication
(a, b) r-)- ab and inverse a r-)- a- 1 are continuous. The axiom is checked by
Lemma 2.16 We have
(1) For (J i- 1, there exists N E U with (J (j. N (so, G is a Hausdorff group);
2.3 A Generalization of Galois Theory 33
(2) N, H E 11 =} N nH E 11.
Proof. By (2.4), L = UM M for finite Galois extensions M/K. If a E

nM Gal(L/M) , a is the identity over all M; so, a = 1 in Gal(L/ K). Thus there
rt
exists a finite Galois extension M / K inside L such that a N = Gal( L / M)
if a =I=- 1. By definition, we have N E 11; so, assertion (1) follows.
Write N = Gal(L/M) and H = Gal(L/M'). Then MM' is a composite of
two finite Galois extensions; so, it is a finite Galois extension. By definition, a
is trivial over MM' ~ a is trivial over M and M'. Then by Lemma 2.14,
we have Gal(L/MM') = NnH, and hence NnH E 11, which proves (2). 0
The topology we have defined was first introduced by Krull, and therefore it
is called the Krull topology on Gal(L/ K).
Proposition 2.17 For an infinite Galois extension L/ K, Gal(L/ K) is a
compact group under the Krull topology.
Proof. Since Gal(L/ K) is Hausdorff, we need to show that for any given infi-
nite subset E in Gal(L/ K), there exists an element a E Gal(L/ K) such that
En H a is infinite for all H E 11; that is, E has a limit point a E Gal(L/ K).
Since H = Gal( L / M) for a finite Galois extension M / K, there are finitely
many automorphisms of M / K. Thus there exists an infinite subset E H of E
that induces a single automorphism aM on M. For any extension a' E EH of
aM to L, the coset Ha' is the collection of T E Gal(L/K) (by Lemma 2.14)
with TIM = aM. We make the construction of aM compatible with an in-
creasing sequence of Galois extensions M j of K inside L. In other words,
we find a sequence of finite Galois extensions M j / K (j = 1,2, ... ) with
K C M j C MJ+l eLand aj = aM, E Gal(Mj/K) such that L = Uj M j and
aj+llM, = aj for all j. We have found an element a E Gal(L/K) such that
aiM, = aM, for all j. This a is a limit point of E. 0
Corollary 2.18 If M is an intermediate field of an infinite Galois extension

L/K, Gal(L/M) is a closed subgroup of Gal(L/K).
Proof. For any finite Galois extension M' / K inside L, we have by definition
the following identity: i- 1 (Gal(L/MM')) = Gal(L/MM') for the inclusion
map i : Gal(L/M) Y Gal(L/ K). For any N E 11, we find a finite Galois
extension M' / K inside L such that N = Gal( L / M'). Since M M' / M is a
finite Galois extension, i-l(N) = Gal(L/MM') is open in Gal(L/M); so, i is
continuous. The image of a compact set by a continuous map is compact; in
particular, it is closed. 0
We now prove a generalization of Theorem 2.13.
Theorem 2.19 Let L/ K be an infinite Galois extension. We have the fol-
lowing canonical one-to-one onto correspondence,
{ intermediate fields M of L / K} ~ {closed subgroups H of Gal( L / K)}
induced by M I---t Gal(L/M) and H I---t LH. Moreover,
(1) Every open subgroup of Gal(L/ K) is closed, and open subgroups corre-
spond to finite extensions LH / K.
(2) Let {Nt hEJ for an index set I be a collection of closed subgroups of the
Galois group Gal(L/K). Writing Nt = Gal(L/Kt ), for the composite M
n
of all {KdtEJ, we have tEJ Nt = Gal(L/M).
(3) If {KthEJ is a collection of intermediate fields of L/ K, then the group
Gal(L/ ntEJ K t ) is the closure of the subgroup in Gal(L/ K) generated by
Gal(L/ K t ) for all i.
(4) For 0- E Gal(L/K), we have Gal(L/o-(M)) = 0-. Gal(L/M)o--l.
(5) If M/K is a Galois extension inside L, then Gal(L/M) is a normal closed
subgroup and Gal(L/K)/Gal(L/M) ~ Gal(M/K) induced by 0- f-t o-IM is
an isomorphism of topological groups.
Proof. Let 9J1 be the set of all intermediate fields of L/ K and Sj be the set
of all closed subgroups of Gal(L/ K). For any M E 9J1, Gal(L/M) is a closed
subgroup of Gal(L/K) by Corollary 2.18; so, M f-t Gal(L/M) defines a map
9J1 -t Sj. We define a reverse map Sj -t 9J1 by H f-t LH.
We first prove that Gal(L/ LH) = H. Since Gal(L/ LH) C Gal(L/ K)
is the collection of all automorphisms of L fixing LH, we confirm that
H C Gal( L / LH). Conversely, for each finite Galois extension M / LH inside
L, H acts on M nontrivially if M -=J LH, and the image under 0- f-t o-IM
in Gal(M/L H ) is a subgroup H' of Gal(M/L H ) and LH = MH'. By Theo-
rem 2.13 (1), H' = Gal(M/ L H ). In other words, for each 0- E Gal(L/ LH),
we find hM E H such that hMIM = o-IM. We consider the infinite set
E = {h M } M C H with M running through all finite Galois extensions M / LH.
Since H is a closed subgroup of a compact group Gal(L/ K), H itself is com-
pact. Then E has a unique accumulation point h. Then for each finite Galois
extension M/LH, o-IM = hiM. Since L = UMM for finite Galois extensions
of M/LH, we find that h = 0- and H = Gal(L/LH).
We now prove LGal(L/M) = M. By Corollary 2.18, H = Gal(L/M) E Sj.
By definition, M C LH. Supposing that LH -=J M, we try to get a con-
tradiction. Pick ~ E LH - M. Then M[~l/M is a finite extension. Thus
M[~lgal C L because L/K is normal. Since ~ tf. M, Gal(M[~lgal/M) -=J {I}.
Pick 0- E Gal(M[~lgal /M) with o-(~) -=J ~. Then by Lemma 2.14, we find
7 E Gal(L/M) = H such that 7(0 = o-(~) -=J ~. This is wrong since o-(~)
has to be ~ because ~ E LH. Thus we find that LGa1(L/M) = M, and the
correspondence is one-to-one and onto.
For an open subgroup H, Gal(L/ K)/ H is discrete and compact; so,
it is finite (see Exercise 4). Since H is the kernel of the continuous map
Gal(L/ K) --;; Gal(L/ K)/ H, H is closed. By H = Gal(L/ LH), we see from
Lemma 2.14 and Corollary 2.15 that Gal(L/K)/H = HomK(LH,L), and
hence LH / K is a finite extension by Corollary 2.15. This shows (1).
Assertion (2) follows from the fact: 0- E Gal(L/ K) is the identity on every
K t if and only if 0- is the identity on the composite of all Ki (Exercise 2).
2.3 A Gellerali;oation of Galois Theory 35
Let Gal(L/ n'EI K,) = H'. Then by definition, H' is a closed subgroup
containing all H, = Gal(L/ K,), and it contains the closure H of the subgroup
generated by H, for all i. Since H is a closed subgroup (so in Sj) containing H"
we see LH C K, and hence LIi c n'EI K,. Thus H ~ H' = Gal(L/ n'EI Ki).
This shows H = H'.
Assertion (4) is plain from the definition of the Galois group, and the last
assertion follows from Lemma 2.14. 0
2.3.2 Automorphism Group of a Field
We study Aut(L/ K) for an arbitrary extension L/ K of fields. An interme-

diate extension M of L/ K is called of finite type over K if there are finitely
many elements Xl, ... ,X n that generate !vI as a field over K. We define a
fundamental system it of open neighborhoods of 1 on Aut(L/ K) by
it = {H = Aut(L/M)IM/ K is of finite type}. (2.5)
Since a composite of two intermediate fields of finite type is again of finite

type, it satisfies the axiom of the fundamental system of neighborhoods (see
the proof of Lemma 2.16 and Exercise 3). Then define a system of neigh-
borhoods of IJ E Aut(L/ K) by lJit U itlJ. This topology gives a topological
group structure on Aut(L/ K). Let Sj be the set of all compact subgroups of
Aut(L/ K), and let SJJ1 be the set of all intermediate fields M in L/ K such
that L / M is a Galois extension. Then we have
Proposition 2.20 (Jacobson) We have a canonical bijective correspon-
dence SJJ1 ~ Sj given by M>--+ Gal(L/M) and H >--+ LH.
Proof. Pick a compact subgroup H of Aut(L/ K). Then for each ~ E L, the
topological group Aut(L/K(O) is an open subgroup of Aut(L/K). Thus H' =
H n Gal(L/K(~)) is an open subgroup of H. In particular, H/H' is compact
and discrete; so, it is finite (Exercise 4). Since f(X) = TIhEH/W(X -h(O) has
coefficients in L H, ~ is algebraic over LH. Since h(~) = h/(O <=? hH' = hIH',
we find that the roots of f(X) are all distinct; so LH (0/ LH is a separable
extension. Since h(O E L for all h E H, all the roots of f(X) are in L. The
Galois closure M of LH (0 over LH is a Galois extension of LH inside L.
Since L = U~EL LH (0, we find that L/ LH is a Galois extension. Then by
Theorem 2.19, we find that H = Gal(L/ L H ). Starting from M E SJJ1, we find
that Gal(L/M) E Sj and M = LGaJ(L/M) by Theorem 2.19. 0
A field extension L/ K is of finite type if L is generated over K by finitely

many elements (as a field). Let Sjo be the set of all open compact subgroups
of Aut(L/ K) and SJJ10 be the subset of SJJ1 made up of fields M that are of
finite type over K.
Corollary 2.21 IfSJJ1° is nonempty, then Aut(L/K) is locally compact, and
the one-to-one onto correspondence in Proposition 2.20 induces SJJ10 ~ Sjo.
Proof If 9)10 is nonempty, pick M E 9)10. Then the identity of Aut(L/ K)

has an open neighborhood Gal(L/M), which is also compact by Proposi-
tion 2.17. By translating this neighborhood to any point (J on Aut(L/K)
through multiplication by (J, every point has an open-compact neighborhood;
so, Aut(L/ K) is locally compact. We need to show that LH / K is of finite type
over K if H E 5)0. Consider F = M LH. Then L / F is a Galois extension; so,
H' = Gal(L/F) is a compact subgroup of Gal(L/M). Since F/M is a finite
extension, F is of finite type over M and hence over K. Thus the subfield LH
of a field F is of finite type over K. 0
Exercises
1. Prove Corollary 2.15.

2. Give a detailed proof of the assertions (2) and (4) of Theorem 2.19.
3. Prove that the Krull topology on Aut(L/ K) given by (2.5) is well-defined.
4. Prove that a compact and discrete set is a finite set.
2.4 Algebraic Curves over a Field
Here we give an exposition of classical theory of algebraic curves over a field

from the viewpoint of the theory of discrete valuation. See [ALF] for a more
detailed exposition on algebraic curves from this point of view. Under this
setting, in the following section, we give a prototype of classification problems
of abelian varieties (description of the moduli of elliptic curves over afield).
To treat classification problems over rings, a more sophisticated language, for
example, the language of schemes, is necessary. From the next chapter on, we
treat the general case via the theory of schemes.
2.4.1 Algebraic Function Fields
Let K be a field of an arbitrary characteristic. An algebraic function field j{

of dimension 1 is a finitely generated nonalgebraic field extension j{/ K such
that for any x E j{ transcendental over K, j{/K(x) is algebraic. We assume
that the algebraic closure of K in j{ is K itself. In this case, we say that j{
is defined over K. Replacing K by its algebraic closure in j{, we may always
assume that j{ is defined over K.
We relate field theory to the geometric theory of projective algebraic curves
by considering local coordinates. A prototypical example of algebraic function
fields is given by the meromorphic function field over C of a compact Riemann
surface. Each point of a Riemann surface has a coordinate neighborhood, and
the collection of all coordinate neighborhoods by definition determines the
Riemann surface. Each meromorphic function has Laurent expansion at a
given point, which gives a well-defined order of vanishing (or order of pole) of
2.4 Algebraic Curves over a Field 37
the function at the given point. Associating the order at a point with mero-
morphic functions gives a valuation of the function field specific to the point.
An idea of how to algebraize a Riemann surface is to consider the set of all val-
uations of its function field trivial over the base field and to replace coordinate
neighborhoods by corresponding valuations. We recall here a formal definition
of discrete valuation rings (DVR) inside an algebraic function field .it over K.
A discrete valuation v of .it trivial on K is a surjective map v : .it -+ Z U {oo}
satisfying the following four conditions:
(VO) v(KX) = 0 (triviality over the base field);
(V1) v(f) = 00 B j = 0;
(V2) v(f + g) ?: min(v(f), v(g)) for all j, g E .it;
(V3) v(f g) = v(f) + v(g) for all j, g E .it.
Here we agree as a convention to have a + 00 = 00 and 00 > a for all a E Z.
By the above properties, Vv = {j E .itlv(f) ?: O} is a subring of .it, and either
x E .it or ~ belongs to V v . In particular, the field of fractions of Vv is equal to
.it. The ring Vv is called a discrete valuation ring, and m = {j E Vv Iv(f) ?: I}
is a unique maximal ideal of Vv (so, Vv is local). The field Vv/m is called the
residue field of the valuation v and is a finite extension of the base field K.
Moreover, every ideal of Vv (except for the zero ideal) is a power of m, and
mn = {j E .itlv(f) ?: n}. See [CRT] Chapter 4 for more about valuation rings.
Our goal in this subsection is to create a space (a geometric object) from
the purely algebraic notion of algebraic function fields .it. The object is called
the Zariski-Riemann space and is the collection of all valuations of .it trivial
over the base field K. The space is an algebraic replacement of the associated
Riemann surface for K = C.
Example 2.22 We start with the simplest Riemann surface: the Riemann
sphere P = P 1 = C u { 00 }. The meromorphic function field of the sphere P is
isomorphic to the rational function field .it = C( x ). Then the polynomial ring
qx] corresponds to the Riemann sphere P with coordinate x in the following
sense. The space P is covered by two coordinate neighborhoods Uo and U=
identical to C, Uo is centered at 0 with complex coordinate x, and the other
U= is centered at 00 with coordinate Xl = ~. Then the field of meromorphic
functions of P is given by C(x), and the polynomial ring qx] is the ring of
functions with the only possible pole at 00. Each meromorphic function ¢ on
Uo finite at 0 has its Taylor expansion in the coordinate x: ¢(x) = ~~=o anx n
whose radius of convergence is positive. If ¢ is not finite at 0, for some pos-
itive m, xm¢(x) is finite at 0, and ¢(x) therefore has its Laurent expansion
~n>_manxn. We can define a valuation Vo on C(x) by assigning the expo-
nent of the leading term of the Laurent expansion to a given meromorphic
function ¢ E C(x). The valuation vo(¢) is just the zero order of the function
¢ holomorphic at O. For any other point a E C = Uo (resp. a = 00 E U oo ),
we can take ta = x - a (resp. too = x- 1 ) as a coordinate around a, and we
can think of the valuation Va giving the order of zero at a. In differential ge-
ometry or in complex analysis, local behavior of functions reflected by Taylor
expansion often determines in the aggregate global properties of the function.

Partial fraction expansion determines elements in C(x); so, local knowledge of
the valuation va.(¢) almost determines C(x). Thus it is natural to expect that
the set Z of valuations {va.}a. determine P. Indeed, at least set-theoretically,
Z is in bijection with P.
We choose a transcendental element x E Jt so that Jt is a finitely generated

separable algebraic extension of K(x). Then Jt/ K(x) is a simple extension;
in other words, we can find a single generator y E Jt such that Jt = K (x) [y].
We may assume that y is integral over K[x] (because some multiple of y
by an element in K[x] is integral over K[x]) , and y satisfies an equation
ym + al(x)ym-l + ... + am(x) = O.
Suppose that K = C. Take a general compact Riemann surface R. Then
on the given Riemann surface R, choosing two generators (x, y) in its mero-
morphic function field, the relation of (x, y) gives a projective algebraic curve
V in p2. Moreover, defining the algebraic function field C(V) of V by the
collection of nontrivial restriction to V of rational functions P(x, y)/Q(x, y)
(with two polynomials P and Q of equal degree), the meromorphic function
field of R is isomorphic to the algebraic function field C(V) of V.
Here is another slightly more nontrivial example.
Example 2.23 Consider the function field Jt = C(x)[y] defined by the equa-
tion y2 = x(x - l)(x - A) for A E C different from 0 and 1. We consider the
square root y(x) = )x(x - l)(x - A) on Uo. This function has two values
on P vanishing at 0,1, A and has a pole at 00; in other words, its inverse
vanishes at 00. Thus the locus of the point (x, y, 1) E p 2 is a two-sheet cov-
ering of P that ramifies at 0,1, A, 00. Around x = 0, if a point circles around
(x, y) = (0,0), for two values of ±y, we have one value of x. In other words,
cutting a line segment [0,1] and [A, 00] from two copies of P and gluing the
two corresponding segments, we get the donut-shape Riemann surface R, on
which the functions y and x both have single values. The field of meromorphic
functions over R is given by C(x)[y]. Moreover, we can embed R into p 2 by
PH (x(P), y(P)), which satisfies the given equation y2 = x(x -l)(x - A). In
other words,
R ~ V = {(x,y) E p 2[y2 = x(x -l)(x - A)},

because x moves around all possible values in P = C U {oo}.
For each point a E C different from 0,1, A, pick a point P from the two
points P, Q of Rover a. Then P has an open neighborhood U with coordinate
t = x-a. Any meromorphic function f defined over U has a Laurent expansion
f(t) = Ln»_=ant n . We define a valuation Vp: C(V) ---+ Zu {oo} given by
v p (f) = m if am 1= 0 and an = 0 if n < m.
Thus we have (VO-3) for vp and

Vp = {J E .RIVp(X) :;;. o} (2.6)
is a valuation ring (Exercise 1) with vp(C X ) = 0 (i.e., Vp is trivial on q.

For P over one of the four ramified points a = 0,1, A, 00, for an open
neighborhood 0 of a that does not contain any of 0,1, A, 00 different from
a, x-I(O) = U is an open neighborhood of P and y gives a local coordinate
t if a =I 00 and y-I gives a local coordinate t of U if a = 00. Then we
define Vp exactly in the same way. By definition, for each x E .R, either
vp(x) :;;. 0 or vp(x) < 0; in other words, x or X-I is in V p . Thus we have
.R = Fmc(Vp ) = H la E Vp, b E Vp ~ {O}}.
Start conversely with a discrete valuation v:.R --+ Z U {oo} with v(C X ) =
O. Then V = {f E .Rlv(f) :;;. O} is a discrete valuation ring. Again for each
x E .R, either x E V or X-I E V. First suppose that x E: V. Then qx] c V.
Let m = {j E Vlv(f) > O}. Then m is the maximal ideal of V, and V 1m
is a field extension of cc. Since qx] is a principal ideal domain, m n qx] is
either (x ~ a) for a E C or (0). If m n qx] = (0), we find that V 1m contains
an isomorphic image of qx]. Thus the transcendental degree of V1m, (i.e.,
dim(V 1m)) is larger than or equal to 1. Since V has Krull dimension 1 (cf.
[CRT] Section 5), this implies m = (0), which is impossible. We find that
m n qx] = (x - a). This implies that V 1m = C and y mod m = {3 which
satisfies (32 = a(a ~ l)(a ~ A). In other words, taking the point P with
coordinate (a,{3) E V, we find V = Vp.
If x tf. V, we find Zl = ~ E V. Then we see easily that V = VeX). We have
found the following fact:
R"'" {VIV is a DVR with a valuation trivial on C} .

This is an algebraic interpretation by Oscar Zariski of a Riemann surface as
a space of all valuations trivial on C, and the space at the right-hand side of
the above identity is called a Zariski-Riemann space. It is intriguing that all
rational primes correspond to discrete valuations of Qi, and we might want to
think that Qi is an arithmetic analogue of a Riemann surface.
For a smooth projective curve V defined over a field K, we write K(V)

for its field of K-rational meromorphic functions, and V(M) denotes the set
of M -rational points of the curve V for an extension M I K. For a point P
of V algebraic over K, we write K(P) for the field generated over K by the
coordinates of P.
Theorem 2.24 For a given algebraic function field .R defined over a field
K, there exists a unique smooth projective curve V defined over K such that
.R "'" K(V). We also have a canonical one-to-one onto correspondence:
V(K) "'" {discrete valuations of.R trivial over K X with residue field K}.
If .R' I.R is an extension of algebraic function fields defined over K, the corre-
sponding smooth projective algebraic curve gives rise to a covering 7r : VI --+ V
such that f M f 0 7r gives the inclusion .R Y .R' . In particular Vp n.R = Vrr(P)

for the valuation ring Vp corresponding to P E V'.
Proof. We generalize the argument (in the two examples) of making a pro-
jective algebraic curve out of valuations of .R for a general algebraic func-
tion field over a field K. We start with K(p l ) = K(x) with one transcen-
dental element x. The main tool is the fact that K[x] is a principal ideal
domain. Pick a nonzero prime ideal P of K[x]. Pick f E P, and factor-
ize f into a product f = c TIt t(xy(t) of irreducible monic polynomials t(x)
with a constant C E KX. Since P is a prime, one of the irreducible fac-
tors t(x) has to be in P. Then K(P) = K[xl/P is covered by K[xl/(t(x)),
which is a finite extension of K. We get a surjective field homomorphism
7r: K[xl/(t(x)) -7> K(P) = K[xl/P. Since a field homomorphism is always in-
jective (a field only has two ideals: (0) or itself), we find K(P) = K[x]/(t(x))
and P = (t(x)). So any nonzero prime ideal is a maximal ideal. We define the
valuation vp : K(x) -+ Z U {oo} by vp(f) = vp(c TIt t(x)e(t)) = e(t). Then
the valuation ring Vp = {J E K(x)lvp(f) ~ O} is associated with the point
P by mp n K[x] = P for the maximal ideal mp = {J E Vplvp(f) > O}.
We study Vp/mp 9:"! K[xl/pn, because mt/mt+l = (K[xl/P)t(x)i =
pi / pJ+1. For a given f E K[x], by the division algorithm, we find a quotient
q E K[x] and a remainder r such that f = tq + r with deg(r) < deg(t(x)); so,
we may regard r as a unique element in K(P). We again apply the division
algorithm to q, getting q = tql + rl; in other words, f = r + rlt + t 2ql. Re-
peating this process, we can expand f into: f = r+rlt+r2t2+r3t3+ .... This
appears to be an infinite series, but is actually a finite sum by comparing the
degrees of the both sides. We write Vp = K(P)[[t]], which is a formal power
series ring, and embed Vp into Vp by the above expansion; so, the image of
Vp is made up of polynomials in t with coefficients in K(P). If one introduces
the projective limit of Vp/mp , we find Vp = ~nVp/mp.
For a given valuation ring V = {J E K(x)lv(f) ~ O} with v(KX) = 0 (but
v(K(x)) = ZU {oo}), we find either x E V or x fj. V. If x E V, then K[x] c V;
so, v is a valuation on K[x]. Since v cannot be trivial over K[x] (otherwise,
it is trivial on K(x)), the intersection mn K[x] is a prime ideal P of K[x].
Then v = Vp because v(f) ~ m ¢=:} f E mm n K[x] = pm because P is
principal. If x fj. V, then X-l E V, and for the prime Q = (x- l ) in K[x- l ], we
have K[x-l]Q = V, and V corresponds to the point 00 = Q.
If K is algebraically closed, each nonzero prime ideal of K[x] is the form
of (x - ex) with ex E K, which tells us
{Vlvaluation rings of K(x) trivial on K} 9:"! K U {oo} = P(K) (canonically).
If K is not algebraically closed, take an algebraic closure K of K. Defining

P(K) by the set of points in P(K) with coordinates in K, we have
{Vlvaluation rings of K(x) trivial on K with K(P) = K} 9:"! P(K).

The set {Vlvaluation rings of K(x) trivial on K} is called the set of closed
points of P. They are associated with a maximal ideal of K[x] or K[X-l].
Of course, if V is associated with (t(x)) and t(x) -=I- x, then the same V is
associated with x-deg(t)t(x) E K[x-l] because vp(x) = 0 if t(x) -=I- x. Thus
we find
{closed points of PI K }
= {maximal ideals of K[x]} U {maximal ideals of K[x-l]}.
We now treat a general case of algebraic function fields not necessarily

a rational function field K(x). For a given valuation v, the residue field of v
means the residue field of the valuation ring of v. Let ~ be an algebraic function
field defined over K, and choose elements x and Y so that ~ = K(x, y) and
y is separably integral over K[x]. Let R be the integral closure of K[x] in K
Since K[x] is a principal ideal domain and the quotient field of R (i.e., Jt) is
a finite extension of K(x), we find that R is free of rank [Jt: K(x)] over K[x].
Since R is a finite integrally closed extension of the principal ideal domain
K[x], R is a Dedekind domain ([CRT] Section 11). So each nonzero fractional
R-ideal a has a prime decomposition a = TIp peeP) into a product of prime
ideals P (e(P) E Z). With each prime ideal of R, we can associate a valuation
Vp : ~ --+ Z U {oo} by vp(f) = e(P) if (f) = f R = TIp pe(P). For any given
valuation ring V associated with a valuation v : ~ --+ Zu {oo} with v(KX) = 0
but v(~X) = Z, we find x E ~ transcendental over K such that x E V. Thus
all valuations v as above are of the form Vp for a suitable choice of x and a
prime ideal P of an integral closure R of K[x]. Thus we may identify
_ { . .
V(K) -
. Iv(K X ) = 0 and v(~X)
v.dlscrete valuatIOn with residue field K
= Z} .
Any prime ideal P of K[x] is decomposed into a product of primes P l , ... ,Pg
in R. If K is algebraically closed, then there are 9 distinct points over a given
point P = (x - a) of P for a E K. If K is not algebraically closed, K(P,)
for some i may be a nontrivial finite extension of K. In other words, in the
algebraic closure K of K, we have at most [K(Pi ) : K] (geometric) points
coming out of P, corresponding to embeddings of K(Pi ) into K.
If furthermore, R = K[x, y] for another element y E ~, we have an equation
f(X, Y) = yg + al (X)yg-l + ... + ag(X) satisfied by (x, y). So for a given
point P = (x - a) E P(K), the point (a, f3) E K2 satisfying f(a, f3) = 0 gives
rise to a point P, over P in p2. In this way, P, f-t (a, f3) supplies us with
a projective embedding of V(K) into p 2(K). The image is the projective
algebraic curve defined by the equation f(X, Y) = 0 in p 2 (or more precisely,
using the homogeneous coordinate (X, Y, Z), it is defined by zg f( -i,~) = 0).
Thus except for finitely many points, there are really 9 distinct points in V(K)
over a given point P E P(K). More generally, if R = K[x, Yl,"" Ym], then
P f-t (x, Yl,'" ,Ym) gives an embedding of V into the (m + I)-dimensional
projective space pm+l.
If we take a proper subring K[x, y] <;; R such that .it = K(x, y), the ring
K[x, y] is not integrally closed. In other words, for P E V(K), x and y generate
a sub ring V~ of the valuation ring. Since mp n K[x] = (t(x)) for an irreducible
polynomial t(x), mp C V~ for some e:::; g (since pg C (t(x))Vp).
Let J(X, Y) be the equation of (x, y). We can think of the projective
algebraic curve C C p2 defined by this equation. Then if K is algebraically
closed, a = (x mod P) E K(P) = K and (3 = (y mod P) E K gives a point
(a,(3) of C in p2. Even if V does not ramify, in C(K), P' = (a,fJ) has extra
ramification, and V~ is not a valuation ring.
We have a natural morphism V --+ C --+ P taking P f--t (a, (3) f--t a. In this
sense, V is the largest projective algebraic curve (over P with coordinate x)
giving rise to the algebraic function field .it.
For a projective algebraic curve C, we consider its function field K( C) and
V~ made up of functions in K(C) finite at P E C. If V~ is a valuation ring, we
call P a smooth point. The projective algebraic curve C giving rise to a given
algebraic function field .it is called a model of .it. Among models of .it, there is a
unique model smooth everywhere, which is called the smooth or non-singular
model of .it. If .it' /.it is a finite extension, by the above construction, we have
a covering map V' --+ V as defined in the theorem. 0
For each point P E V(K) for V as in the theorem, J E .it has an expansion
J(t) = I:n»-oo ant n with an E K(P) for a generator t of PVp . An element
t E .it giving rise to the generator t of P is called a uniJormizer at P.
Corollary 2.25 Let the notation and assumption be as in the theorem. Then
the morphism n : V' --+ V is a polynomial map oj the projective coordinates.
Proof. Let R (resp. R') be the integral closure of K[x] in .it (resp . .it').
Choose generators so that R = K[X,Yl, ... ,Ym] and R' = K[x,y~, ... ,y~].
Since R c R', we find x = Jo(x,y~, ... ,y~) and y, = J,(x,y~, ... ,y~). Then
n((x, y~, ... , y~)) = (x, Yl, ... , Ym), as desired. 0
Since any field embedding a : .it --+ .it' brings .it into a subfield a (.it) ,
we have a morphism n : V' --+ Vo- of projective algebraic curves V' and Vo-
corresponding to .it' and a(.it) as long as a leaves the field of definition stable
(even if a is nontrivial on K). If .it is defined by J(X, Y1 , ... , Ym ), then a(.it)
is obviously defined by the polynomial a(f)(X, Y1 , ... , Y m ) obtained from J
by applying a to the coefficients in K of J(X, Y1 , ... , Ym). Thus Vo- as above
is actually given by the conjugate a(V) (i.e., a(V)(K) = a(V(K)) for any
choice of extension of a to K). We have proven:
Corollary 2.26 Let VI K and V; K be smooth projective curves. Then we
have a canonical isomorphism HomK(V, V') ~ HomK(K(V'), K(V)) given
by K(V') :3 ¢ f--t ¢ 0 J Jor a morphism J : V --+ V' oj projective curves, where
HomK(V, V') is the collection oj morphisms oj projective curves defined over
K and HomK(K(V'), K(V)) is the set oj all field homomorphisms over K.
2.4 Algebraic Curves over a Field 4:3
For each f E HomK(V, VI), taking the corresponding field homomorphism

C5 : K(V/) '-+ K(V), we define the degree of the morphism f by the field
extension degree [K(V) : C5(K(V/))]. By definition, deg(J) also can be given
by rank vp ' (EBPEf-'(P')Vp) for any choice of a closed point pi E VI.
Exercises
1. Prove the surjectivity of Vp so that Vp in (2.6) is a DVR.

2. For a valuation ring V of K(x) of a valuation trivial on K but nontrivial
on K(x), if x ric V, show that V = K[z](z) for z = x-l.
3. Describe the intersection in P(K):
{maximal ideals of K[x]} n {maximal ideals of K[X-l]}.
4. Give a more detailed proof of Corollary 2.25.
2.4.2 Zariski Topology
Let R be an algebraic function field over an algebraically closed field K. We

write V for its Zariski-Riemann space or equivalently the set of K-rational
points of the associated smooth projective algebraic curve. We equip V(K)
with a topology in the following way. We cover V by "open" affine subvarieties.
In the language of schemes, an open affine subvariety U C V is given by the
spectrum of an integral domain R; that is, U = Spec(R) and V - U is a
proper closed subset of V (i.e., heuristically, a subset defined by the zero set
of polynomial equations whose meaning becomes clear later).
Recall that vp is the valuation Vp : R .....;; Z U {CXJ} associated with each
point P E V(K). We put
Vp = {J E o} (the discrete valuation ring of vp),

J-tlvp(J) 2:
mp = {J E Vplvp(J) > o} (the maximal ideal of V p ), and (2.7)
K(P) = Vp/mp (the residue field of Vp).
For each nonconstant x E R, we can think of
(2.8)
Then Rx is the integral closure of the polynomial ring K[x] in R (Exercise

1), and V p is the localization of Rx at P x = P n Rx (Exercise 2). Then
the set Spec(Rx)(K) of all maximal ideals of Rx is in bijection with Vx =
{p E VIV p :3 x} (see [GME]1.2 and 1.4.3 for Spec(Rx)). We also have V =
UYER-K Vy = Vx U Vl/x' For each P E V and for x E V p (so, P E Vx), Vx is
called an open affine neighborhood of P. Any finite intersection VXI n· .. n VXm
of Vx) :3 P (j = 1,2, ... , m) is also called an open affine neighborhood of P.
Each open affine neighborhood U of x is in bijection with the set of maximal
ideals Spec(R)(K) of a subring R C .it finitely generated over K. For example,

Vx ~ Spec(Rx)(K) and Vx n Vy ~ Spec(Rx . Ry)(K) (Rx . Ry is the composite
ring of Rx and Ry in .it), because Vp :J Rx . Ry {o} Vp :J Rx and Vp :J R y.
This topology we defined on V is called the Zariski topology on V.
2.4.3 Divisors
We now introduce divisors on an algebraic curve V. We start with interpreting

elements in .it as a function defined on V with values in the projective space
pl. Each f E .it - K gives rise to an inclusion K(f) '---+ .it and hence gives rise
to a (rational or meromorphic) function f : V -+ pI with
if P VI,
E
if P E VI/I,
where p~ = {P E p l lP:3 g}. This is well-defined by the following. If P E VI,

then P n K[J] = (f - a) for a E K = P}, and a = (f mod P) = f(P). If
P E VI n VI/I' then P n K[t] = (t -
~) and still f(P) = a. If P fj. VI, then
f(P) = 00. When f(P) = 0, we call P a zero of f and when f(P) = 00, we
call P a pole of f. For f E K, we associate a constant function f : V -+ pI
with value f everywhere. In this way, we can identify .it with the set of all
merom orphic functions on V.
For each P E V, take t = tp with vp(tp) = 1 (such a tp always exists
because vp:.it -+.z U {oo} is surjective). By definition, mp = (tp) in Vp.
Then we can expand f E .it into a Laurent series of tp as follows. First suppose
that P is not a pole of f· Then f - f(P) has zero at P; so, !I = (f - f(P))/tp
does not have a pole at P; so, we have f - (f(P) + !I (P)tp) is divisible by t~
and so on. We thus have f = Ln>o cntp E K[[tp]]. If f has a pole at P, killing
the pole by multiplying f by t~p(J)I, 9 = t~p(J)lf does not have a pole at P;
so, we apply the above argument, and after expanding g into a power series
of tp, we divide the power series by t~p(J)1 to get the Laurent expansion of f.
We have f = Ln2:vp(J) cntp E K((tp)) = K[[tp]][t~]' If we choose another
t'p with vp( t'p) = 1, we find that tp = t'p( Co + CI t'p + C2t'p 2 + ... ) with Co -=J O.
Thus t'p/tp E V; and if we write the power series expansion of f with respect
to tp as f(tp), then we get f(t'p) by substituting tp(t'p) for tp (check that
this substitution gives a well-defined power series in t'p because tp(P) = 0).
The divisor div(f) of f -=J 0 is defined by a formal sum Lp vp(f)P (in
some classical books in number theory, they use multiplicative notation, such
as div(f) = TIp pVP(J), because it corresponds to the prime decomposition of
the ideal (f), but we use additive symbols following the geometric tradition).
Since f satisfies a polynomial equation of finite degree over K[x] for a suitably
chosen x E .it, f has only finitely many zeros and poles; thus the sum defining
div(f) is actually a finite sum (because vp(f) = 0 except for finitely many
points P E V).
We define the divisor group Div(V) = Div(Jt) by the free abelian group of
all formal finite sums D = LPEV epP, where ep = 0 except for finitely many
P E V. We define the degree of a divisor D = LPEV epP by deg(D) = Lp ep.
We have a homomorphism deg : Div(V) -+ £, and we define Divo(V)
Ker(deg).
Proposition 2.27 If f E Jt, then deg(div(J)) = 0; so, div(J) E Divo(V).
Proof. We may assume that f E Jt - K, because the assertion is clear for

f E KX. The inclusion K(J) Y Jt induces a projection f : V -+ pl. Then
for each p E pl, regarding it as a prime ideal of K[J], we find pR = TIp pep
with prime ideals P in the integral closure R of K[J]. Then by definition, we
have f(P) = 0 {:} ep > 0 and Lp ep = [R: K[f]] = [Jt: K(J)]. In particular,
writing 00 = TIQ QeQ for the points Q with f( Q) = 00, we find that
deg(div(J)) = L ep - Leo = [Jt : K(J)] - [Jt: K(J)] =0

p Q
as desired. 0
Write P(V) = {div(J)lf E Jt X }. Then P(V) is a subgroup of Div°(V).

The quotient group Pic(V) = Div(V)jP(V) (resp. Jac(V) = Pico(V) =
Div°(V)jP(V)) is called the Picard group (resp. the Jacobian) of V. When
D - D' E P(V), we say that D is linearly equivalent to D', and if deg(D) =
deg(D'), we call D is algebraically equivalent to D'.
2.4.4 Differentials
We introduce a notion of differential forms on an algebraic curve V in a purely

algebraic way. For the moment, until the end of the proof of Theorem 2.29,
we suppose that K is algebraically closed.
For a given f E Jt, we formally define df as a collection of power series
df (t p) = d~~t:) dt p. Here P runs through all points P E V, and
d(Ln p
dt cnt )dt p = ('""""' n l
6 cn nt p - ) dt p.
p n
In particular, df does not have the term t~' If we change coordinate tp into
t'p, the expression of df will change according to the chain rule:
df(t' ) = df(t'p) dt' = df(t'p(tp)) dt'p dt .

J p dt'p p dt'p dtp p
We also think of w = gdf as a collection of power series {g( t p ) d~~t:) dt p } FE v.

In other words, if gdf and g'df' give rise to the same power series at every
P E V, we identify gdf = g'df'. The totality of all differentials on V gives
rise to a vector space over it. We see gdf / g' dl' = .if if,
E it, because f and
l' satisfy the polynomial relation F(f,I') = 0 over K. In other words, if
F(X, Y) = I;2,J a2,JX2YJ, then taking a formal derivative of F(f, 1') = 0, we
have
0= dF(f, 1') = L
(ia 2,Jr- 1I'J df + ja 2,)r 1')-1 dl') ,
2,)
and df and dl' are linearly dependent over it. Thus this vector space over it
is one-dimensional.
We define v p (gdf) = v p (g dfd~t:)). Note here that v p (~~;) = 0 for two
parameters tp and t'p at P E V. By the chain rule, d'tp = ; /p ~~'p,
p we have
dfp )
v p ( g dt = vp (gd f
dt'p dt'p)
dt p = vp (gd f
dt'p ) + vp ( dt
dt'pp ) = vp (dJ
g dt'p ) .
Since the last expression of the above formula is the value vp(gdf) computed
with respect to the coordinate t'p, the value vp(gdf) is well-defined indepen-
dently of the choice of the parameter tp around P. In particular, we have
gdf = df (t p ) ) dtp
( g~ = (""
~' cntp
n ) dtp.
n~vp(gdf)
We define div(w) = I;p vp(w)P.

Since differentials form a one-dimensional vector space over it, for any
two differentials wand Wi, div(w) and div(w ' ) are linearly equivalent. Thus
deg( div(w)) is independent of the choice of w. Writing deg(w) = 2g ~ 2, we
define the genus g = g(V) of V.
A differential w is called holomorphic or of the first kind if vp(w) ?:': 0 for
all P E V. If Wi is holomorphic, writing Wi = fw, we find div(f) + div(w) =
div(w ' ) ?:': 0 (we write D ?:': 0 for D = I;p epP if ep ?:': 0 for all P). We have
f E L(div(w)) = {g E itl div(g)?:,: div(w)} .
The space Dv / K of all holomorphic differentials is isomorphic to L( div( w))

by fw H f for any choice w of nonzero differentials, and dimK L(div(w)) =
dimK DV/K is finite and independent of the choice of w (see Lemma 2.30).
We later show (see Theorem 2.31) that this dimension is given by the genus
g; so, g ?:': O.
For a differential w = gdf, we expand it into a power series in tp at P E V,
and write
w = ( L cntp) dtp.
n~vp(w)
Then we define Resp(w) = C-1.

Proposition 2.28 The residue Resp(w) is well-defined independently of the

choice of the parameter tp.
Since w has only finitely many poles, Res(w) = Lp Resp(w) E K is well-
defined.
Proof. We give a proof valid only when K is of characteristic O. See [ALF]
Appendix and [ALG] Theorems III.7.14.1-2 for the proof valid for arbitrary
characteristics. Choose two parameters t = tp and t' = t'p, and expand t' =
alt + a2t2 + ... with al i- O. If the expansion of w = gdf with respect to t is
given by 9 ;{ dt' = Ln Cn tIn dt', the one with respect to t is given by
Let us look into the term involving rl;
cl(al t + a2 t2 + ... )-1 (Lja]tJ- I ) dt = c:a 1 + higher terms.

]~1 1
We need to show that cn(a1t + a2t2 + ... )n (L]ja]t]-I) = cnt,n~t: for

.
n i- -1does not mvolve the term r 1. ThIS. follows from t' n ddtt = n+1 1 -d-t
I dt +
- ,
tn 1
since the Laurent series expansion of ~r does not involve the term rl which
is not a derivative of a power of t. This finishes the proof.
Since ft./K(x) is a separable finite extension for a nonconstant x E ft., we
have the trace map TrstjK(x) ; ft. --+ K(x) for any nonconstant function x.
We study how the residue map behaves under field extensions using the trace
map. The inclusion K(x) y ft. is induced by the projection x ; V --+ pl. For
a prime ideal p of K[x], we consider X-l(p) = {P}; thus, pR = ITppe p for
the integral closure R of K[x] in ft..
By the Chinese remainder theorem, R/pR ~ ITp R/ pep and R/pn R ~
ITp R/ pnep. Thus the completion of ft. with respect to p is given by fi.p =
ITp:x(P)oo=p ft.p. In particular, we have Trstj K(x) (f) = Lpx(p)oo=p Trstpj K(x)p (f)
for f E ft., because TrstjK(x)(f) is Tr(p(f)) for the matrix p(f) E Mn(K(x))
given by (fgl, fg2,"" jgd) = (gl,"" gd)p(f) for a base g] of ft./ K(x).
In ft.p = K[[tlJ (t = tp), we find t e = utp (e = ep) for a unit power series
u = Co + Cl t + C2t2 + ... with Co i- O. We can take an eth root v of u in
K[[t]] because K is algebraically closed. Changing t by vt, we may assume
that K(x)p = K((t ep )) and tp = teo Then we realize that K((t)) is a Galois
extension of K(x)p = K(W)) with the Galois group isomorphic to the group
of eth roots of unity {Le. Each ( E {Le acts on t by t t-+ (t. Then
TrstpjK(x)pt n = ( L
(EfJ.e
en) t n {~tn
=
e f n,
if eln.
We see that for t n E .itp, Resp(Tr.l\pjK(x)p(tn)dW)) = e or 0 accordingly as

n = -e or not. On the other hand, we see Resp(tndW)) = Resp(etn+e-ldt) =
e or 0 accordingly as n = -e or not, because t n <iJte = et n+e - l . Thus
Resp(Tr.l\jK(x)(f)dtp) = Lp:x(p)=pResp(fdtp), and hence we get, for all
I E.it,
L Resp(Tr.l\/K(x)(f)dx) = L Resp(fdx). (2.9)
PEP' PEV
Theorem 2.29 (Residue Theorem) For every differential w on V, we
have LPEV Resp(w) = O.
Proof. Pick x E .it - K. Since the space of differentials is one-dimensional

over .it, we can write w = Idx for I E .it. Then by (2.9), we may assume that
.it = K(x). Since pl = K U {oo}, for each a E K, we take ta. = x - a to
be our parameter. At 00, we choose too = ~ to be the parameter there. For
each I E K(x), expand I = Ln cnt~ and define [/la. = Ln<o cnt~. Then
[/la. = 0 except for finitely many a and ¢ = 1- La.EP' [/la. E K, because
1- [/la. E K[xl(x-a.) = Va. and L.e#a.[/l.e E Va. (which implies the difference ¢
is a rational function without a pole; so, constant by deg(¢) = 0). We expand
I = La.EK,J (xc~~)J + LJ;:::o aJxJ. We only need to check the vanishing for
t-;;jdx = t-;;jdta. and xJdx for j ::::: 1. By computation, we have Resa.(xJdx) = 0
for all a E K U {oo} because x Jdx = -t-;;]-2dt oo ' Similarly (Exercise 5),
if f3 = a and j = 1,
if f3 = 00 and j = 1, (2.10)
otherwise.
This finishes the proof.
We now treat the general case; so, K is an arbitrary field inside its algebraic
closure K. We suppose that K is algebraically closed in .it. The composite
K.it ~ .it ®K K is an algebraic function field over K. We write V (resp.
V) for the Zariski-Riemann space of.it = K(V) (resp. K.it = K(V». Then
we have a natural inclusion V(K) '-+ V(K). Note here that V(K) = {P E
VIK(P) = K}. If K(P) "2 K, then there are [K(P) : K] points corresponding
to P. In other words, for points P l , ... ,Pd with d = [K(P} : Kl in V(K),
PJ n.it = P {:} vPJ 1.1\ = Vp or, equivalently, Pj corresponds to each field
embedding K(P) '-+ K. We define Resp(w) = Lj ResPJ (w), where w is the
differential of K.it corresponding to w = gdl. Since ResPJ (w) for different j
are conjugates of each other, we find Resp(w) E K. Similarly, we send P to
the sum L J Pj (trace of PJ ) in Div(V) and in this way, we embed Div(V) into
Div(V). This embedding sends P(V) into P(V), and we have deg on Div(V)
by pulling back deg defined on Div(V). In this way, everything we stated (in
particular, the residue theorem, Theorem 2.29) is valid over the general field
K (not necessarily algebraically closed).
Similarly to L(div(w)), we define for a general divisor D,
L(D) = {j E ~I div(f) :::: -D},
where for two divisors D = I:p epP and D' = I:p e'pP, we write D :::: D' if
ep :::: e'p for all P. Plainly L(D) is a K-vector space.
Lemma 2.30 The dimension dimK L(D) is finite. In particular, we have
L(O) = K and dimK L(O) = 1.
Proof. If f is nonconstant, div(f) is nontrivial (in V) and deg( div(f)) = o.
Thus L(O) c K n ~ = K (by the definition of algebraic function field, the
integral closure in ~ of the base field is itself). For sufficiently positive D', we
see L(D) c L(D'). Writing D' = I:p e'pP and expanding f E L(D') into a
Laurent series of tp at P E D', we have e'p linear forms taking f E L(D') to
the coefficient of t pj for 0 < j ~ e'p. If all such linear forms vanish at f, we
see f E L(O) and hence f is a constant. Thus dimK L(D') < 00, which proves
the desired assertion. 0
Since f . L(D) = L(D - div(f)), the dimension dimK L(D) depends only
on the linear equivalence class of D. For the linear equivalence class f2 of
differential divisors, we have a well-defined number dimK L(f2 - D).
We introduce the following Riemann-Roch theorem (see [ENT] Chap-
ter VI, [FAN] 7.2 and [GME] 2.1.3 for different proofs).
Theorem 2.31 We have for all DE Div(V),
dimK L(D) = deg(D) - 9 + 1 + dimK L(f2 - D),
where 9 is the genus of the curve V.

Applying this theorem to the trivial divisor 0, we find
1 = dimK L(O) = deg(O) - 9 + 1 + dimK L(f2),
which shows that 9 = dimK L(f2) = dimK f2V/K as we claimed before.
Exercises
1. Prove that Rx in (2.8) is the integral closure of K[x] in ~.

2. Prove that Vp is the localization of Rx at P x = P n Rx.
3. Show that V is not a Hausdorff space under its Zariski topology.
4. Prove Theorem 2.28 for an algebraic function field of characteristic p > O.
5. Give a detailed proof of (2.10).
2.4.5 Adele Rings of Algebraic Function Fields
We insert a brief interpretation of differentials by the language of adeles (due

to A. Weil; see [ALF] Chapter 2). We write stp for the Laurent series ring:
K(P)((tp)) and consider the product ring TI PEV stp, which we consider as a
st-algebra by embedding st diagonally. We define the adele ring stA of st by
stA = st + V c II stp,
PEV
where V = TIPEV Vp for Vp = K(P)[[tp]]. Here we do not assume that K is

algebraically closed. Similarly to the case of Q, we can identify
stA = {(fp)pEvlfp E Vp except for finitely many p}. (2.11)
For a divisor D = Lp epP, we define V(D) = {(Jp) E Vlvp(Jp) 2': -ep}. We

equip with stA the topology whose system of neighborhoods of f E stA is given
by f+V(D) for all negative divisors D. Here we call D negative if ep ::; 0 for all
P. By definition, stA becomes a topological ring. For each f = (Jp)p E st~, we
define its divisor div(J) = Lp vp(fp )P. This is well-defined since vp(Jp) = 0
for almost all P if f E st~. In particular, VX = {J E stAI div(J) = o}. Thus
we have Div(st~) = Div(V) and hence
(2.12)
For each differential w on V (supposing K to be algebraically closed), we

define a K-linear form w : stA --+ K by f '"""* Lp Resp(Jw). Since fp E Vp
for almost all P, Resp(Jw) = 0 for almost all P; so, the linear form w is
well-defined. By definition, if div(Jw) = div(J) + div(w) 2': 0, then w(J) = O.
Thus w(V(div(w))) = 0, and hence w is a continuous linear form on stA with
respect to the discrete topology on K and vanishes over (the diagonal image)
st by Theorem 2.29.
We consider the space I:- of continuous linear forms f : stA/ st --+ K. We
know that I:- is a vector space over st. Actually I:- is isomorphic to the space
of differentials by w '"""* W. We could have defined the differentials as K-linear
forms on stA/ st, and the proof of the Riemann-Roch theorem can be given
from this point of view (see [BNT] Chapter VI and [ALF] Chapter 2).
Exercises
1. Give a detailed proof of (2.11).
2. Prove that stA is locally compact if and only if K is a finite field.
3. Prove that V(D) + V(D') = V([D, D'D for the least common multiple
[D,D'] of D and D'.
4. Prove that V(D) n V(D') = V((D,D')) for the greatest common divisor
(D,D') of D and D'.
2.5 Elliptic Curves over a Field 51
2.5 Elliptic Curves over a Field
An algebraic function field Rj K is called elliptic if its genus is equal to l. The

corresponding algebraic curve E j K with one designated point 0 E E(K) is
called an elliptic curve. The elliptic curve is a pair (E, 0) j K. The point 0 is
called the origin of E, and two elliptic curves E and EI are isomorphic if we
have an isomorphism ¢ : E ~ EI of algebraic curves that sends the origin to
the origin. We study elliptic curves over a field in detail here. When we regard
the point 0 as a divisor, we write [0] instead of O.
2.5.1 Dimension Formulas
For divisors m[O], we study L(m[O]). Since 0[0] is the trivial divisor 0, we find
L(O[O]) = K. By definition, [2EjK is one-dimensional over K. Since deg(w) =
2g - 2 = 0 for a differential w on E, if w has a zero, then w has to have a pole.
If w is holomorphic nonzero, it vanishes nowhere. In other words, nowhere-
vanishing differentials are all nonzero multiples of w. In particular, div(w) = O.
Recall the linear equivalence class of div(w l ) for all meromorphic differentials
Wi on E. Then [2 is the linear equivalence class of all meromorphic functions
in R, and L([2 - D) ~ L( -D). If D > 0, then dimK L( -D) = 0 (Exercise 1).
We fix a nowhere-vanishing differential w. If f E L([O]), then f has pos-
sibly only one simple pole at O. Thus Reso(fw) = LPEE Resp(fw) = 0 by
Theorem 2.29. The function f is in L(O) and is constant, and dimK L([O]) =
dimK L(O) = l. By the Riemann-Roch theorem, we find
dimK L(m[O]) = deg(m[O]) - g + 1 + dimK L( -m[O]) = m if m > O. (2.13)
More generally, by the same proof, we get

Proposition 2.32 Let E be an elliptic curve over a field K. If D > 0 is a
positive divisor, we have dimK L(D) = deg(D).
We may normalize the uniformizing parameter at the origin 0 in terms
of a holomorphic differential w. Choose a parameter t at the origin. Since w
does not vanish at 0, we have w = (co + CIt + C2t2 + ... )dt with Co i- O.
Making a variable change t = cr/T,
we get w = (1 + alT + a2T2 + ... )dT.
The parameter T with the above property is unique modulo T2 in K[[T]].
Conversely, once we have chosen a formal parameter T modulo T2, there
is a unique nowhere-vanishing differential w of the above form because of
dimK [2EjK = 1 (Exercise 2). Such a T is called "T adapted to w."
Exercises
l. Show that dimK L( -D) = 0 for an elliptic curve E if D > O.

2. Show that for a given parameter t at 0, there exists a unique nowhere-
vanishing differential w so that its expansion in t has constant term 1.
2.5.2 Weierstrass Equations of Elliptic Curves
We now embed ElK into the two-dimensional projective space PjK using a
base of L(3[0]) and determine the equation of the image in Pj K" We first
consider L(n[O]) which has dimension n if n > O. We have L([O]) = K and
L(2[0]) = K1 + Kx. Since x has to have a pole of order 2 at 0, we may
normalize x so that x = T- 2(1 + higher terms) in K[[T]]. Here x is unique up
to translation: x ~ x+a with a E K. Then L(3[0]) = K1+Kx+Ky. We may
then normalize y so that y = -T-3(1+higher terms) (following the tradition,
we later rewrite y for 2y; thus, the normalization will be y = -2T- 3(1 +
higher terms) at the end). Then y is unique up to the affine transformation:
y ~ y + ax + b (a, bE K).
Proposition 2.33 Suppose that the characteristic of the base field K is dif-
ferent from 2 and 3. Then for a given pair (E,w) of an elliptic curve E and
a nowhere-vanishing differential w both defined over K, we can find a unique
base (1, x, y) of L(3[0]) such that E is embedded into Pj K by (1, x, y) whose
image is defined by the affine equation
(2.14)
and w on the image is given by ~. Conversely, a projective algebraic curve

defined by the above equation is an elliptic curve with a specific nowhere-
vanishing differential d:
if and only if the discriminant L1 (E, w) = g~ - 27g~
of 4X 3 - g2 X - g3 does not vanish.
An equation of an elliptic curve E as in (2.14) is called a Weierstrass equation
of E, which is determined by the pair (E, w).
Proof. By the dimension formulas, counting the order of poles at 0 of mono-
mials of x and y, we have
L(4[0]) =K + Kx + Ky + Kx 2 ,
L(5[0]) = K + Kx + Ky + Kx 2 + Kxy and
L(6[0]) = K + Kx + Ky + Kx 2 + Kxy + Kx 3
=K + K x + K y + K x 2 + K xy + K y2,
from which the following relation results,
(2.15)
because the poles of order 6 of y2 and x 3 have to be canceled. Therefore ElK

can be embedded into PjK via P ~ (x(P), yep)). The image is defined by
the equation (2.15).
Now we make a variable change y ~ y + ax + b in order to remove the
terms of xy and y (i.e., we are going to make al = a3 = 0):
(y + ax + b)2 + alX(Y + ax + b) + a3(y + ax + b)

= y2 + (2a + adxy + (2b + a3)y + polynomial in x.
Assuming that 2 is invertible in K, we take a = - a2, and b = - a23 • The

resulting equation is of the form y2 = x 3 + b2x2 + b4 x + b6 . We now make the
change of variable x H x + a' to make b2 = 0:
Assuming that 3 is invertible in K, we take a' = -~. We can rewrite the

equation as in (2.14) (making a variable change 2y H y). By the variable
change as above, we have y = -2T- 3 (1 + higher terms), and from this, we
conclude w = ~. The numbers g2 and g3 are determined by T adapted to a
given nowhere-vanishing differential form w.
Conversely, we have seen that any curve defined by equation (2.14) is
smooth in Example 2.23 if the cubic polynomial F(X) = 4X 3 - g2X - g3 has
three distinct roots in K. In other words, if the discriminant L1 (E, w) of F (X)
does not vanish, E is smooth.
For a given equation, y2 = F(X), the algebraic curve E defined by the
homogeneous equation y2 Z = 4X3 - g2X Z2 - g3Z3 in P7
K has a rational
point 0 = (0,1,0) E E(K), which is 00 in p2. Thus E is smooth over K if
°
and only if L1 (E, w) #- (Exercise 2).
We show that there is a canonical nowhere-vanishing differential wEDEl K
if E is defined by (2.14). If such an w exists, all other holomorphic differ-
entials w' are of the form fw with div(f) ;:::: 0, which implies f E K; so,
g = dimK DEIK = 1, and ElK is an elliptic curve. It is an easy exercise to
show that y-1dx does not vanish on E (Exercise 2).
We summarize what we have seen. Returning to the starting elliptic curve
ElK, for the parameter T at the origin, we see by definition
x = T- 2 (1 + higher degree terms) and y = -2T- 3 (1 + higher degree terms).

This shows
dx -2T-3(1 + ... )
-y = -2T-3( 1 + ... ) dT = (1 + higher degree terms)dT = w.
Thus the nowhere-vanishing differential form w to which T is adapted is given
by d;. Conversely, if L1 #- 0, the curve defined by y2 = 4x 3 - g2X - g3 is an
elliptic curve over K with origin 0 = 00 and a standard nowhere-vanishing
differential form w = d;. This finishes the proof.
Exercises
1. Show that dx/y does not vanish at any point on E.
2. Show that if L1 = 0, the curve defined by y2 = 4x 3 - g2X - g3 is not smooth
at the multiple root a of 4x 3 - g2X - g3 = 0.
2.5.3 Moduli of Weierstrass Type
We continue to assume that the characteristic of K is different from 2 and 3.

Suppose that we are given two elliptic curves (E, w) j K and (E', w') j K with
nowhere-vanishing differential forms wand w'. We call two pairs (E, w) and
(E',w') isomorphic if we have an isomorphism ip : E -t E' with ip*w' = w.
Here for w' = jd9, ip*w' = (J 0 ip)d(9 0 ip); in other words, if (J : .it' -t .it is the
isomorphism of the function fields associated with ip, ip*w' = (J(J)d((J(9)). Let
T' be the parameter at the origin 0 of E' adapted to w'. If ip : (E, w) "=' (E', w'),
then the parameter T = ip*T' mod T2 is adapted to w (because ip*w' = w).
We choose coordinates (x, y) for E and (x', y') for E' relative to T and T' as
above. By the uniqueness of the choice of (x, y) and (x', y'), we know ip* x' = x
and ip*Y' = y. Thus the Weierstrass equations of (E, w) and (E', w') coincide.
We write 92(E,w) and 93(E,w) for the 92 and 93 of the coefficients of the
Weierstrass equation of (E,w). Considering a polynomial ring K[g2,g3J with
variables g2 and 93, if K has a characteristic different from 2 and 3, we have
where [.J indicates the set of isomorphism classes of the objects inside the
bracket and Spec(R)(K) for a ring R is the set of all algebra homomorphisms:
R -t K. The last isomorphism sends (92,93) to the algebra homomorphism ¢
with ¢(X) = g2 and ¢(Y) = 93.
We now classify elliptic curves E eliminating the contribution of the dif-
ferential from the pair (E,w). If ip : E "=' E' for (E,w) and (E',w'), we
have ip*w' = AW with A E K X , because ip*w' is another nowhere-vanishing
differential. Therefore we study KX-orbit: (E,w) mod K X under the action
of A E K X given by (E,W)jK ~ (E,AW)jK, computing the dependence of
9J(E,AW) (j = 2,3) on A for a given pair (E,W)jK' Let T be the parameter
adapted to w. Then AT is adapted to Aw. We see
E )_ (l+T¢(T)) ( \)_ (1 + higher terms) _ \-2 (

X( ,w -
)
T2 =>x E,AW - (ATF - A x E,w ,
E )=(-2+T1/J(T)) (EA)=(-2+higherterms)=A_3( )
y ( ,w T3 =>y, w (AT)3 yE,w.
Since y2 = 4x 3 - 92(E,w)x - g3(E,w), we have
(A- 3y)2 = 4r 6 x 3 - 92(E,w)A- 6 X - A- 6 g3(E,w)

= 4(A -2X)3 - A-4 92 (E, W)(A -2x) - A-6 g3 (E, w).
This shows
(2.16)
Thus we have
Theorem 2.34 If two elliptic curves ElK and ElK are isomorphic, then
choosing nowhere-vanishing differentials wIEand wI E" we have gJ (E', w') =
). -2J gJ (E, w) for). E K x. The constant). is given by <p*w' = ).w.
We define the J-invariant of E by J(E) = (12~(t:,:;;))3. Then J only depends

on E (not the chosen differential w). If J(E) = J(E'), then we have
(12g2(E,w))3 = (12g2(E',w,))3 (E' ') = ).-2J (E )

LJ.(E,w) LJ.(E',w') <==? g] ,w 9J ,w
for a twelfth root ). of LJ. (E, w) / LJ. (E' , w'). Note that the twelfth root ). may
not be in K if K is not algebraically closed.
Conversely, for a given j tic {O, I}, the elliptic curve defined by y2 = 4x 3 -
gx - 9 for 9 = J2~Jl has J-invariant 12 3j. If j = 0 or 1, we can take the
following elliptic curve with J = 0 or 12 3 . If J = 0, then y2 = 4x 3 - 1 and if
J = 12 3 , then y2 = 4x 3 - x. Thus we have
Corollary 2.35 If K is algebraically closed, then J(E) = J(E') ¢} E ~ E'
for two elliptic curves over K . Moreover, for any field K, there exists an
elliptic curve E with a given J (E) E K.
Exercises
l. Prove that gJ(E',w') = ).-2]9J(E,w) for suitable wand w' and a suitable
twelfth root). of LJ.(E,w)/LJ.(E',w') if J(E) = J(E').
2. Explain what happens if J(E) = J(E') but E 'f'- E' over a field Knot
necessarily algebraically closed.
2.5.4 Group Structure on Elliptic Curves
We now introduce algebrogeometric group structure on elliptic curves. First

assume that K is algebraically closed. We know that L(2[0]) = K + K x. For a
given point PEE - {O}, x is finite at P. We put f = x - x(P). The function
f has a zero at P. Since deg(div(f)) = 0 and f has at most an order 2 pole at
0, we have two possibilities: div(f) = P + P' - 2[0] or div(f) = P - [0]. If the
latter case happens, [J't : K(f)] = 1 and hence E ~ pl, which is impossible
since the genus of pl is 0 (Exercise 1). Thus we find a unique P' from P, and
we define P' = -Po When P = 0, the above argument just gives P' = 0; so,
the definition of P >-7 -Pis valid for all P.
Now we define addition of two points P, Q E E. For P, Q E E - {O}, we
solve a system of simultaneous linear equations:
{
X(P)X + y(P)Y + Z = 0,
x(Q)X + y(Q)Y + Z = O.
Here (1, x, y) is a base of L(3[0]). Then for a nontrivial solution (a, b, c) of the
above system, we define 9 = ax+by+c. We assume that Q 1= -Po Then b 1= 0
(because b = 0 =} Q = - P as is clear from the definition of - P). Thus 9 has
a pole of order 3 at O. Since 9 is nonconstant with two zeros P, Q, we find a
unique point R such that div(g) = P+Q+R-3[0]. Then we define P+Q to
be - R. By Abel's theorem we prove in the following subsection, this addition
gives rise to an abelian group structure on E. When Q = - P, we just define
P+Q = O.
Here is a geometric interpretation of the above definition of the addition.
Embed E into p2 by 1, x, y. Then we draw a line aX + bY + cZ = 0 passing
through P, Q in p2, where (X, Y, Z) is the homogeneous coordinate of p2.
Then we consider the function ¢ : E -+ pI given by the equation ¢(X, Y, Z) =
(aX + bY + cZ)/Z, which is an element in .R = K(E). We see that ¢ vanishes
at two points P, Q, and the line intersects with E at a unique point Ragain.
Then div(¢) = P+Q+R-3[0]. In particular, the coordinate of R is a rational
function of the coordinates of P and Q; so, + : E x E -+ E is a morphism
of algebraic geometry (in other words, it is induced from a field embedding
.R '-t .R ® K .R). Similarly, P H -Pis an automorphism of the elliptic curve
(therefore of the algebraic function field .R).
After embedding E into p2, for any field automorphism (J" of K, we can
apply (J" to the coordinate of E. Then we get a new elliptic curve EeJ. If E
is defined, for example, by F(X, Y) = :L",J ai,JXiYJ = 0, then E eJ is defined
by FeJ(X, Y) = :L",j (J"(ai,J)Xiyj = O. Since everything we have proved for E
shifts to EeJ, the morphism + : E x E H E will be sent to + : EeJ x EeJ -+ EeJ.
If E is defined over a smaller field k c k = K, the rational functions of the
coordinates of E giving rise to + : Ex E -+ E are therefore invariant under (J";
so, it is a rational function with coefficients in k. Thus what we have said so
far is valid for any elliptic curve defined over any perfect field (not necessarily
over an algebraically closed field).
Exercises
1. Prove that pI has genus O.

2. Give a detailed argument why the morphism giving addition on E is well-
defined over its field of definition.
2.5.5 Abel's Theorem
We now prove the following Abel's theorem for elliptic curves.

Theorem 2.36 Let E be an elliptic curve with origin 0 defined over a per-
fect field K. Then i : E(K) -+ Jac(E)(K) given by i(P) = P - [0] is an
isomorphism of groups.
By this theorem, we find that P + Q + R rv 3[0] (here rv is the linear equiva-
lence) if and only if i(P) + i(Q) + i(R) = 0 in Jac(E); so, the group structure
we have defined in the previous subsection is induced from the group structure
of the Jacobian Jac(E).
Proof. By the same argument (as in the last part of the previous subsection)
conjugating by automorphisms of K, everything we prove over an algebraically
closed field K will be valid over any subfield k such that K j k is algebraic.
Thus we may assume that K is algebraically closed. We first remark that
L(D) i= {O} =} deg(D) ;:::: O. (2.17)
Indeed, if 0 i= f E L(D) implies that div(f) + D ;:::: 0; so, taking the degree,
we have deg(D) ;:::: - deg(div(f)) = O. This shows
deg(D) < 0 =} dimK L(D) = O. (2.18)
Take a degree 1 divisor D = 'Lt=1

Pt - 'L~:; Qj. By the Riemann-Roch
theorem, we have dim L(D) = deg(D) + dim L( -D) = deg(D) = 1 by (2.18).
We find a nonconstant f E L(D). This f has to vanish at all Qj. We have two
possibilities:
• div(f) = 'L~:; QJ - 'Lt::11 Pi after renumbering P t ; so, D rv Pd ;
• f has an extra zero R. Then div(f) = 'L~:i QJ +R- 'Lt=1 P t ; so, D rv R.
Thus D is linearly equivalent to one point. In other words, for any divisor
Do of degree 0, Do + [0] rv P for a point P, and hence, the class of Do is in
the image of i; so, i is surjective. If i(P) = i(Q) for P i= Q, then we have
P - Q rv 0, and hence dim L(P - Q) = dim L(O) = 1. Therefore we have a
function f with a simple pole at P and a simple zero at Q. Thus [.it: K(f)] = 1
and E '::::' pI, a contradiction; so, i is injective. 0
2.5.6 Torsion Points on Elliptic Curves
Let T be a parameter of E over K at the ongm O. Then taking a copy

5 of T, we may think that 5, T is a set of parameters of E x E at (0,0).
Then (P, Q) H f(P + Q) gives rise to a meromorphic function on E x E.
In other words, the function is a pullback of f by +. In particular, we have
To + = <P E K[[T,5]]. For f(T), we find f 0 + (T, 5) = f(<P(T,5)), and by
associativity, we get <P(T, <P(5, U)) = <P(<P(T, 5), U) (Exercise 1). In particular,
sending (T,5) H <P(T,5) = aT + b5 E (T, 5) j (T, 5)2, we get a new asso-
ciative and commutative group law on K, which is linear with respect to K.
Since such a group law is unique on the field K, we find that <P(T, 5) T + 5 =
mod (T,5)2 and, hence, the addition of the elliptic curve coincides with the
usual addition on K in this sense.
,-_-,A,-_ _
N
Let N be a positive integer. Then x H N· x = ;; + x + ... + i is a mor-

phism of E into itself, and N induces on the vector space (T)j(T2) the mul-
tiplication by N. The morphism x H N . x induces an embedding N* of the
function field .R = K(E) into itself. The degree [.R: N*.R] = deg(N) is finite,
because .R is finitely generated over K and of transcendental degree 1. Thus
we find that N : E -+ E is surjective for all positive integer N. In particular,
E(K) is an N-divisible group if K is algebraically closed, and we have proven
this fact as long as N is prime to the characteristic of K.
A naive question is: What is deg(N) = [.R: N*(.R)]? The following theorem
gives us that information and slightly more.
Theorem 2.37 Suppose that K is algebraically closed. Then deg(N) = N 2.
Moreover, if N is outside the characteristic of K, we have E[N](K) ~
(71,1 N71,)2 as abelian groups, and if p > 0 is the characteristic of the field
K, then E lP] is either trivial or isomorphic to 71,1 p71,.
We prove this theorem in the following subsection when K is a subfield of C.
See [GME] 2.6 for more general cases.
In the rest of this subsection, we explore consequences of the theorem.
Start with an arbitrary field K and write K for its algebraic closure. Take an
elliptic curve E defined over K with function field .R. We write the composite
.RK as ~, which gives rise to an elliptic curve E over K. As the algebraic curve
inside pm (the integer m can be taken to be 2 as we have seen), E is just
defined by equations giving relations among generators of.R over K. Since ~
has the same generators (as .R) over K, they give rise to the same algebraic
curve inside pm. For any extension F 1K inside K, we define E (F) to be the
set of points of E with coordinates in F. Thus E(F) is the subset of E(K)
fixed by Gal(K 1F).
Take two points P, Q in E(K). Since P + Q is determined by the third
intersection with E of the line passing through P and Q, if we conjugate the
coordinate by a E Gal(K1K), the resulting image goes to the third intersec-
tion with E of the line passing through O'(P) and O'(Q) (when P = Q, the line
is the line tangent to P E E(K)). This shows that a(P) + a(Q) = O'(P + Q),
and conjugation by Galois automorphism is compatible with the addition of
E. In particular, O'(N . P) = N . O'(P), so the Galois action preserves E[N]
(since 0'(0) = 0 because 0 E E(K)). In other words, a : E[N] -+ E[N] is
an injective group homomorphism. Since E[N] is a finite group, a induces an
automorphism of the group E[N].
Let P, Q be a base of E[N] over 71,IN71,. Then O'(P), O'(Q) is another base.
Write p( a) the base-change matrix:
( a(p))
a(Q)
= (ap+b Q )
cP+dQ
= (acdb) (P)
Q
with p(o') = (~~). By definition, we find p(O')p(T) = p(O'T) and p(l) = h,

where 12 = (6 ~).
Theorem 2.38 Let E be an elliptic curve defined over K. Suppose that the
integer N is outside the characteristic of K. Then a I---t p(O') is a homo-
morphism p : Gal(KIK) -+ GL 2(71,IN71,), where GL 2(71,IN71,) is the group
of invertible two-by-two matrices with coefficients in the ring Z/NZ. More-

over, detp(O") = XN(O"), where XN(O") is determined by O"((N) = (it((J) for a
primitive root of unity (N'
By the above theorem, the field F = K(E[N]) generated by coordinates of
all points of E[N] is a Galois extension of K with Gal( F / K) ~ Im(p) C
GL 2 (Z/NZ), and F contains all Nth roots of unity. We later explore some
cases where p is surjective.
Proof. We only need to prove the last assertion about det p = XN for the
cyclotomic character XN. We construct a bilinear form e = eN : E[N] x
E[N] --t fLN for the group of Nth roots of unity fLN such that
(PI) e(x, x) = I for all x E E[N] (alternating);
(P2) e(x, E[N]) = I =} x = 0 (nondegeneracy);
(P3) O"(e(x, y)) = e(O"(x), O"(y)) (Galois equivariance).
If this is worked out, we find e(P, Q) = (N is a primitive Nth root (P2) and
O"((N) = O"(e(P, Q)) = e(O"(P), O"(Q))

= e(aP + bQ, cP + dQ) = e(P, Qtde(Q, p)bc = ('f:.f- bc
as desired (because I = e(P + Q, P + Q) = e(P, Q)e(Q, P)). We need to
construct e and to prove (PI-3) above. Here is a construction given by A.
Weil: we need to consider a point PEE both as a geometric point and also a
divisor; so, we write [P] when we consider P as a divisor. By Abel's theorem,
(div) if L, c,x, = 0 for x, E E(K) and L, c, = 0 for integers c,' then
0= L, c,([x,] - [0]) = L, c,[xJ
Pick x E E[N]; so, N[x] - N[O] = div(fx) for a function fx E ~. Since E(K)
is divisible, we can pick t E E(K) such that x = Nt. We consider the divisor
D = LUEE[N][t + u]- LUEE[N][U]. Then by (div), we find a function gx E ~
such that div(gx) = D. We see that div(fxoN) is made up of points that go to
o and t with multiplicity N; so, div(fx 0 N) = N . div(gx). Thus g;; / fx 0 N is
a nonzero constant c, and gx(p+y)N = cfx(N(P+y)) = cfx(NP) = gx(p)N
for y E E[N]. We then define e(x, y) = gx(P+y)/gx(P) which is an Nth roots
of unity. By this construction, g(J(x) = CIO"(gx) for a nonzero constant Cl E K;
so, we may choose g(J(x) to be dgx). Then Galois equivariance follows from
O"(e(x, y))O"(gx(P)) = O"(gx(P + y))

= O"(gx)(O"(P) + O"(y)) = e(dx), O"(y))O"(gx(P)).
For y,y' E E[N], gx(P+y+y') = e(x,y')gx(P+Y) = e(x,y)e(x,y')gx(P);

so, the pairing is linear with respect to the right variable. We now take care
of the linearity for the left variable. Let z = x + w for x, w E E[N]. Since
x + w - z - 0 = 0, applying (div), we find h E ~ such that div(h) = [x] +
[w]- [z] - [0]. Then
div(Jxfwf;l) = N([x] - [0] + [w] - [0] + [0] - [z]) = N . div(h)
and N ·div(gxgwg;i) = N ·div(hoN). Thus gxgwg;i = c(hoN) for a nonzero

constant c E K. In other words,
e(x, y)-ie( w, y)-ie(z, y)gx(P)gw(p)g;l (P)

= gx(P + y)gw(P + y)g;i(p + y) = c· h(P) = gx(P)gw(p)g;l(p),
which shows the desired left linearity.

If e(x, y) = 1 for all y E E[N], gx factors through the image of N; that
is, gx = g' 0 N, and hence fx = g'N (so, N . div(g') = div(Jx)) because
N : E -+ E is surjective. We have div(g') = [x] - [0], and by Abel's theorem
x = O. This proves the nondegeneracy.
We now prove (PI). Observe
N-i N-i
div( II fx(P - nx)) = N L ([nx + x]- [nx]) = O.
n=O n=O
Then TI~:Ol fx(P - nx) is a constant. Since we have chosen t so that Nt = x,

we have gx(P-nt) = fx(N(P-nt))N = fx(N P-nx)N, and TI~:Ol gx(p-nt)
has to be a constant. Now we plug P - x in P, and we get
gx(P)gx(P - t)··· gx(P - (N - l)t)

= gx(P - t)gx(P - 2t)··· gx(P - (N - l)t)gx(P - x).
This shows that gx(P) = gx(P - x) and hence e(x, x) = 1. 0
Exercises
1. Prove that p(T, P(S, R)) = p(p(T, S), R) in the power series ring of three
variables T, S, Rover K.
2. Prove p(U)p(T) = p(UT).
2.5.7 Classical Weierstrass Theory
Let Eire be an elliptic curve over C. Since E is of genus 1, it is a quotient of

C by a lattice L c C. Here a lattice L c C is a Z-submodule of C generated
by a base (Wi, W2) of Cover lR. We write Lat for the set of all lattices in C.
Taking a nowhere-vanishing differential w on E(C), we rediscover the lat-
tice L of C (associated with (E,w)) by
Abel's isomorphism is then given by P f--t J:

W E Cj L = Jac(E). By Abel's
theorem, we find that E[N] = tyLjL ~ (71jN71)2, which gives a proof of
Theorem 2.37 if K = C. For a general field K of characteristic 0 inside C and
an elliptic curve ElK, the composite Cit gives rise to the elliptic curve Be
over C defined by the same equation; so, E[N](K) ~ EdN], which proves the
theorem basically for all fields of characteristic O.
Conversely, for a given L E Lat, we define the Weierstrass P-functions by
XL(U)=P(U)=~+ '" { 1 _~}=~+92u2+93u4+ ...

2 u ~ (u - £)2 £2 2 u 20 28 '
€EL-{O}
U = P '()
YL () u = -2
u3 - 2 '~
" (u _1 £)3 = -2u -3 + ... ,
€EL-{O}
where
and
Then r.p = YI - 4xi + g2XL + g3 is holomorphic everywhere. Since these

functions factor through the compact space Cj L, r.p has to be constant, because
any nonconstant holomorphic function is an open map (the existence of power
series expansion and the implicit function theorem). Since XL and YL do not
have constant terms, we conclude r.p = o. We have obtained a holomorphic
map (XL, yL) : Cj L - {O} -+ c;c-Looking at the order of poles at 0, we know
the above map is of degree 1, that is, an isomorphism onto its image. It also
extends to the embedding P = (xL,YL, l) = (U 3XL,U 3YL,U 3): CjL -+ Pic-
Thus we get an elliptic curve EL = p(CjL) = E(g2(L),g3(L)). We then have
This shows
Theorem 2.39 (Weierstrass) We have
L t ~ [(E )
a
IE:
an elliptic curve over C, ]
,W Ie w: a nowhere-vanishing differential '
where the straight brackets [ ] indicate the set of all isomorphism classes of
the objects inside.
We now make the space Lat more explicit. Two complex numbers WI, W2
span a lattice if and only if Im(wdw2) -=J. O. Let Sj = {z E q Im(z) > O}.
Interchanging WI and W2 if necessary, we may assume that Im(wdw2) > O. So
we get a natural isomorphism of complex manifolds via (:;;; ) f--t (W2' wd W2):
B = {v = C:;;) E UCX)21 Im(wdw2) > o} ~ <ex x S).
Since v and v' span the same lattice L if and only if v' = av for a E SL 2(Z),
we see that Lat ~ SL 2(Z)\B. The action of a = (~~) E SL 2(Z) on B is given
on <ex x S) by a(u,z) = (cu+d,a(z)) for a(z) = ~;:~.
We now consider the variables g2 and g3 and the J-invariant J defined
below Theorem 2.34 as a function on the upper half complex plane S). In
particular, by Corollary 2.35, J satisfies J(-y(z)) = J(z) for,,( E SL2(Z),
2.6 Elliptic Modular Function Field

Combining the analytic consideration in the previous subsection with the
algebraic study in the earlier ones, we now state and partially prove the first
non-abelian global reciprocity law due to Shimura.
For a lattice L spanned by (Wi, W2) with Im(wdw2) > 0, let E(L) be
the elliptic curve given by E(C) = <ejL. Picking a point u(a, b) = aW 1 j:/w 2 E
E(L)[N] (for 0 -=f. u = (a,b) E (ZjNZ)2), we define a function fu : Lat ---+ <e
by
g2(W)g3(W)
fu(w) = ,1(w) xL(u(a, b)) and xu(w) = xL(u(a, b))
for w = t(Wi,W2) and,1 = g~ - 27g~. Then for a E SL 2(Z),

fu(aw) = fu(w) for 0 -=f. \;/u = (a, b) E (ZjNZ)2 {:} a == ±1 2 mod N.
(2.19)
Also we see easily that xu().w) = ).-2xu(w) for nonzero scalar). E <ex. We
call a meromorphic function f on B a modular form of weight k on r(N)
if f satisfies f(-yw) = f(w) for all "( E SL 2(Z) with "( - 12 E NM2(Z) and
f().w) = ).-kf(w) for)' E <ex. In particular, if k = 0, they are called modular
functions on r(N). Thus fu is a modular function on r(N), and Xu is a
modular form of weight 2 on r(N). We often let r(N) denote the subgroup
of SL 2 (Z) made of matrices congruent to b modulo N. We have the following
exact sequence,
1 ---+ r(N) ---+ SL 2 (Z) ~ SL 2 (ZjNZ) ---+ 1.
The surjectivity of the last "mod N" map is nontrivial (Exercise 1).
Since B = S) x <ex, one may regard a modular form as a meromorphic
function on the upper half complex plane S) by putting f(z) = f (21fi (1' )).
Since a (n = (cz + d) (,t)) for a = (~~), we find that f : S) ---+ <e is a
modular form of weight k on r(N) if the following conditions are satisfied,
(G1) f(a(z)) = f(z)(cz + d)k for all a = (~~) E r(N).
If f is a modular form on r(N), f(z+N) = f(z); so, f is a function on S)jNZ,
where NZ acts by translation. By the variable change, q = qN = expe;;Z),
2.6 Elliptic Modular Function Field 63
we may identify SJ/NZ with the open unit disk {q E rcllql < I} punctured at
o (0 corresponds to 00), and we may regard f as a function of q defined over
the punctured disk. It has a Laurent expansion around 0: f(q) = Ln anqn,
which is called the q-expansion of f. We assume that f(q) is finite tailed; that
is, there exists N > 0 such that an = 0 as long as n < -N.
We can compute explicitly the q-expansion of g2, g3 and Ll for q = ql:
12g 2 = 1 + 240 f {L
n=l O<dln
d 3 } qn
-6393=1-504f{ L d 5 }qn
(2.20)
n=l O<dln
II (1 -
00
Ll = q qn)24.
n=l
From this we can conclude J E q-1Z[[q]] (d. [IAT] Section 4.6).

We consider the Riemann surface Y(N) = r(N)\SJ. Regarding J = 12 3j
as a variable, and considering the elliptic curve E : y2 = 4x 3 - gx - 9 for
9 = J2~Jl defined over Q(J), we can think of Q(J)(E[N]). This is a Galois
extension of Q( J) with
Gal(Q(J)(E[N])/Q(J)) C GL 2(Z/NZ).
We also consider the field JtN generated over Q( J) (now J is a function on

SJ) by fu for all u E (Z/ NZ)2. Since fu is the x-coordinate of the point
corresponding to u = u( a, b) E E[ N] (up to the factor g2 / g3 - 27 g2 in Q( J)),
we may regard JtN C Q(J)(E[N]) and JtN C C(Y(N)). It is easy to see that
a E SL 2 (Z) acts on Y(N) by z >--t a(z); so, it induces an automorphism of
C(Y(N))/C(J).
Theorem 2.40 We have Gal(JtN/Q(J)) ~ GL 2(Z/NZ)/{±1}, and a E
SL 2(Z) acts on f E JtN by f >--t f 0 a. This action of SL 2 (Z) factors through
PSL 2(Z/NZ), which is the subgroup of Gal(JtN /Q(J)).
Proof. We check fu 0 a = fua. Since fua = fu for all u implies that ua = ±u

by definition, we find that Gal(C(Y(N))/C(J)) contains PSL 2 (Z/NZ). Since
Y(N) is the covering of degree IPSL 2(Z/NZ)1 of pl(J), we find that
C(Y(N)) = C(J)(fuIO #u E (Z/NZ)2).
By computing the q-expansion of fu, we find that fu E Q[(N]((q)):
1 nqNn (b qa ((bnqan + (-bnq-an)nqn

+ (1 +""'
00 00
X ---2~ N N N
u - 12 L 1 _ qNn _ (b qa )2 ~ 1 _ qn
n=l N N n=l
for u = (a, b) and q = exp(27fiz/N). By Theorem 2.38, we know that fiN :J

IQ[(N]; so, <C n fiN = <C n IQ[(N] ((q» = IQ[(N]. Since Gal(fiN/IQ(J» contains
all PSL 2 (Z/NZ) and also matrices with any given determinant modulo N,
we find
This finishes the proof.
By the above proof, q gives a parameter at CXl of Y(N) defined over IQ((N);
so, we have
fiN = <C(Y(N» n IQ[(N] ((q» (q

27fiZ) ).
= exp ( N (2.21 )
This shows
Corollary 2.41 The curve Y(N)/c is actually defined over IQ[(N], and
We consider the union fi = UN fiN (note here that fiN C fiM if NIM); so, fi
is a field.
Corollary 2.42 We have Gal(fi/IQ(J» ~ GL 2 (Z)/{±I} and fi n <C = IQcyc.
Moreover, each matrix 9 E G L2 (Z) acts on IQcyc by the action of det(g) under
the identification Gal (lQcyc /IQ) = ZX .
For a E M 2 (Z) with det(a) > 0, we can always write by elementary divisor
theory, a = "( (7i ~ h' with "(, "(' E S L2 (Z). We write 8 = (7i ~). Since
the effect of f H f ° 8 on q-expansion is q H qm/n, the action f H f ° a
preserves the coefficient field IQ[(N]. Since 8- 1 r(N)8 :J r(mnN) (Exercise 2),
fiN 08 C fimnN. Since "(, "(' E SL 2 (Z) preserve fiN, we see that fiNoa C fimnN
and hence f H f ° a is an automorphism T(a) of it. For any a E GL 2(1Q)+,
its integer multiple is in M 2 (Z). Thus the linear transformation Z H a(z) is
induced by an integer matrix; so, T(a) E Aut(fi) is well defined on fi; that is,
GL 2(1Q)+ acts on fi by f H f ° a. We can check (e.g., [IAT] 6.4 or [MFG]
Corollary 3.3):
(AI) GL2(1Q)+GL2(Z)GL2(lR)+ = GL 2(A)+, where + indicates the posi-

tivity of the determinant (at CXl);
(A2) GL 2 (Z x lR)+ n IQx = {±1}.
Thus we can let GL 2 (A)+ act on fi. Here is a theorem of Shimura.
Theorem 2.43 (Elliptic Reciprocity Law) We have an exact sequence
2.6 Elliptic Modular Function Field 65
The action of GL 2 (A)+ on yt is as given above. The difficult part of the proof
is the surjectivity of T for which we refer the reader to [I AT] Theorem 6.23.
We later give a sketch of a proof in a more general setting, Theorem 4.14.
For each open compact subgroup S of GL 2 (Ah/(YGL 2 (lR)+, we have an
algebraic function field yts, which gives rise to a projective curve Vs defined
over the fixed subfield ks of IQcyc by det(S) c A x /lQx lR~ = Gal(lQcyc /IQ).
This curve is the modular curve of level S, and Vs(C) = rs\SJ (forgetting
the points above 00 E P1(J)). Here rs = S n SL 2 (1Q). In particular, Y(N)
corresponds to {u E G L2 (2) lu == 12 mod N}, and we have the following
identity,
[(E, cp : (7L/N7L)2 ~ E[N])IK!eN(cp(l, 0), cp(O, 1)) = (N 1 ~ Y(N)(K),
where K is any field extension of IQ[(N] for a specific primitive Nth root of
unity (N, and [.] indicates the set of isomorphism classes of the pairs (E, cp).
We have (E, cp) I K ~ (E', cp') I K if there exists an isomorphism f : E --+ E'
defined over K such that f 0 cp = cp'.
Since A = A ( 00 ) x lR, we have GL 2 ( / x
A)+ (A(oo))
IQ GL 2 (lR) + = GL hlQl) ,writ-
ing Z(IQ) for the center of GL 2 (1Q) isomorphic to IQx. Consider T giving the
isomorphism
GL 2 (A (00))
T: Z(IQ) ~ Aut(yt).
The fixed field yts of T( S) is an algebraic function field; so, it is the function
field of a unique smooth projective curve Vs defined over ks = yts n IQ
(which is the fixed field of &(det(S)) for &in Theorem 2.12). Therefore the
isomorphism T in Theorem 2.43 gives rise to a tower of algebraic curves {Vs} s
GL (A(oo))
defined over ks indexed by open compact subgroups S c ~(IQI) . The tower
{Vs} s is called Shimura's canonical model of the tower of modular curves (or
simply just the tower of modular curves), which classifies elliptic curves with
additional structures and is the simplest example of the Shimura varieties
we study in the rest of the book. We write Ys (C) for the image of r s \SJ
f
in Vs ( e ) for rs = PGL 2 ( IQ ) n S in
GL (A(oo))
~(IQI) . Then Vs - Ys is a set 0
finite points (called cusps of Vs), and Ys is an open algebraic curve (a quasi-
projective curve) defined over k s . In other words, VS/ks is the unique smooth
compactification of Y Slks '
We may regard Vs as defined over IQ forgetting the requirement of the field
of definition to be the algebraic closure of IQ in the function field ks (Vs). By
the strong approximation theorem (i.e., the density of SL 2 (1Q) in SL 2 (A(00));
e.g., [MFG] 3.1.2), GL 2 (1Q)\GL 2 (A(00))/Z(IQ)S is in bijection with Gal(ks/IQ)
by 9 H &( det(g)). We write 9 S for the conjugate 9 . S· g-l. Since ks Q91Q1 e =
ITaEGal(ks/lQl) e by k Q9 x --+ (a(k)x)a, we have
yts Q91Q1 e= II eytg s,

gEGL 2 (1QI)\ GL~(t~)OO)) IS
which implies
U VS/C = U
aEGal(ks/QI) EGL ("")\ GL2(A(oo» /S
9 2 '" Z(Q)
(2.22)
U VoS/c(C),
where each Vos/c(C) = VgS/ks(C) gives one connected component of VS/QI(C)·

Since Vs(C) with finitely many points above 00 (called cusps) removed is
exactly the quotient Ys = rs \fJ, we find
by (g, z) f-t (z mod rgs). Thus we get the following expression:
(2.23)
We may think of the proalgebraic curve Y = ~ S YS/QI as a model of the

pro-Riemann surface GL 2 (Q)+ \ (GL 2 (A.(00)) x fJ) /Z(Q). This is the point of
view of Deligne ([DI] and [D2]). Then ~ S Vs is the smooth compactification
of the open proalgebraic curve Y. In the rest of this book, we replace the
algebraic group G L(2) /QI by a more general reductive group G /QI and study
the canonical models of G(Q)\ (G(A. (00)) x X) / Z(Q) for the symmetric space
X of G(JR) and their (smooth and minimal) compactifications. Here Z(Q) is
the topological closure of the center Z(Q) of G(Q) in G(A.(oo)). In the case
of G = GL(2) /QI' we have Z(Q) = Z(Q); so, our formulation is consistent. In
our study of p-adic automorphic forms on G, the two formulations, one due
to Deligne and the other due to Shimura, both play fundamental roles.
Exercises
1. Prove the surjectivity of the mod N map from SL 2 (Z) to SL 2 (Z/NZ).

2. Let <5 = ('0 ~). Prove that <5- 1 r(N)<5:J r(mnN).
3. Prove (AI) and (A2).
3
Modular Curves
In this chapter, we translate the result in the latter part of the previous
chapter into the language of schemes, and at the same time, we sketch a proof
of the vertical control theorem in the p-ordinary case of elliptic modular forms.
There are several different proofs of the vertical control theorem:
(1) Through the moduli theory of elliptic curves; this is what we show (Ann.
Sci. Ec. Norm. Sup. 19 (1986) and [GME] Chapter 3);
(2) Through the study of topological cohomology groups and Jacobians of
modular curves. This way has the advantage of simultaneously producing
at the same time Galois representations into GL 2 (IT), where IT is a quotient
of the universal p-ordinary Hecke algebra. The ring IT could be large and
may be free of finite type over W[[X]] (Inventiones 85 (1986));
(3) Through the theory of p-adic Eisenstein measures and p-adic Rankin con-
volution theory. This method was found by A. Wiles in [Wi] and is pre-
sented in the elliptic modular case in my book [LFE] in Chapter 7;
(4) As an application of the identification of Hecke algebras and universal
Galois deformation rings at many different weights (done by Wiles and
Taylor). This method is presented in my book [MFG] 5.3.5.
We follow the first method. Chapter 3 of the book [GME] contains a more
down-to-earth description of the proof. This first method seems most effec-
tive when we try to generalize the VCT to more general reductive groups
(as long as the group admits Shimura varieties). We generalize this proof to
automorphic forms on more general Shimura varieties (see Sections 4.1, 4.2,
and 8.3).
3.1 Basics of Elliptic Curves over a Scheme
We give a brief description of the theory of the moduli of elliptic curves.

Chapter 2 of the book [GME] contains a thorough exposition of the theory.

68 3 Modular Curve~
Studying curves G over a field k in Section 2.4, we have analyzed its set
G(K) of K-points for a field extension K/k. The main idea behind the theory
in Section 2.4 is to associate with each point P E G(K) a discrete valuation
of the function field K(G). However, to study a curve G over a base ring
B, obviously we need to study G(R) directly for an arbitrary B-algebra R.
The main point of scheme theory is to consider G as a covariant functor
R I-t G(R) from the category B-ALG (also written as ALG/ B ) of B-algebras
into the category SETS of sets. We have many interesting functors F : B-
ALG -+ SETS. For example, we may associate with R the set [E/Rl of all
isomorphism classes of elliptic curves defined over R (or Spec(R)). This is
a covariant functor since a B-algebra homomorphism R ~ R' induces the
base change E/R I-t (E ®R R') which is an elliptic curve defined over R'. We
can extend this functor to a contravariant functor from the category B-SG H
(also written as SGH/ B ) of schemes over B to SETS in an obvious manner.
The functor is contravariant, because R I-t Spec(R) reverses the direction of
morphisms. If the functor R I-t [E/Rl is isomorphic to a functor R I-t 9)1(R)
for a B-scheme 9)1, we could use algebraic geometry to study the functor (and
at the same time we may use the functoriality to study the geometry of the
moduli scheme 9)1). Finding 9)1 is a classification problem of elliptic curves.
Of course, we may add some extra structures ¢ to E to classify pairs (E, ¢ ),
and if successful, the moduli 9)1' of the functor R I-t [(E, ¢)/Rl constitutes a
scheme over 9)1 by the morphism (E, ¢) I-t E. We study this type of problem
in this and the following sections. In this section, we summarize necessary
facts on elliptic curves defined over a scheme.
3.1.1 Definition of Elliptic Curves

For a given scheme S, a proper smooth curve f : E -+ S is called an elliptic
curve if it satisfies the following three conditions:
(E1) E has a section D = DE E E(S) (thUS f 0 D = Is);
(E2) dims E = 1, and E is geometrically connected (this means that each
geometric fiber of E over an algebraically closed field is connected and
of dimension 1);
(E3) fJ1E/S (equivalently Rl f*OE by Grothendieck-Serre duality) is lo-
cally free of rank 1 (genus = 1).
Compare this definition to the one already given in Section 2.5. There is no
harm in assuming that S is connected, as we do from now on. For any S-
scheme ¢: T -+ S, the fibered product ET = E xsT is again an elliptic curve
with the zero section DT = DE X IT. For two elliptic curves E and E' over S,
an S-morphism h : E -+ E' is always supposed to take DE to DE'.
3.1.2 Cartier Divisors

A closed subscheme DeE is called an effective Cartier divisor (relative
to S) on E if f*OD = f*(OE/I(D)) given by an invertible sheaf of ideals
3.1 Basics of Elliptic Curves over a Scheme 69
J(D) C OE is S-£l.at (thus locally free). We define C(D) = J(D)-l and put
deg(D) = deg(C(D)) = ranks f*(OD). In particular, the 0 section gives rise
0#
to a divisor [0] of degree 1 given by 0[0] ~ Os. We then consider J(m[O]) =
J([o])m and C(m[O]) = J(m[O])-l for m E Z. The line bundle C(m[O]) can be
regarded as the sheaf of meromorphic functions on E smooth outside 0 and
with possible singularity at 0 having a pole of order ::; m.
If S = Spec(k) for an algebraically closed field k, k-rational effective divi-
sors can be identified with positive linear combinations of points on E(k). We
have deg(Lp mp[P]) = Lp mp. We can thus think of the group Div(Ejk)
of all formal linear combinations (including negative coefficients) of points
on E. Then deg : Div(Ejk) -+ Z is a well-defined homomorphism given by
the above formula. In particular, for any divisor D E Div(Ejk), we have
C(D) = C(D+) 0 C(D_)-l writing D = D+ - D_ for effective divisors D+
and D_, and we can verify deg(C(D)) = deg(D). By Abel's theorem (Theo-
rem 2.36), any invertible sheaf C on E/k is isomorphic to C(D) for a divisor
D, and we define deg(C) = deg(C(D)) if E is defined over an algebraically
closed field k.
3.1.3 Picard Schemes
For any scheme X, we define Pic(X) as the set of all isomorphism classes
of invertible sheaves on X. The association X H Pic(X) is a contravariant
functor by the pullback of invertible sheaves, and Pic(X) is actually a group
by tensor product.
Let E .4 S be an elliptic curve. We define, for each S-scheme ¢ : T -+ S,
PicE/s(T) = Pic(ET)j f; Pic(T).
We can extend the degree map to deg : PicE/s(T) -+ Z 7r o(T) for the set of
connected components 7ro(T). Indeed, for any algebraically closed field k and
a geometric point s : Spec(k) Y T, the fiber E(s) = E xs,¢os s = ET XT S
is an elliptic curve over the field k and deg(C) = deg(C(s)) for the pull back
C( s) at s, which is well-defined independently of the choice of s in a connected
component. We embed Z in Z 7r o(T) diagonally, and we define
PicE/s(T) = {C E PicE/s(T) I deg(C) = r}.

Here is a version of Abel's theorem (e.g., Section 2.5.5 and [GME] 2.2.2):
(Abel) PicE/s(T) ~ E(T) = Homs(T, E) by C([P]) 0 C([OlY- 1 +-+ P.
Thus an elliptic curve is a group scheme with the identity 0 (see [GME]
1.6 for the basics of group schemes). If ¢ : G -+ G' is a nonconstant S-
morphism of two smooth geometrically connected curves, I¢-l (s') I is constant
for geometric points s' of G'; that is, ¢*Oc is locally free of finite rank (e.g.,
[GME] Lemma 2.6.1). We write this number as deg(¢). If the base scheme S
70 3 Modular Curves
is a spectrum of a field, the valuation ring Vs' associated with a closed point
s' E C' is given by the stalk Dc, ,s'; so, this definition of deg( ¢) is consistent
with the one given for algebraic curves over a field just after Corollary 2.26.
Thus Adeg(¢) ¢*Oc is an invertible sheaf on C'. If ¢ : E --+ E' is an S-
morphism of elliptic curves, by our convention, ¢ takes 0 E to 0 E' and, hence,
at the side of the Picard scheme, it is just L --+ Adeg(¢) ¢*L; so, obviously ¢
is a homomorphism of group functors. In particular, a morphism of elliptic
curves (taking zero to zero) is a homomorphism of group schemes.
3.1.4 Invariant Differentials
By (E3), for a dense affine open subset Spec(B) of S, HO(E, DE/B) = Bw

for a I-differential w. For each point P E E(S), Tp : x H x + P gives an
automorphism on E. Since we can therefore bring any given cotangent vector
at 0 isomorphically to a cotangent vector at P, each cotangent vector at 0
extends to a global section of nEls' Thus Tpw = w (see [GME] 2.2.3 for a
more rigorous proof of this invariance).
3.1.5 Classification Functors
An important fact from functorial algebraic geometry is that we can associate,

with each S-scheme X, a contravariant functor X : S-SCH --+ SETS such
that X(T) = Homs(T, X). This association is fully faithful; in other words,
writing CT F for the category of contravariant functors from S-schemes to
SETS, we have
Homs(X, Y) ~ HomcTF(X, Y)
via (X ~ Y) --+ (¢(T): X(T) --+ Y(T)) given by ¢(T)(T -Lr X) = ¢of. This
is intuitively clear because an algebraic variety is just a function associating to
each ring R its R- integral points X (R) = X (Spec( R)). The verification of this
is left to the reader as an exercise (the inverse is given by HomcTF(X, Y) :3
F H F(X)(Ix) where F(X) : X(X) --+ Y(X) = Homs(X, Y); see, for exam-
ple, [MFG] 4.1.3 or [GME] 1.4.3).
This fact can be used as follows. Let N be a positive integer. Since E(T)
is a group, x H N· x gives a functorial map N(T) : E(T) --+ E(T); so, an
endomorphism of elliptic curves N : E --+ E. We define its kernel E[N]
E XE,N.O S:
E[N] ----+ E
1 IN
S ----+ E.
o
It is clear that E[N](T) = Ker(N(T)). It is known that deg N = N 2 (e.g.,
[GME] Theorem 2.6.2) and if N is invertible over S, E[N](k) ~ (Z/NZ)2 for
all algebraically closed fields k with Spec(k) --+ S.
3.1 Basics of Elliptic Curves over a Scheme 71
We consider the following functor
from the category ALG of Z-algebras into SETS, where w is a nowhere-

vanishing invariant differential, P is a point of order exactly N, that is, m H
mP induces an isomorphism (Zj NZ) / R '-+ E of group schemes defined over R,
and [ ] = { }j s=! is the set of all isomorphism classes of the objects inside
the brackets. Here ZjNZ as a group functor associates with T the group
(ZjNZ) oCT). Therefore 0CZ/NZ)/z = EBZ/NZ Z (and 0CZ/NZ)/R = EBZ/NZ R);
7f
so, the structure sheaf of Zj NZ is free of finite rank N. Such a group scheme
is called a locally iree group scheme (of rank N).
There is another example of locally free group schemes: starting with the
multiplicative group G m (as a group functor Gm(R) = R X and as an affine
scheme G m = Spec(Z[t, el])), we consider the kernel JLN of N (as a group
functor JLN(R) = {( E RI(N = 1} and as a scheme JLN = Spec(Z[t]j(tN -1)) =
Spec(Z[(ZjNZ)])). Then JLN is a locally free group scheme of rank N. If
N> 1, it is nonisomorphic to ZjNZ, since for any prime p, JLp(lFp) = {1} but
(ZjpZ)(lFp) = ZjpZ. We consider a version of the functor P~dN) defined as
follows,
Pr,CN)(R) = [(E,¢N: JLN '-+ E[N],w)/R]'
3.1.6 Cartier Duality
The two functors Pr,CN) and PhCN) are isomorphic by Cartier duality (see
[GME]l. 7): If G is a locally free group scheme of rank N over S, there exists a
group scheme G/ s such that G(T) = HomT(GT,G m/ T ) = HomT(GT'~N/T)'
where G m/ S = G m x Sand JLN/S = JLN x S over Spec(Z). We have G s=! G
in an obvious manner, and 'i:jiiZ = JLN by ((m) = (m for ( E JLN(R) and
mE (ZjNZ)(R).
Let E -4
S be an elliptic curve. The section 0 : S -+ E induces a section
of 1* : Pic(S) -+ Pic(E); so, we have a splitting:
Pic(ET ) = r; Pic(T) EEl Ker(Or) and Ker(Or) = PicE/S(T),
regarding Pic E / S (T) as a set of isomorphism classes of invertible sheaves whose

restriction to 0 is trivial; that is, 0* C is isomorphic to Os.
Let P E Ker(7r) for a nonconstant S-morphism 7r : E -+ E'. Then
7r: = P'ICE' /S -+ p.ICE/S =

* E' rv rv E
is an S-homomorphism. We show that Ker(7r*) = ~). Take C E Ker(7r*),

and take an open covering E' = Ut Ut such that Clu, = it-lOU,. Since 0* C =
Os, we may normalize it so that it 0 OE' = iJ 0 OE' for all i #- jon Ut n UJ .
72 3 Modular Curves
Let h, = i, 07r; we have 7r*£lv, = h;:lOv, for V. = 7r- 1 (U,). Let P E Ker(7r);
then
h, 0 P = ii 07r 0 P = i, oOE' = i1 oOE' = h j 0 P.
--
This implies that the h, 0 Ps glue to give a global section hoP
Gm(T), getting a homomorphism Ker(7r*) -+ Ker(7r), which can be easily
verified to be an isomorphism (because twice this operation yields an identity
E r(T, O~) =
map of Ker(7r*)). Since N* = N as we can see easily, we get = E[N]. EfiV]

Writing the pairing as ( , ) : E[N] Xs E[N] -+ J.1NIs, we get (¢(P), Q) =
(P, ¢*(Q)); so, ¢ f--t ¢* is an involution with ¢* 0 ¢ = deg(¢) 2: 0 (a positive
involution; see [GME] 2.6.3). The pairing (".) is the scheme-theoretic version
of the pairing eN(',') we studied in (Pl-3) in the proof of Theorem 2.38.
For a given additive level N-structure ¢N : ZjNZ y E[N], by duality,
we get 7rN : E[N] -» J.1N which has a section ¢~ well-determined modulo
C = ¢N(ZjNZ). Thus (EjC,¢~: J.1N Y (EjC)[N],w') is well-defined as an
element of Pr,(N) (R), where w' coincides with w at the identity (because the
projection E -+ E j C is a local isomorphism; that is, an etale morphism; see
[GME] Proposition 1.8.4). The inverse map: P -+ P' is given similarly by
(E',¢~: J.1N Y E'[N],w') H (E" = E'jlm(¢~)'¢N: ZjNZ Y EI[N],w").

Since (EjC)/¢~(J.1N) = EjE[N] ~ E, we have P' ~ P ~ P' and hence
equivalence.
3.2 Moduli of Elliptic Curves and the Igusa Tower

As already described, we regard a scheme M over a ring B as a contravariant
functor M : SCHI13 -+ SETS by M(S) = Hom13-scH(S, M). The scheme
MI13 is a coarse modulus of a contravariant functor F : SCHI13 -+ SETS if
we have a morphism i : F -+ M of functors such that i(k) : F(Spec(k)) ~
M(Spec(k)) for any algebraically closed field k and for any other morphism g :
F -+ M' for a B-scheme M', we have a morphism M ~ M' with g = 7r 0 i (see
[GME] 2.3.2 (CMl-3) for more details on coarse moduli schemes). If the above
morphism i is an isomorphism of functors, we say that F is representable by
MI13, and MI13 is a fine modulus scheme (see Section 4.4.1). A fine or coarse
modulus of a functor is uniquely determined up to isomorphisms of B-schemes
(Yoneda's lemma; see [GME] Lemma 1.4.1).
We now study the scheme Y1 (N) /Z[ -b 1 representing P r , (N)' This eventually
leads us to the vertical control theorems in the elliptic modular cases.
3.2.1 Moduli of Levell over Z[~]
Hereafter, we assume until Section 3.3 (for simplicity) that 6 is invertible in

any algebra we consider (see [GME] 2.6-9 for the theory valid over Z). Let
3.2 Moduli of Elliptic Curves and the Igusa Tower 73
(E,W)/R be a pair of an elliptic curve and a nowhere-vanishing differential.

We choose a parameter T at 0 so that
W = (1 + higher terms of T)dT.

By the integral version of the Riemann-Roch theorem (see [GME] 2.1.4),
we verify (as in Proposition 2.32) that rankHO(E,£(m[O])) = m ifm > O.
Therefore we have two morphisms x, y : E ---+ pI such that
1. x has a pole of order 2 at 0 with the leading term T- 2 in its Laurent
expansion in T (removing the constant term by translation);
2. y has a pole of order 3 with leading term - T- 3 .
We repeat here the computation of the Weierstrass equation of E done in
Section 2.5.2 more carefully over the ring R as follows. Out of these functions
x and y, we can create bases of HO(E, £(m[O])):
• HO(E, £(2[0])) = R + Rx, HO(E, £(3[0])) = R + Rx + Ry. This implies
that x has a pole of order 2 at 0 and y has order 3 at o. They are regular
outside 0;
• From these functions 1, x, y, we create functions with a pole of order n at
o as follows:
n:::; 4: 1,x,y,x 2 (dim = 4)
n:::; 5: 1,x,y,x 2,xy (dim = 5)
n:::; 6: 1,x,y,x 2,xy,x 3,y2 (dim = 6).
Comparing the leading term of T-6, one sees that the seven sections
in the space HO(E, £(6[0])) have to be linearly dependent and satisfy the
following relation,
We can kill in a unique way the terms involving xy and y by a variable change
y H Y + ax + b. Indeed, by the variable change y H Y - TX - a23 , we get a
simplified equation:
y2 = x 3 + b2x2 + b4 x + b6 .
Again a variable change x H x- pt simplifies the equation to
y2 = x 3 + C2 X + C3.
Since £(3[0]) is very ample (deg(£(3[0])) = 3 ~ 2g + 1), by finally making a
variable change 2y H Y (so now the T-expansion of y begins with -2T- 3 ),
we get a unique equation from (E,W)/R:
74 3 Modular Curves
In other words, E c P7 R is given by
Proj(R[X, Y, Zl!(Zy 2 - 4X 3 + g2(E, w)X Z2 + g3(E, w)Z3)).

It is easy to see by computation that this equation gives a smooth curve
of genus 1 having the origin 0 = 00 = (0,1,0) in p 2 if Ll = Ll(E,w) =
g~ - 27g§ E RX. We recover the differential w by d;. This shows that, writing
R = Z[i, g2, g3, i] for variables g2 and g3,
where Ml = Spec(R) for R = Z[i,g2,g3, i].

We have the universal elliptic
curve and the universal differential w given by
(E,W)/M, = ( Proj(R[X,Y,Zl!(ZY 2 -4X 3 +g2XZ 2 +g3Z3 ))'Y dX) .

For each pair (E,W)/R, we have a unique rp E M1(R) = Homs(Spec(R),Md
(S = Spec(Z[i])) such that
(E,W)/R ~ rp*(E,w) = (E,w) xM, Spec(R).
If we change W by AW for A E R X = Gm(R), the parameter T will be

changed to AT and hence (x,y) is changed to (A- 2X,A- 3y). Thus (E,AW)/R
will be defined by
This has to be equivalent to the original equation by the uniqueness of the

Weierstrass equation, and we have
Again by the uniqueness of the Weierstrass equation, we find that
Aut((E, W)/R) = {lE}
as long as 6 is invertible in R.
3.2.2 Moduli of P r, (N)
Consider (E, P, w) / R for a point P E E[f] (R) of order f for a prime f. We have
a unique rp E M1(R) such that
rpE : (E,W)/R ~ rp*(E,w) = (E,w) xM, Spec(R).

We thus have a commutative diagram:
p
'PE T
Spec(R) --+ E
Then P induces a unique morphism 'PP = 'PE 0 P : Spec(R) -+ (E[e] - {O}).

This shows that, over Z[iE]'
Similarly, over Z[6~],
Pr,(N)(R) ~ Ph(N)(R) ~ (E[N]- U E[d]) (R).

N>dIN
We put MrdN) = E[N] - UN>dIN E[d]. Thus we have proven

Theorem 3.1 There is an affine scheme MrdN) = Spec(RrdN)) defined
over Z[ 6~] such that
PrdN)(R) ~ P~,(N)(R) ~ HomZ[shl-alg(RrdN), R) = MrdN)(R)

for all Z[6~ ]-algebras R. The scheme MrdN)/ Ml is an etale covering of
degree 'P(N) for the Euler function 'P.
The fact that the covering is etale finite follows from the same fact for E[N]
since E[N](k) ~ (Z/NZ)2 for all algebraically closed fields with characteristic
not dividing N. Since Ml is affine, any finite covering of Ml is affine.
If we find an S-scheme M for a given contravariant functor P : S-SC H -+
SETS such that we have an isomorphism of functors M ~ P for M(S') =
Homs(S',M), we say that P is representable (or represented) by an S-scheme
M. The scheme M is called the moduli scheme of the functor P. Then the
statement of the theorem is equivalent to the representability of the functors
Pr,(N) and Ph(N) by an affine Z[6~]-scheme MrdN).
Even if P is not representable by a scheme, we often find a rough approx-
imation of the moduli called a coarse moduli scheme so that P(Spec(k)) ~
M(k) for the spectrum of any algebraically closed field k that is an S-scheme.
Here we do not give the scheme-theoretic characterization of coarse moduli
schemes, only referring to [GME] 2.3.2, though we show many coarse moduli
schemes in this book.
3.2.3 Action of G'1n
The group scheme G m acts on the functor PrdN) in the following way,
76 3 Modular Curves
for ), E Gm(R). This induces (by the functoriality described in Section 3.l.5)
an action of G m on MrdN) and hence on Rr,(N).
Here is a general fact on the action of G m (see [GME] l.6.5). Let X be
a B-module for a base ring B. Regard X as a functor from B-ALG to the
category of B-modules B- MOD by X (R) = X (SJ B R. If a group scheme G / B
has an action coming from the following morphism of functors, G x X -+ X
(which induces the set-theoretic action G(R) x X(R) -+ X(R) for each B-
algebra R), we call X a schematic representation of G. It is known that if X
has a schematic action of G m / B , then
such that X[K:] = {x E XI)'· x = A"x}; that is, X[K:] is the eigenspace for the
character Gm(B) -+ B X taking Z E Gm(B) = B X to z".
The action of G m / B on MrdN) gives rise to a schematic action on RrdN)
(because it was defined by functorial action; see Section 3.l.5). Thus we can
split
RrdN)/R = EB R,,(rl(N); R),
"EZ
where on f E R,,(r1(N); R), G m acts by the character -K:.
Since fER" (rl (N); B) is a functorial morphism,
we may regard f as a function of (E,¢N,W)/R with f((E,¢N,W)/R) E R

satisfying
(GO) f((E,¢N,),W)/R) = )'-"f((E,¢N,W)/R) for)' E R X = Gm(R);
(G1) If (E,¢N,W)/R S=' (E',¢~,W')/R' then we have
(G2) If p : R -+ R' is a morphism of B-algebras, then we have
If a graded ring A = EBJ AJ has a unit u of degree 1, we see that A =

Ao (SJz Z[u, u- 1 ] and Spec(A) = Spec(Ao) x G m by definition; so, Proj (A) =
Spec(A)/G m = Spec(Ao). If A has a unit of degree n > 0, then we still have
Proj(A) = Proj(A(n)) = Spec(Ao)
for A(n) = EBJ AnJ (see [GME] Lemma l.3.1). Since Ll- 1 ERe RrdN) , the
graded ring RrdN) has a unit of degree 12, and hence, we have
We consider the functor defined over Z[ ~ ]-ALG given by
By definition, ErdN) = (Gm \Pr,(N)' Since Proj(Rr,(N)) gives the quotient

by (Gm of Spec(Rr,(N)) (cf. [GME] Theorem 1.8.2), we conclude
Theorem 3.2 (Shimura, Igusa) We have an open curve
defined over Z[6~], which is locally free of finite rank over Ml = Proj(R) =
pl(J) ~ {oo}. For all geometric points Spec(k) of Spec(Z[6~])' we have
Er,(N)(k) = [(E,¢N)/k] = Yl(N)(k). The above assertion holds for any
Z[6~ ]-algebras R in place of algebraically closed fields k if N ;::: 4.
Here a "geometric point" means that k is an algebraically closed field. This
theorem over Z[iI] for an unspecified M with NIM was proved by Shimura
in the late 1950s (see [CPS] I [57a]) , and the result over Z[~] was proved by
Igusa in [I] soon after Shimura's work.
Returning to the classical setting, we define an arithmetic subgroup r 1 (N)
of SL 2 (rfJ) by
rl(N)={(~~) ESL 2 (Z)lc:=0 modN, a:=d:=l modN}.
Since SL 2 (Z) acts on the upper half complex plane 5) discretely by z >--+ ~;:~,
we can make a quotient Riemann surface r 1 (N)\5) (see [IAT] Chapter 1). By
the association: z >--+ (Ez(rc) = C/(Zz+Z), P z = ~ E Ez(rc)), it is well known
that r 1 (N) \5) classifies all elliptic curves with a point P of order N over C
(cf. [IAT] Chapter 4 and [GME] 2.4); so, we conclude
Thus Y1 (N)(rc) is an open Riemann surface.
3.2.4 Compactification
For any Z[i]-algebra R, we put
Let Gr,(N)(Z[6~]) be the integral closure of the ring G(Z) in the graded
r
ring R r , (N) /7/.,[ 1v l' To see that G r , (N) (Z[ 6~]) is a graded ring, we write for
the nontrivial homogeneous projection of highest degree of r E R r ,(N)/7/.,[61vl'
78 3 Modular Curves
If r E R r ,(N)/Z[6JvI is integral over G(Z), r satisfies an equation P(X)

xn + alXn-l + ... + an = 0 with aJ E G(Z). Then r satisfies P(X)
xn + al Xn - l + ... + an = 0, and r is integral over G(Z). Then by induction
r,
of the degree of we see that G r, (N) (Z[ 6~]) is graded (cf. [BCM] V .1.8).
We put for any Z[6~ ]-algebra R,
We then define X l (N)/R = Proj(Gr,(N) (R)). By definition, X l (N) is the

normalization of Proj(G) = Proj(G(12)) = Pl(J) (J = (12~d) for G(12) =
EB %:0 G 12k in Yl (N). As classically known, J- l has a q-expansion starting
with q, that is, J- l E qZ[[q]] (see [IAT] (4.6.1)). Thus the completion of the
local ring of Pl(J) at the cusp 00 is isomorphic to Z[i][[q]]. Moreover, we
have the Tate curve (e.g., [GME] 2.5):
. 1
Tate(q) = ProJ (Z[[q]][6"][X, Y, Z]j(Zy2 - 4X 3 + g2(q)XZ 2 + g3(q)Z3)),
which extends the universal curve over pI (J) - {oo} to pI (J) locally at the
cusp 00.
Since Tate(q)(R[[q]]) :J (R[[qW)/qZ (see [GME] Theorem 2.5.1), we may
regard Tate(q) as the algebraization of the formal quotient Gm/Z[[q)]/qZ of the
formal multiplicative group Gm ; so, it has a canonical level structure
¢rr : /hN Y Gm -» Tate(q).
The Tate couple (Tate(q),¢N'n)/Z[[q)] is an elliptic curve over Z[[q]][q-l]; so,

by the universality of Yl(N), we have a morphism
Since we may regard the Tate curve as a universal formal deformation of a

stable curve of genus 1 (with the level structure ¢N'n) centered at the Z[k]-
point represented by an ideal (q) of Z[k][[q]] ([GME] 2.5.2-3), the morphism
Loo is an infinitesimal isomorphism centered at the cusp 00 (by the univer-
sality of the Yl (N) and the universality of the Tate curve). Since XI (N) is
the normalization of pl(J) in Yl (N), we conclude that the formal comple-
tion along the cusp 00 on X l (N) is canonically identified with R[[q]] by Loo.
Replacing the level structure ¢N'n by ¢N'n 0 a for a E SL 2(Z/NZ), basically
by the same argument, the local ring at the cusp a(oo) of X l (N)/R is given
by R[/hd][[ql/d]] for a suitable divisor diN. We need to extend the scalar to
R[/hd][[ql/d]] because the Tate curve (Tate(q), ¢N'n 0 a) is only defined over
R[/hd][[ql/d]] for a suitable divisor diN dependent on the choice of a. This
point is slightly more technical, and we refer the reader to a more detailed
account, which can be found in [AME] Chapter 10 and [GME] 3.1.1. Thus
Xl (N) is smooth at the cusps, and moreover f E Ck(rl (N); R) is a function
of (E,¢N,w) satisfying (GO-2) and
(G3) f(Tate(q),¢N,w) E R[(N][[ql/NlJ for any choice of ¢N and w.
Since rl(N)\(Sj U Pl(Q)) is a smooth compact Riemann surface and is the
normalization of pl(J) in Yl(N)(C) ([IAT] Chapter 1), we conclude
The space Ck(rl (N); C) is the classical space of modular forms on r l (N) of
weight k. Since Tate(q)(R[[q]]) ::J (R[[qW)jl~, we may consider Tate(q) to be
a "quotient" of Gm/Z[[q)] j qZ; so, it has a canonical level structure ¢'tr : J1N y
G m ...... Tate(q) and a canonical differential Wean induced by !if identifying
G m = Spec(Z[t, e l ]). In particular,
00
f(q) = f(Tate(q),¢'tr,w ean ) = 'La(n;f)qn

n=O
coincides with the Fourier expansion of f at infinity if f E Ck(rl(N);C).
3.2.5 Moduli of r(N)-Level Structure
We consider the following classification problem of level r(N),
for all Z[6~'(N]-algebras R, where (N = exp(27ri-:h). Writing f for the map

(".) : E[N] xM, E[N] -t J1N as a morphism of group schemes over Z[-:h]'
it is obvious that Pr(N) is represented over Z[6~' (N] by the closed affine
subscheme Mr(N)/Z[6iv ,(N] = r
1((N) of E[N] xM, E[N]. Writing Mr(N) =
Spec(Rr(N)) for a graded algebra Rr(N), we define an affine curve
which represents the following functor,
Er(N)(R) = [(E, ¢N : (ZjNZ)2 ~ E[N])I(¢N(l, 0), ¢N(O, 1)) = (NJ '
if N 2: 3 (see [GME] Theorem 2.6.8). Here Ro(r(N)) is the degree 0-

component of the graded algebra Rr(N)' Taking the integral closure C r(N)
of the graded algebra C(Z[ 6~' (N]) in Rr(N), we define the compactification
X(N) = Proj(C r(N)) over Z[6~' (N]. Again we have well-defined q-expansion
at 00.
We caulet a constant group a E SL 2(ZjNZ) act on YeN) (and hence on
X(N)) by (E, ¢) H (E, ¢oa). Since (¢oa(l, 0), ¢oa(O, 1)) = (~et(Q), the same
80 3 Modular Curves
action of a E GL 2 (7l./N7l.) induces an automorphism of Y(N) (and X(N))

regarded as schemes over 7l.[6~] (not over 7l.[6~' (N]), which coincides with the
Galois action (N r-t (~et(Q) on 7l.[(N]. It is easy to see that Q[(N](X(N)) = fi.N
in Corollary 2.41 and the action of GL 2 (7l./N7l.) introduced here coincides with
the identification Gal(fi.N/Q(J)) ~ Gal(7l./N7l.)/{±l} in Theorem 2.40.
3.2.6 Hasse Invariant
Let R be a ring of characteristic p and (E,w) be an elliptic curve over S =

Spec(R). On each affine open subset U = Spec(r(U, OE)) in E, the Frobenius
endomorphism x r-t x P induces a morphism Fabs : U -+ U. These glue each
other to the absolute Frobenius endomorphism Fabs : EjR -+ EjR. Note
here that Fabs acts nontrivially on the coefficient ring R. We can define the
relative Frobenius map E -+ E(p) = Ex S,Fabs S by Fabs x S f for the structure
morphism f : E -+ S. This relative Frobenius map is the classical map taking
homogeneous coordinates on E to their p-powers.
Let TEjs be the relative tangent bundle. By definition, its sections are
derivations; in particular, the space of global sections HO(E, TEjs) is the R-
dual of HO(E, [lEjS), and HO(E, TEjs) is spanned by the dual base T] =
T](w) of w. One can identify HO(E, TEjs) with the module of Os-derivations
Deros(OE,O,OS) (cf. [GME]l.5.1). For each derivation D of OE,O, by the
Leibniz formula, we have
Thus DP is again a derivation. The association D r-t DP induces an Fabs-linear

endomorphism F* ofTEjs. Then we define H(E,w) E R by F*T] = H(E,w)T].
Since T]()..w) = )..-IT](W), we see
H(E, )..w)T]()..w) = F*T]()..w) = F*()" -IT](W)) = ).. -P F*T](w)

= )..-PH(E,w)T](w) = ),,-PH(E,w) ..T]()..w) = )..l-PH(E,w)T]()..w).
Thus we get
H(E, )..w) = )..l-p H(E, w).
Then H is a modular form of weight p - 1 defined over lFp:
We compute H(Tate(q), wean) for Wean = if,

writing the coordinate of G m
as t (i.e., G m = Spec(7l.[t, rl])). The dual of if
is given by D = t~. The
action of F keeps D intact, because D(t) = t (so DP(t) = t). On the tangent
space, F acts as identity, and hence H(Tate(q),w ean ) = l.
An important fact (cf. [GME] Proposition 2.9.1) is:
H (E, w) =0 {==} E is su persingular.
This is because:
1. If Ejiffp is ordinary, then E[P] ~ jJ..p x (ZlpZ) over lFp;
2. jJ..p = Spec(lFp[t]/(t P - 1)) shares the tangent space with G mjIFp ' because
they are both of dimension 1 infinitesimally;
3. p2 = P up to units in the supersingular case.
The zero locus of a nonzero section of a line bundle is an effective divisor;
t
hence, for any algebraically closed field k of characteristic p for p N, the
points in Xl(N)(k) carrying supersingular elliptic curves are finitely many
and algebraic over lFp .
3.2.7 Igusa Curves
Let W = Zp and Wm = Wlpmw. Fix N with p tN. We have a lift of the

Hasse invariant in G p- 1(rl (1); Zp), which is the Eisenstein series [normalized
so that a(O, [) = 1. The standard Eisenstein series Ek of weight k on SL 2 (Z)
has the following q-expansion ([LFE] Chapter 5),
Define [(q) = ((2~p)Ep-1. By the Von Staut theorem, the q-expansion [(q) of
[ is congruent to 1 modulo p; so, [ mod p coincides with H. Let (E, ¢N) jM
be the genus 1 semi-stable curve (completed by appropriate Tate curves at
the cusps) over M = X 1 (N)jw. Let
Mm = Xl (N)jW= = X1(N) Xw Wm .
Define Sm C Mm by the maximal open subscheme of Mm on which [ is

invertible, and symbolically write Sm = Mm [i-]. Similarly we write So =
M [i-] for the maximal open subscheme of M on which [ is invertible. The
congruence [ == [' == H mod p tells us that [' = [ . (1 + n) for n nilpotent;
so, [ and [' are invertible at the same time. Thus the scheme Sm does not
depend on the choice of the lift [ as long as p is nilpotent in the base ring.
We write S= for the formal completion ~mSm of So along Sl.
Since we have defined X 1 (N) by Proj(G r1 (N)), the invertible sheaf I,Jd.k
(k > 0) associated with the kth graded piece is ample. To see for which
k, I,Jd.k becomes very ample, we recall that an invertible sheaf of degree ;:::
2g + lover a curve of genus 9 is very ample by the Riemann-Roch theorem
(see [GME] Proposition 2.1.4). Computing the genus of X 1 (N) (e.g., [GME]
Theorem 3.1.2), the invertible sheaf 1,Jd.7R corresponding to G k (r1 (N);R) is
very ample if k ;::: 2 and N ;::: 4 (or k > 2). Thus Sm is affine, and Sm =
Spec(Vm,o) for a Wm-flat algebra Vm,o.
82 3 Modular Curves
Let Wm-pALG be the category of p-adically complete Wm-algebras R

with p-adically continuous morphisms. When we write Woo, it means W =
¥!!!
mW m. Here the words "p-adically complete" mean that R = nRj pn R.¥!!!
We consider the functors £~rd,£'o;d : Wm-pALG -+ SETS given by
where P is a point of order pQ. For a given test triple (E, P, ¢N )/R, we have
a unique commutative diagram (up to isomorphisms)
E ~ E
1 1
Spec(R) -----+ YI (N)
and t(P) E (E[pQ]et - E[pQ-I]et) (R). The association (E, P, ¢N) / R rl t(P)
gives rise to the following isomorphism,
if m is finite. Here E[pQ]et is the maximal etale quotient of E[PQ]; thus,

writing E[PQ] as a relative affine spectrum Specs", (R) (cf. [GME] 1.5.4)
of a sheaf of Os", -algebras R, we have E[pQ]et = Specs", (Ret), where Ret
is the maximal subalgebra of R etale over Os", (see [T] (1.4)). We write
Tm,Q/ s'" = (E[pQ]et - E[pQ-I]et) / S",' which is an etale Galois covering with
the Galois group (7l,jpQ7l,y. Then the scheme Tm,Q represents the functors
£,~rd and £~rd over W m . The curve Tm,Q/w", is called the Igusa curve of level
NpQ. We also define a procurve Tm,oo/w", = ¥!!!QTm,Q which represents the
functor £g;;d over W m.
By a result of Igusa (e.g. [GME] 2.9.3), the curve TI,Q/s, (LV = 1,2, ... ,00)
is irreducible (which we prove in the following subsection).
3.2.8 Irreducibility of Igusa Curves
Since Tm,Q has the same topological space as TI,Q, we may assume m = 1
in order to prove the irreducibility of Tm,Q; namely, we prove the following
theorem.
Theorem 3.3 (Igusa) Let p be a prime outside Nand iFp be an algebraic
closure oJlF p • Then the Jgusa curve TQ/ JFp := TI,Q over Xl (N) /JFp is irreducible
Jar all LV.
This theorem was first shown by Igusa in [1] using the fact that the inertia
group at each supersingular point of Xl (N)/JF p in Gal(TQjXI(N)) is equal to
the full group (7l,jpQ7l,) x. We prove this theorem assuming p > 3, since we
have constructed modular curves only over Z[6~]. Our proof is new and can
be generalized easily to more general Shimura varieties (see Theorem 4.21,
Theorem 6.27, and Corollary 8.17).
Proof. We write Ta = Spec(Va) for an etale extension of Vo over lFp . For two
integers 0 < NIN' prime to p, Ta over X 1 (N') is a finite covering of Ta over
Xl (N), the irreducibility of the Igusa curve over Xl (N') implies that of the
Igusa curve over Xl (N). Thus we may assume that N is large so that Xl (N) is
smooth over Z[6~]. We may construct the Igusa tower T~ over S' = X(N)[i]
in exactly the same manner as Ta/X,(N) , replacing X 1 (N) by X(N). Again
Ta is a finite quotient of T~; so, we only need to prove the irreducibility of
T~. By Zariski's connectedness theorem ([ALG] III.11.5), the irreducibility
of the projective curve X(N) over Q combined with f*OX(N) = Z[(N, 6~]
for f : X(N) --+ Spec(Z[(N' 6~]) implies the irreducibility of Slover lFp (an
algebraic closure of lFp). We extend the scalar to lFp and again write Vo for
V1 ,0 ®IF p IFp. Then, Vo is an integral domain over IFp. Since T~ / S~ is etale, T~ is
a disjoint union of irreducible curves over S~/lFp. By the existence of the Tate
curve, the cusp 00 gives a point of T~, and there exists a unique irreducible
component Ca = Cr(N),a of T~ containing 00. The action of Z; on the level
p-structure permutes irreducible components of T~; so, all other irreducible
components are isomorphic to Ca. In this way, we get a tower of irreducible
curves
... --+ Ca --+ ... --+ Co = S~.
Since (Z/paz) x acts transitively on points of T~ over a given closed point of
S~, C a / S~ is a Galois covering with Galois group
Thus H = ¥!!! aHa is a subgroup of Z;. We need to show that H = Z;.

We fix an embedding ip : Q Y Qp for an algebraic closure Qp of Qp. Let
Q~r be the maximal unramified extension of Qp inside Qp, and write W for the
p-adic completion of the p-adic integer ring of Q~r. We write W = i:;; 1 (W); so,
W is a strict Henselization of Z(p) C Q, which is the localization in Q (not a
completion) at p of Z. Let Ro be the affine ring of So/wi so, Vo = Ro/pRo and
So = Spec(Ro) for the W-algebra Ro. The ring W[p,p""] is a valuation ring (no
longer discrete). We write Ra for the integral closure of Ro [p,p""] in X (N pa) / K
for the field of fractions J( of W[p,p"" ]. Each element f E Ra has its q-expansion
f(q) = Ln»-oo an(f)qn. We then define a valuation v = Va : Ra --+ Q by
va(f) = Infn ordp(ip(an(f))), where ordp is the valuation of W normalized so
that ordp(p) = 1. Obviously, by the same definition, this valuation extends
to v : J't --+ Q U {oo} for the modular function field J't in Theorem 2.40.
Note that Ro/pRo = Vo and Va is an extension of the valuation Vo to R a ·
Thus for the ideal Pa = {f E Ralva(f) > O} of R a , we have an embedding
Ra/Pa Y lFp(Ca ) by the q-expansion (irreducibility of Ca ). Let "'a be the
field of fractions of Ra/Pa. Since lFp(Coo ) is Galois over "'0 = lFp(Co), the
84 3 Modular Curves
Galois group H surjects down to Gal(K: oo / K:o) by the restriction map; so, we
need to show that Aut (K: oo / K:o) ::) Z;, where K:oo = U", K:",. Let V", be the
valuation ring of v'" in K(X(Np"')). By definition, we have K:", = V",/I,fJ", for
the maximal ideall,fJ", of V",. Note that Qj = Gal(Jt/Q(J)) = GL 2 (Z)/{±1} by
Theorem 2.40. By the description of the action of T( a) on Jt in Section 2.6, we
find that the action of (~ a~l ) with a E Z;is induced by the action ¢p I--t ¢p 0
(~ a~l) on level r(poo)-structures ¢p. By our choice, C", contains the infinity
cusp where the Tate curve Tate(q) /Z[[ql/ p "lJ has the level structure ¢can : !1p" x
Z/p"'Z -+ Tate(q)[p"'] given by ((, m) I--t ((qm/ p") E Gm(W[[ql/p"lJ)/qz. Thus
the action of a diagonal matrix (~ a~l ) brings the etale part ¢~~n = ¢canlz/p"z
to ¢~~n 0 a for a E Z;, which induces the action of a E Z;
on Voo = U", V",.
Hence we need to prove that D / I ::) Z; by a I--t (~ a~l ) for the decomposition
group D C Qj and the inertia group I C Qj of 1,fJ00/1,fJ0. In other words, we need
to show that for a diagonal matrix a E SL 2(Qp), T(a) preserves the valuation
v:Jt-+QU{oo}.
Let Dp be the decomposition subgroup in Aut(Jt) of v; that is,
Dp = {O" E Aut(Jt)Iv 00" = v}.

We note here that Dp is a closed subgroup of Aut(Jt). Indeed, by Hilbert's
theory of decomposition groups (e.g. [BCM] V.2.3), for any subfield e c Jt
over which Jt is algebraic, the intersection De = Dp n Aut(Jt/e) is compact
by the surjectivity of De onto Aut(K:Jt/K:e) for the residue fields K:Jt and K:e
of the valuation v restricted to Jt and e, respectively. Note here {Gal(Jt/e)h
for e finitely generated over Q with algebraic Jt/e gives a system of open
neighborhoods of the identity of Aut(Jt); so, Dp is a closed subgroup (see
Section 2.3 for the topology of the automorphism group of a field).
As we have seen in Section 2.6, the action of a rational upper triangular
matrix a = (o~) E GL 2 (Q)+ on f(q) is given by f I--t r("') = f 0 a which
sends Lm am(J)qm to Lm exp(27ri rr:/)am (J)qma/d. Thus T(a) preserves v.
In other words, Dp::) B(Q) = {(o~) E GL 2(Q)+la,b,d E Q}. Since U(Q) ~
Q is dense in U(A(oo») ~ A(oo) for the upper unipotent radical U, we find
U(A(oo») CDp.
Since SL 2(A(poo») in Aut(Jt) commutes with the diagonal torus T(Zp)
given by {( ~ a~l ) la E Z;}, the action of g E SL 2 (A (poo») can be lifted to an
action ofthe function field UN,,,, iFp(Cr(N),,,,); so, v is preserved by SL 2 (A (poo»)
(this point is addressed more carefully in the proof of Theorem 4.21). Since the
closure of SL 2(A(poo»)B(Q) in GL2(A)+/QxlR~ = Aut(Jt) (see Theorem 2.40)
is contained in the closed subgroup Dp and contains the diagonal torus T(Zp),
we conclude that T(Zp) preserves v. 0
The following corollary is called the q-expansion principle and is a consequence
of the irreducibility of T", combined with the fact that the local q-parameter
gives the uniformizing parameter at a cusp of X1(N) over Z[~].
Corollary 3.4 Let £ IF be a line bundle realized in the function field of TalFp .
Let s be a cusp and U be a Zariski open neighborhood of s E TalFp' If the q-
expansion at s of a section f E HO(U, 12) vanishes, then f vanishes on U.
3.2.9 p-Adic Elliptic Modular Forms
Since Sm is affine, Tm,a is also affine. We write Tm,a = Spec(Vm,a)' We have

a tower of Wm-fiat algebras:
Vm,o C Vm,l C ... C Vm,a C ....
These algebras are unramified over Vm,o and Gal(Vm,a/Vm,o) ~ (7l.,/pa7l.,) X •

By construction, we have Tm,oo = Spec(Vm,oo) for Vm,oo = Ua Vm,a' Write
simply PO'. for E(pa]et as a locally free etale group scheme over Sm. Over Vm,a,
we have a canonical isomorphism (the universal level p-structure)
lean = ¢P'" : 7l.,/pa7l., ~ PO'. = E(part.

We then define Vm,oo = U a Vm,a and
v= Vr,CN) = ~mVm,oo and V = Vr,CN) = ~mVm,oo.
The space Vr,cN) is the space of p-adic modular forms on rl(N). By taking
the Cartier dual of 7l.,/pa7l., '-+ E(pa], we may regard f E Vm,a as a function
of (E, ¢p : /-Lp'" '-+ E(pa], ¢N) satisfying the following conditions similar to
(Gl-3) for p-adically complete W-algebras R = ~nR/pnR.
(Gpl) (E,¢p,¢N)IR ~ (E',¢~,¢'tv)IR =? f(E,¢p,¢N) = f(E',¢~,¢'tv) E R;
(G p 2) If p : R --+ R' is a continuous W-algebra homomorphism, then
(G p 3) For all level N -structures ¢ N of type r 1 (N) on the Tate curve Tate (q),
f(Tate(q), ¢~an, ¢N) E W[[q1IN]].
Each element f E VrlCN)0wB for a W-algebra B = ~mB/pmB is a func-

tion of (E, ¢P' ¢N) I R satisfying the conditions (G p l-3) for p-adically complete
B-algebras R (replacing W in the conditions by B).
Again we note the consequence of the irreducibility of the Igusa curve.
Corollary 3.5 (q-Expansion Principle) If the q-expansion at the infin-
ity cusp of a p-adic modular form f over a p-adically complete W -algebra
B vanishes, f itself vanishes; that is, f(E, ¢P' ¢N) = 0 for all test objects
(E, ¢p, ¢N) I R defined over all p-adically complete B-algebras R.
86 3 Modular Curves
3.3 p-Ordinary Elliptic Modular Forms
We have a p-divisible module VrdN) on which Gal(Vm,oo/Vm,o) = Z;

= T(Zp)
acts continuously. Here T = G m . We construct a projector e acting on V out of
the Hecke operator U(p) commuting with the action of Z;
= Gal(Vm,oo/Vm,o).
The important features of e are
• e = lim n -+ cxo U(p)n! (we recall the definition of U(p) in Section 3.3.3);
• v ord = eV has the Pontryagin dual which is projective over W[[Gm(Zp)]J;
• For any k 2': 3, there is a canonical isomorphism
vord[k] "'" eHo(So,(;,l Q?illp ) = eHo(M,i,,:l Q?illp ) (llp = Qp/Zp),
where V[k] = {j E VlzJ = zk J Vz E Z;}.

Here is how to prove the convergence of the projector e = lim n -+ cxo U (p )n!.
For any finite p-torsion module X with an endomorphism u, the power un!
stabilizes as n grows, giving an idempotent e u in the endomorphism algebra
End(X), since End(X) is a finite ring. For a p-divisible module X with an
endomorphism u, we assume that X = lli¥~EIX" for finite modules Xl stable
under u. Then the endomorphism eu = lim n -+ cxo un! is well-defined over X,
for all i and hence over X. For any compact p-profinite module X with a
continuous endomorphism u, we take the Pontryagin dual pair (X*, u*) which
is p-divisible (cf. [FAN] Chapters 1 and 3). Thus eu. exists on X*. Then
eu = lim n -+ cxo un! (which is the dual of eu ') exists as an endomorphism of X.
3.3.1 Axiomatic'Ireatment
Let r,dk = O(k) for the embedding of X1(N) = PrO~dN)) into the pro-
jective space. Thus r,dk is the quasi-coherent sheaf G rdN) (k) associated with
the graded piece GrdN)(k) ([GME] 1.3.3). Then r,dk = r,d®k. Computing the
genus of Xl(N), the Riemann-Roch theorem tells us that r,dk is very ample if
k 2': 3 (see [GME] Proposition 2.1.4 and Theorem 3.1.2). Therefore r,dk is the
pull back of O( k) of the target projective space.
Let (E,cPN'w) be the universal elliptic curve over Yl(N). For each triple
(E, ¢N, w) defined over R (called a test object), we have a unique morphism [ :
Spec(R) -t Y1(N) such that [*(E,cPN'w) = (E,¢N,W). For each section J E
HO(Y1(N),r,dk), we define [*J = J(E,¢N,W)w®k. The function (E,¢N,W)-t
J(E,¢N,W) satisfies (GO-2). The condition (G3) ensures that J extends to
Xl (N). This shows
HO (X 1 (N) / R, r,dk) "'" G dr1 (N); R) (canonically) (3.1)
for all Z[ 67v ]-algebras R.

Let (E, cP p , cPN) be the universal elliptic curve over Sm. Pick a section
J E HO (Sm, r,dk). Since /Lp= carries a canonical differential Wean = '¥'
writing
3.3 p-Ordinary Elliptic Modular Forms 87
/-LpG = Spec(Z[t]/(t pD -1)), we may regard f as a function of (E,rPp,rPN)

by f(E, rPp, rPN) = f(E, rPN, rPp,*wcan ). For each (E, rPP' rPN) E E:;;d(R) for
a Wm-algebra R, we have a unique morphism L : Spec(R) ~ Tm,oo =
Spec(Vm,CXJ) such that (E,rPp,rPN) = L*(E,rpp,rpN). Then L*f is just a
function on f(E,rPp,rPN) such that f(E,z-lrPp,rPN) = zkf(E,rPp,rPN) for
z E Gal(Vm,oo/Vm,o) = Z;. We recall that the action of T(Zp) = Z;
is nor-
malized so that z E Z; acts on the dual identification rP~t : E[poo]et ~ r!Jp/Zp
of rPp : /-Lpoo Y E[pOO] via the multiplication by z; so, z acts on rPp by
rPp r l rPp 0 Z-l. In other words, we have z . f(E, rPP' rPN) = f(E, z-lrPp, rPN) =
zk f(E, rPp, rPN). This shows that
Vm,CXJ[k] = HO(Sm,r"i)
and V[k] = HO(SCXJ/w,i:'d. k I8i 1I'p) = HO(Soo/w,i:'d. k ) I8i 1I'p, (3.2)
where 1I'p = r!Jp/Zp. The last identity follows from the affineness of So. When
we generalize this point to classical groups bigger than SL(2)/Q, the affineness
is no longer true; so, we need to work out this point (the base change property)
carefully: how to bring 1I'p inside the cohomology group. In the elliptic modular
case, simply by the affineness of So, (3.2) shows that V[k] is p-divisible, and
its direct summand eV[k] is also p-divisible.
We consider the following condition:
(F) corank w eV[k] = rankw Hom(eV[k], 1I'p) is finite for an integer k.
In practice, this condition is often proven by showing
where
The left-hand side (LHS) of (3.3) is p-divisible, since So is affine. The right-
hand side (RHS) is of finite corank since Xl(N) is projective. Thus eV[k] is
p-divisible of finite corank.
Decompose Z; = r T x Ll for a p-profinite group r T and a prime-to-p
finite group Ll. For simplicity suppose that p > 2. Then rT is isomorphic to
Zp and for its generator " we have W[[rTlJ ~ W[[XlJ = A via, r l 1 + X
(i.e.,,S ~ (1 + X)S = L~o G)XJ), and W[[Z;lJ = A[Ll]. Let vord be the
Pontryagin dual module of eV.
See [FAN] Chapters 1 and 3 for generality on topological groups (in-
cluding Pontryagin duality and profinite groups). My books, [MFG] Chap-
ter 2 and [LFE] Chapter 8, also have some short exposition on this sub-
ject. The Pontryagin dual for p-torsion modules X (equipped with discrete
topology) is given by X* Hom con t(X,1I'p) for 1I'p = r!Jp/Zp with dis-
crete topology. Writing X = ~'EIX, for finite submodules X" we have
88 3 Modular Curves
X· = ~"nHom(X"p-nZ/Z), and the topology on X· is the profinite topol-

ogy with respect to the finite discrete groups Hom(X" p-nZ/Z). Then X· is
a compact profinite group. The duality is perfect; so, if X '-t Y -» Z is an
exact sequence of p-torsion modules, Z· '-t y. -» X* is an exact sequence of
profinite groups. In particular, if X is p-divisible and p-torsion, the multipli-
cation by p is surjective; so, by duality, multiplication by p on X· is injective,
which shows that X· is torsion-free if X is p-divisible and p-torsion.
Suppose that a profinite (Noetherian) W-algebra R acts on X continu-
ously. For any closed ideal a c R, Pontryagin's dual of X[a] = {x E Xlax = O}
is given by X· I aX·, where R acts on X· continuously by the pullback (dual)
action. To see this, we choose a finite set of generators (aI, ... , a r ) of a and
1ook .mto the exact sequence: 0 -+ X [a] -+ X X1-4(alx, ... ,a r x) ) X r. T a k'mg t he
dual, we find that X·
r (Xl, ... ,X r )1-4L:, a,x, I
) X· -+ X· aX· -+ 0 is exact, whIch
.
shows X[a]· = X·laX·.
For any (continuous) W-algebra homomorphism x: R -+ W, applying the
above fact to a = Ker(x), the Pontryagin dual of the x-eigen subspace
X[x] = {x E Xlax = x(a)x for all a E R}
is given by the maximal x-eigen quotient X· I aX· for the x-augmentation

ideal a = Ker(x) = (a - x(a))aER C R.
Let Z = Hom(Ll, W X ), and write V[X] for the x-eigen subspace of X E Z:
V[x] = {v E VI<5v = X(<5)v for all <5 ELl}.
If eV[k] is of finite corank for one k, taking X E Z with X = kiLl, we have

V[k] = V[X][k], and then by duality, we have
vo rd ®W[[T(Zp)]],k W = vord[xl/(X + 1 - ,/)vord[X]

= Homcont(Vord[X, (X +1- '/)]' ll'p)
~ Homzp(~mH~rd(Xl(N)/w>n'~~w,J, ~mp-mZ/Z)
~ ~mHomzp(H~rd(Xl(N)/w>n'~~w,J,p-mZ/Z)
~ ~mHomw(H~rd(Xl(N)/w>n'~~w>n)' W m)
= Homw(H~rd(XdN),~k), W) ~ Homw(Gkrd(r1(N); W), W). (3.4)
In the middle equality, we have assumed (3.3). Here the subscript or super-
script ord indicates the image of e.
Write Z = Hom(Ll, WX). Note here that each element X E Z is induced
by (infinitely many) positive k E X(T). Decompose vo rd by the character of
Ll as follows,
Vo rd = EB vord[X]·
xELl
This is possible as we assumed that .1 has order prime to p. If z f-t zk coincides

with X on .1, then vord[X] 0A,k W = v ord 0W[[T(Zp)]],k W. By Nakayama's
lemma, we have a surjective homomorphism of A-modules:
where s = s(X) = corankweV[k]. If (F) holds for one k, it holds for all
K, E X(T) inducing X, and 7r has to be an isomorphism for the following reason:
The number s is the minimum number of generators of v ord [X]0 A,I< W over W.
We know that this module is W-free, because its dual V[K,] is p-divisible; so, it
is free ofrank s. The morphism 7r induces an isomorphism modulo (1+X) - "(I<
for all K, inducing X. Then Ker(7r) C nl< Ker(7r mod (1 + X - "(1<)) = 0, and
we get
Theorem 3.6 Suppose that (F) holds for one k. Then v ord is A-free of finite
rank, and if (3.3) holds for K, E X(T) with K,I~ = kl~, then
v ord 0 w [[z,;JJ,1< W ~ Homw(C~rd(rl(N); W), W).
3.3.2 Bounding the p-Ordinary Rank
Since So is affine, we have HO(Sl,,,,l) = HO(So/w,(",l) 0w WI. If we have

a sequence 11,12"'" IJ of linearly independent sections in H~rd(Sl,f,)l), we
can lift them to Ii E HO(So/w, ~k) so that Ii = (it mod p). Since So = M[·tJ,
we have
HO(SO/W,~k) = ~nHO(M/w,~k+n(p-l))/£n.
Thus £m ft E HO(M/w,~k+m(p-l)) for all i = 1, ... ,j for sufficiently large m,

and they are linearly independent. We now assume
(C) e(£J) = £(eJ) for all f E HO(Sl,~k).
By this, e(£mft) are still linearly independent in H~rd(M/K,~k+m(p-l)); so,
we have dimK H~rd(M/K,~k+m(p-l)) 2: j. If rank w1 H~rd(Sl,~k) is infinite,
the rank of H~rd(M/K,~k+m(p-l)) grows as m grows. Thus the condition (F)
for all k follows from
(F') dimKCrd(rl(N);Qip) is bounded (independently of k).
Indeed, the Eichler-Shimura isomorphism combined with a calculation of
group cohomology H~rd (rl (N), L( k; K)) proves much stronger
(E) If k 2: 3, dimK Ckrd(rl(N); K) depends only on k mod p - 1
for the projector e. Actually we can ease the condition k 2: 3 to k 2: 2 if we
replace Crd(rl (N); K) by Crd(rl (N) n ro(p); K) in (E) (see Section 4.3 in
the text and [LFE] 7.2).
90 3 Modular Curves
3.3.3 p-Ordinary Projector
Let (E,¢p,¢N)jR be a test object. Each subgroup C of order p outside the

image of ¢p is etale; so, we can think of the quotient (E IC, ¢P' ¢N) defined
over a (possibly radiciel) finite extension R' of R. We define the U(p) operator
acting on a (p-adic or classical) geometric modular form f by
The sum Lc in the above definition has to be understood as a trace from R'
to R (see (4.18)) if R is not a Q-algebra (this subtle point is discussed in detail
later in Section 8.3.1). From the definition, as long as the prime p is invertible
in the base ring R, we can verify the conditions (GO-3) (or (G p l-3)) for
fIU(p). Computing the q-expansion, we know a(n, fIU(p)) = a(np, f). Then
by the q-expansion principle, the operator U(p) is well-defined p-integrally
over W even if the definition of the operator involves division by the prime p.
Since £ == 1 mod p, we confirm (C). Let
= EB Ck(r1(N); R).
00
Cr,(N)(R)
k=O
One can prove the p-adic density of D = Cr'(N)(W)[~] n V in V (see Corol-

lary 8.4 in the text and [GME] Corollary 3.2.4). Using this fact, we can
show that we can write V as a union of finite modules stable under U(p);
so, e = limn-+oo U(p)n! exists. The final result is as follows.
Theorem 3.7 (VeT) We have
if k ?: 3,
if k = 2.
Similarly, if we write vg~~p for the subspace of cusp forms in v ord and write
v
Vc':s~ for its Pontryagin dual (i.e., the cuspidal quotient of ord ), the above
v
result is still valid replacing ord and cord
k v
by cusp
ord and Bard
k'
3.3.4 Families of p-Ordinary Modular Forms
r
Recall the decomposition T(7!..p) = 7!..; = T x L1 for a finite group L1 and
rT ~ 7!..p- We have chosen a generator "I E r T so that 7!..p :3 s H "Is E r T
induces the isomorphism 7!..p ~ r T and A = W[[rTJJ ~ W[[X]] by "I +-+ 1 + X.
Let a( n) : v ord --+ 1I'p be the linear map associating with f its coefficient
of qn in the q-expansion; so, a(n) is in the dual v ord . We now consider
With each tJ5 E C(X; A), we associate its q-expansion
L
00
tJ5(q) = a(n; tJ5)qn E A[[q]] with a(n, tJ5) = tJ5(a(n)) EA.

n=O
Theorem 3.8 For each k ::::: 2, we have
(1) C(X;A) is A-free offinite rank;
(2) C(X; A) 0A,k W ~ Crd(r1(N) n rO(p) , xw-k; W), where w is the Te-
ichmiiller character modulo p;
(3) the above identification is induced by tJ5 f--} Ln a( n, tJ5) (-l- 1 )qn E W[[q]].
Proof. The A-freeness follows from the freeness of vord[X]. We only prove the
assertion when k induces X on L1; so, xw-k is trivial. We have
C(x; A) Q9A,k W ~ Hom A(V0rd [X], A) 0A,k W

~ Homw(Vo rd [X]0A,k W, W) ~ Homw(Homw(Crd(r1(N); W), W))
= Crd(r1(N); W) ~ Crd(r1(N) n ro(p); W).
Here is the explanation of how to obtain the last isomorphism in the above
sequence of isomorphisms. As is well known (see (4.59)), the Hecke operator
T(p) acts on f E Ck(r1(N);W) by
a(n, fIT(p)) = a(pn, J) + pk-1a(!!., J),

p
where we put a(~, J) = 0 if ~ ~ Z. Thus in W[[q]], T(p) == U(p) mod p if

k ::::: 2, and the natural inclusion Ck(r1(N); W) y Ck(r1(N) n ro(p); W)
induces an inclusion:
By the theory of primitive forms ([MFM] Theorem 4.6.17; see also [H85] Sec-
tion 3), this inclusion is surjective if k ::::: 3. We leave the verification of asser-
tion (3) to the reader. 0
For a given element tJ5 E C(X; A), we get a family of p-ordinary elliptic modular
forms {tJ5( -yk - 1) h>2 whose q-expansion coefficients depend (p-adic) analyt-
ically on the weight k E Homcont(T(Zp),Z;). Each element tJ5 E C(X;A)
is called a p-ordinary A -adic form, and {tJ5(rk - 1) h?:2 is called the p-adic
analytic family of modular forms associated with a A-adic form tJ5.
There is a version of this for ro(N) and cusp forms. For k ::::: 2, we have
1. F(ro(N), x; A) is A-free of finite rank;
2. F(ro(N), X; A) 0A,k W ~ Fkrd(ro(Np), xw-k; W);
3. the above identification is induced by tJ5 f--} Ln a( n, tJ5) ("(k -1 )qn E W[[q]] ,
where X is a character modulo Np and F =C and S (see [GME] Chapter 3
for a proof).
92 3 Modular Curves
3.4 Elliptic A-Adic Forms and p-Adic L-functions
We describe how to view A-adic forms as p-adic modular forms defined over
A. Once this is done, we can evaluate A-adic forms at elliptic curves, which
gives us a convenient method of constructing and analyzing p-adic L-functions.
Then, we give a short account of the A-adic Eisenstein series and examples
of A-adic L-functions. All the arguments presented here can be generalized to
the Hilbert modular case (see Chapter 4), the Siegel-Hilbert modular case,
and the quasi-split unitary case (see Chapter 8).
3.4.1 Generality of A-Adie Forms
For simplicity, we assume that p > 2 and only consider the A-adic forms
of level poo; thus, we have N = 1, and rT = 1 + pZp is torsion-free. Let
A = Zp[[TlJ. In the previous section, we have introduced the space G(X; A) of
p-ordinary A-adic forms, which is a free A-module of finite rank with
for all k ;::: 2. Here k : A --+ Zp is the evaluation at "(k - 1 of the power series.
If we identify A with the Iwasawa algebra Zp[[rTll by sending the generator
"( of rT to 1 + T, k is induced by the character rT :3 z H zk E 71.,;.
We write G(A) for the A-module made of formal q-expansions
q> = L a(n; q»(T)qn E A[[qlJ,

n2:0
such that tP("(k-1) E V[kl for infinitely many k. We have EBx G(X; A) c G(A),
where X runs over (actually even) powers of Teichmiiller characters.
We now consider the space of p-adic modular forms VI A over A of level pDQ.
In other words, we make a base change Tm,nlZ p to Tm,nlA = Tm,nlZ p xZp A
and consider p-adic modular forms over A. The functions in VIA = V®zpA
classify couples (E, ¢ : J.lP= Y E[POO]) I R defined over p-adic A-algebras R.
This J E VIA is a functorial rule assigning the value J(E,¢) E R for each
couple (E, ¢)/R as above.
This space has two A-module structures: one coming from the base ring A
and another coming from the action of Gal(Tm,oo/Sm) = 71.,; by the diamond
operators (z). Let v : rT --+ AX be the universal character given by v(z) =
[zl E rT. Then we can define
G(A) = {J E VIAIJI(z) = v(z)J Vz E rT}' (3.5)
Each q> E G(A) has a q-expansion at 00: q>(T, q) = Ln>o a(n; q>)(T)qn. By
definition, we have a natural map VI A 0 A,s Zp --+ Vlzp for each s : A --+ Zp
taking tP(T) to q>("(s - 1) for s E Zp. Here the tensor product is taken using
3.4 Elliptic il-Adic Forms and p-Adic L-functions 93
a A-module structure induced by the diamond operators. Since on G(A), the

two A-module structures coincide, the map is injective by the q-expansion
principle. This map brings ifJ E G(A) to a p-adic modular form of weight s.
Therefore, ifJ is a A-adic form.
Conversely, starting from a A-adic form ifJ, we regard ifJ as a bounded mea-
sure on r T having values in Vlzp' Here we use the fact that Zp[[rT ]] is canon-
ically isomorphic to the measure space on rT by a(T) >---+ J xSda = ab s - 1)
(see [LFE] Sections 3.3-5). Thus, writing C(X, Y) for the space of continuous
functions X -4 Y for two topological spaces X and Y, we may regard ifJ as a
Zp-linear map: f >---+ J fdifJ of C(rT , Zp) into Vlzp' Then for each test object
(E, ¢) I R for a p-adic A-algebra R, regarding R as a p-adic Zp-algebra, we can
evaluate J fdifJ E Vlzp at (E'¢)IR, getting a bounded Zp-linear form from
the space C(rT , Zp) into R, which we write ifJ(E, ¢)(T) E R®zp A = R[[T]].
Since R is already a A-algebra, the A-module structure A ®R ----t R given by
A ® l' = Ar induces a surjective algebra homomorphism m : R ®zp A -» R. We
then define ifJ(E, ¢) by m(ifJ(E, ¢)(T)). Then the assignment: (E, ¢) >---+ ifJ(E, ¢)
satisfies the axioms (G p l-3) governing the p-adic modular forms defined over
A. It is easy to check that this p-adic modular form is in G(A) having the
same q-expansion at (X) as ifJ. Thus we have found:
Theorem 3.9 The subspace G(A) C VIA is isomorphic to the space G(A) of
all A-adic forms via q-expansion at the cusp 00. In particular, we have
E9 G(X; A) ~ e(G(A))
x
for the p-ordinary projector e : VI A -» Vr~·d.

Let (E, w) IW be an elliptic curve with complex multiplication by an imag-
inary quadratic field M = Q[ V- D]. We suppose that w is defined over
W = W n ij fixing an embedding ip : ij Y Qp. Suppose that p splits in
M and write p = pp. Under this assumption, E is p-ordinary. We may assume
that W is the p-adic integer ring of the p-adic completion of the maximal
unramified extension of Qp (so W = W(Fp), the ring of Witt vectors with
coefficients in Fp; see [BCM] IX.l) and that E[pOO] is the etale part of E[pOO]
over W. Thus we have ¢ : {Lp= ~ E[pOO]. In this way, we can evaluate a given
A-adic form ifJ at (E, ¢).
Corollary 3.10 IfifJ(E,¢) = 0 for infinitely many distinct Es with complex
multiplication, then ifJ = O. There exist finitely many elliptic curves (Ei' ¢2) IW
such that any given linear form G(x; A) ----t A is a A-linear combination of
evaluation at (E2' ¢2) .
IfifJ E G(X;A) and furthermore, if f = ifJb k -1) E Gdro(p),Xw-k;W), then
f (E, w) E W. The morphism ¢ : {Lp= Y E induces a canonical differential
Wean = ¢* 1f.Then w = flpwcan' and we have a result of Katz:
94 3 Modular Curves
wean) = cJ>(E, ¢ Kl
f( E ,w ) = f(E,[lk - 1) EWe ffl'l
(EQ1)
[lk "",.
p p
We may assume that E(C) = Cj R for the integer ring R of M. Let w be

the variable of C. Then dw induces a canonical differential Woo on E(C),
w = [loow oo , and we get a result of Shimura (actually dating back to a work
of Hurwitz [Hz] for elliptic modular forms):
f(E,w) = f(~:oo) = f(E~~ean) EWe Q. (EQ2)

00 p
The lattice 0 = HI (E, 71,) c C is generated over 0 by a single element "'( = 1

and [loo = I, w, because I, dw = 1.
3.4.2 Some p-Adic L-Functions
In this subsection, we describe very briefly some p-adic L-functions whose con-
struction stems from the theory of A-adic forms. More general and thorough
discussion can be found in [LFE] Chapters 7 and 10 and [SGL].
For simplicity, we assume that p > 2 and only consider the A-adic Eisen-
stein series of level poo. Let us fix an even power X = w a of the Teichmiiller
character. For simplicity, we choose a¢O mod p - 1; so, X is nontrivial. We
consider the Kubota-Leopoldt p-adic L-function -~Lp(1- s, X) = ao("'(8 - 1)
("'( = 1 + p) with ao E 71,p[[T]] (e.g., [LFE] Sections 3.5 and 4.4). We have, for
positive integers k,
Then we define an element Ex E G(X; A) by
a(n,Ex) = L X(d)d- l (l +T)lOg(d)/log(r) and a(O, Ex) = ao(T).

O<dln,pfd
We want to relate Ex ("'(k - 1) to the following classical Eisenstein series:

1 1
Ek(E,w) =-2 L
(m,n)i(O,O)
(
mWI
+ nW2 )k'
where (E,w)/c corresponds to the lattice L = 71,wI + 71,w2 by Weierstrass'

theory ([GME] Section 2.4), that is, E(C) = Cj Land w = dw for the variable
wE C. As is well known ([IAT] (2.2.1)), for even k > 2,
for c(k) = (2:J]W' This shows that if w k = X, then

3.4 Elliptic A-Adic Forms and p-Adic L-functions 95
If we take the elliptic curve (E, w) defined by y2 = 4x 3 - x (which is equiv-

alent to y2 = 1 - x4: Gauss' lemniscate) with w = d;, then it has complex
multiplication by Q[A] (A: (x, y) f-7 (Ax, y) for the model given by
y2 = 1 - X4 and for k > 2 with w k = X, we have
~E (E ) = L(k, Ak)
2 k ,w [2k'
00
where Ak is a Hecke character of conductor 1 such that Ak (a) = a k . Since we

can find, E SL 2(Z) (g n
SL 2 (Z) such that (W1,W2h is a base of p, we re-
discover p-adic interpolation of Hurwitz numbers (done by many outstanding
mathematicians) :
Ex(E, w)(rk - 1) =2 (k)-l L(k, Ad - pk-1 Ak(fJ)p-k L(k, Ak)

[2k c [2k
p 00
=2c(k)-1(1- p-1 Ak(fJ)) L(~kAk)

00
This is a A-adic version of Katz's way of constructing the p-adic Hecke L-

function: Lp(s) = Ex(E,w)(rs - 1) ([K3], [HT], and [SGL] Chapter 8).
A p-adic Rankin product can be constructed similarly. Let tf> be a nor-
malized Hecke eigenform in G(X; A). Write lL for the field of fractions of A.
As shown in Corollary 5.4, the Hecke algebra acts semi-simply on G(X; A);
so, we can uniquely decompose G(X; A) ®A lL = lLtf> EB X as Hecke mod-
ules. Let £ : G(X; A) --+ lL be the linear form defined by !Jt = £(!Jt)tf> + x
for x E X. We consider two copies of A, say Zp[[S]] and Zp[[T]]. Take two
Hecke eigenforms tf> E G(X;Zp[[T]]) and!Jt E G(7j;;Zp[[S]]). Linearly extend £
to G(X; Zp[[T]]) 0z p Zp[[S]] --+ lL0zp Zp[[S]]. We define
Lp(S, T) = £( e(!Jt(S)Ex'l/J-1 ((1 + T) (1 + S) -1 - 1))),
where e : V/zp[[Tll --+ V/'i:llTll is the p-ordinary projector. Then we see
!Jt(S)Ex'l/J-1 (~ : ~ - 1)ls=,m-1,T=,k-1 = !Jt(rm - l)Ex'l/J-' (rk-m - 1).
Thus !Jt(rm - l)Ex'l/J-' (r-m(1 + T) - 1) E G(X; Zp[[T]]) and hence the p-adic
L-value Lp(rm _l"k - 1) is the coefficient of w(rm - l)Ek- m in tf>(rk - 1)
for a suitable Eisenstein series Ek-m of weight k - m. As is shown by Shimura
[Sh7], this coefficient can be computed by the Rankin product value
D(k - 1, tf>(rk - 1), !Jt(rm - 1))

(k > m)
(tf>(rk - 1), tf>(rk - 1))
96 3 Modular Curves
for the Petersson inner product ( , ) up to an explicit constant; so, Lp gives

p-adic interpolation of the Rankin product. For an explicit evaluation formula
for Lp(S, T), see [LFE] Chapters 7 and 10 and [SGL] Chapter 6.
We summarize the basic ingredients we have used in the construction of
p-adic L-functions and in the study of A-adic forms:
1. The vertical control theorem;
2. The q-expansion principle (irreducibility of the Igusa tower).
We generalize these points to more general automorphic forms in the following
chapters.
4
Hilbert Modular Varieties
We present here a general theory of the Hilbert modular varieties together

with a sketch of a proof of the geometric reciprocity laws and the vertical
control theorems, assuming some of the results which are proven for more
general reductive groups G in later chapters.
We start with an affine group scheme H over a ring R'. For a subalgebra
R of R', if the covariant functor C f-t H(C Q9R R') defined on the category of
R-algebras is representable by a scheme H/R' we write H/R = ResR'/RH and
call it the Weil restriction of H with respect to R' I R (this is not changing
the base ring of H/R' to the subalgebra R). Since C f-t H(CQ9RR') is a group
functor, Res R' / RH is a group scheme if it exists. If R' IRis locally free of
finite rank, an affine group scheme ResR' / RH always exists (see [NMD] 7.6,
Theorem 4).
To connect with what we have said in the introduction (Chapter 1), we
describe briefly the algebraic groups G, G 1 and their torus T which gives the
notion of weight for Hilbert modular forms. Hilbert modular forms are defined
with respect to a fixed base totally real field F with discriminant d(F). We
always assume in this section that the fixed prime p is un ramified in F IQ; so,
p f d(F). Write 0 = OF for the integer ring of F. Our classical groups are given
by G = Reso/zGL(2), G 1 = Reso/zSL(2), and PG = Gad = Reso/zPGL(2).
In other words, for commutative algebras R, we define (as functors from al-
gebras into groups) G(R) = GL 2 (R Q9z 0), G1(R) = SL 2 (R Q9z 0), and
so on. The torus T is the subgroup made up of diagonal matrices in G 1 ;
so, T(R) = (R Q9z O)X, and hence T(2p) = 0; for the p-adic completion
Op = ~nOlpno. Writing 1 for the set of all embeddings of F into Q, we
may identify X(T) with the formal free module 2[1] generated by elements
of 1, associating a character (x f-t xk = Ilo.a(x)kO") with k = Lakaa E 2[1].
Thus the weight of Hilbert modular forms is given by a g-tuple k of integers
for 9 = [F : Q].
We usually use the symbol PG for G I Z in place of Gad in order to avoid
confusion with the adjoint group of the derived semi-simple group G 1 of G,

98 4 Hilbert Modular Varieties
although in the description of Shimura varieties, we use Gad, which is more

common in papers treating Shimura varieties.
Here is some more notation generally used in this chapter. The real Lie
group G(JR) is isomorphic to GL2(JR)I by an isomorphism sending a E G(Q) =
GL 2 (F) to a tuple (a")"EI of conjugates. The identity-connected component
G(JR)+ is made up of g-tuples of matrices (a")"EI with det(a,,) > 0 for all
a E I. We put G(Q)+ = G(JR)+ n G(Q). We later define in Section 4.2.2 a
slightly different group G(Q)+ for more general reductive groups G, but in
this Hilbert modular case, we actually have G(Q)+ = G(Q)+; so we use the
symbol G(Q)+ instead of G(Q)+.
For any lattice a c F, we define a* = {x E FITtF/IQ(ax) C Z} and ,,-1 =
0* (so, " is the absolute different of F /Q). As before, W is a discrete valuation
ring inside Q (on which the embedding ip : ij Y Qp is p-adically continuous)
and W = ~nWn for Wn = W/pnW, which is the p-adic completion of W.
4.1 Hilbert-Blumenthal Moduli

We first describe basic definitions and properties of abelian varieties with real
multiplication (AVRM), which are also called by a different name, Hilbert-
Blumenthal abelian schemes, and are a direct generalization of elliptic curves,
because their Tate modules supply us with p-adic Galois representations into
GL(2). Then we describe some features of the moduli of AVRM necessary to
prove VCT, putting the construction of the moduli scheme off to Remark 6.21
and Section 7.1.
4.1.1 Abelian Variety with Real Multiplication
Let A/ s be an abelian scheme. Here a group scheme over a base scheme S

is called an abelian scheme if it is a proper smooth geometrically irreducible
group scheme over S (the words "geometrically irreducible" mean that its
fiber at every geometric point of S is irreducible). A one-dimensional abelian
scheme is an elliptic curve. We consider the following Picard functor from
s'
SCH/ s into SETS: PiCA/S(S') = Pic(As' )/7r Pic(S') = Ker(Os') for 7r :
A ---+ S. If f : A ---+ Y is a morphism of abelian schemes, £ f--t f* £ for an
invertible sheaf £ induces the dual morphism t f : Pic y / s ---+ PicA/So Thus
Pic A/ S is a contravariant functor. It is known that PiCA/S is represented by
a (locally Noetherian) reduced group scheme ([FGA] Exp. 232, 1961/62, or
[NMD] Chapter 8, or [ABV] Section 13, or [DAV] 1.9). Let tA be the identity-
connected component Pic~/s of the group scheme representing PicA/So Then
t A/ s is an abelian scheme (see [ABV] Section 13). We write t f again for the
dual morphism t f : ty ---+ t A of a morphism f : A ---+ y.
Let F /Q be a totally real finite extension unramified at the fixed prime p.
Write 0 for the integer ring of F. We consider the following fibered category
AF of abelian schemes over the category of schemes. Here a fibered category C
4.1 Hilbert-Blumenthal Moduli 99
over a base category Co (cf. [SCA] VI) means that we have a specified covariant
functor (fiber functor) C ---t Co. An object of AF is a triple (A, A, L)/S, where
(rm1) L = LA : 0 y End(A/s) is an embedding of algebras taking the
identity to the identity;
(rm2) A is an O-linear symmetric isogeny A : A ---t t A induced by an ample
line bundle fiber-by-fiber geometrically (see [GIT] 6.2). Here A is called
symmetric if A = t A. Such a A is called a polarization of A;
(rm3) The image of LA is stable under the Rosati involution on the endo-
morphism algebra End(A) I8Iz Q: a H a* = A-Iota 0 A;
(rm4) As 0 I8Iz Os-modules, we have an isomorphism: Lie(A) ~ 0 I8Iz Os
(B 1f* (.flA/ s) ~ ll-1 I8Iz Os with 1f : A ---t S for the absolute different II
of F) locally under the Zariski topology of S, where the sheaf Lie(A)
of Lie algebras of A (i.e., the direct image of the tangent bundle over
A/ s ) is an O-module by the action induced from L.
The fiber functor is given by AF 3 (A, A, L)/S H S E SCH.
Here is a more detailed description of the above four conditions. Either by
localizing at a rational prime or by extending the scalar to an etale extension,
any projective O-module becomes free; so, we often pretend that the isomor-
phism in (rm4) is global, although Lie(A) is required to be only a locally
free module over (0 I8Iz Os) of rank 1. Since an O-module free of finite rank
is determined by the characteristic polynomials in Os[t] of multiplication by
a E 0, (rm4) is also equivalent to
(det) The characteristic polynomial of each a E 0 on Lie(A) over Os is
given by the image ofI1a(t-o-(a)) E ;E[t] in Os[t], where (j runs over
all embeddings of F into ij.
A morphism f : (A, A, L)/S ---t (A', A', L')/s of AF is an O-linear morphism
f : A/ s ---t A is of abelian schemes over S with A = 1* A' := t fOA' 0 f.
A polarization is an O-linear isogeny A = AL : A ---t t A induced etale
locally by a symmetric line bundle L / A; that is, we have (-1) * L ~ Land
AL(X) = T;(L) 181 L- 1 , where Tx(Y) = x + y. This definition is equivalent to
Mumford's definition ([CIT] 6.3) requiring that A be induced by a symmetric
line bundle LS/As for each geometric point s E S (see [DAV] 1.1.6). Indeed, for
the universal Poincare bundle P on A Xs t A/A, 2A is associated with (1 x A)*P
globally over S (see [GIT] Proposition 6.10).
The morphism A is an isogeny if and only if L is ample (cf. [ABV] Section
6). We have AUg)U = AL + AU, and the monoid of Hom(A, t A) generated by
polarizations forms a cone P(A). If A : A ---t t A is a polarization, Ker(A) is
given by A[c 1 ] for an integral ideal c- 1 i- 0, because Ker(A) is self-dual under
Cartier duality. Then A induces t A ~ A 181 c for the following reason. Tensoring
A over 0 with the exact sequence 0 ---t 0 ---t c ---t c/O ---t 0, by the divisibility
of A, we get another exact sequence 0 ---t Torl(c/O, A) ---t A ---t A 1810 c ---t 0
of fppf abelian sheaves (see Section 8.2.1 for fppf abelian sheaves). Since
o is a Dedekind domain, we have c/O ~ O/c-1, A[c- 1 ] = Torl(c/O,A)
(see [LFE] 1.1.1) and tA ~ AIA[c- l ] ~ A 00 c. Such a polarization is

often called a c-polarization. Starting from an O-linear polarization A, we
write C(A) for its polarization ideal; so, ker(A) = A[C(A)-l] if C(A)-l is in-
tegral. If S is a Q-scheme, the module Lie(A) is a faithful module over
End~(A) = Endo(A) 0z Q. An element ¢ E End~(A) is called symmet-
ric if ¢* = ¢ for the Rosati involution * in (rm3). We write End~_sym(A) for
the F-subspace of F-linear symmetric endomorphisms in End~(A). We put
Endo_sym(A) = End~_sym(A) n Endo(A). In particular, the algebra of sym-
metric endomorphisms End~_sym(A) is equal to the image of ~A and is of di-
mension 1, which implies Endo_sym(A) = O. Therefore if A is a c-polarization,
Homo_sym(A, t A) = Endo_sym(A) 0 c = c. For a E 0, we see easily that
aoTx = Ta(x) oa. From this, we have Aa* L = a2 AL. The monoid ring of totally
positive elements in the ideal c is generated over N (additively) by square
elements. If A and A' are two O-linear polarizations, we have A' = t~(a)A
for a E FX. By the positivity of the Rosati involution (i.e., the positive def-
initeness of the quadratic form 0: M Tr(o:o:*) on End(A) 0 Q; see [ABV]
Section 21), we find that all conjugates of a in lR are positive (i.e., a is totally
positive: a » 0). Hence we have a canonical identification peA) ~ c+ for the
monoid c+ of totally positive elements in c. Write F~ c F for the group of to-
tally positive elements. Then the class of polarizations "X = {A 0 ~(0:)10: E F~}
is determined by the pair (c, c+ = peA)) modulo multiplication by F~ and
hence only depends on the strict ideal class of c (because (t ~(o:)) 0 A = A 0 ~(o:)
is an o:-lc-polarization for 0: E F~).
We can extend the above definition of the polarization class to a smaller
subclass "X = {A 0 ~(0:)10: E O(E)+} for a set of rational primes E, requiring A
to have degree prime to E. Here 1::(17) = Q n 1::17 for 1::17 = I1PE E l:: p , 0(17) =
o 0z 1::(17), and 0(17)+ = F~ n 0(17). Shimura was the first to consider such
a polarization class, which he called a weak polarization on a given abelian
variety (cf. [Sh4] 4.7-8). This is natural since Shimura varieties classify abelian
varieties up to isogenies (as we show later in Section 4.2); so, polarization
cannot be kept without having some leeway. Suppose that A is defined over
a l::(E)-scheme. Thus A E "X is an etale isogeny; so, for each geometric point
s E S, the algebraic fundamental group 7rl(S, s) (in Section 4.4.2) acts on "X.
We say that "X is defined over S if it is stable under the action of 7rl (S, s) for all
geometric points s E S. Changing the geometric point s changes 7rl(S, s) by
an isomorphism (see Section 4.4.2), and we only need to require the stability
taking one geometric point on each connected component of S (for more details
of this type of results, see Section 6.4.1 where the rationality of level structures
is dealt with by such an argument).
Suppose that "X is defined over S. Then we find an irreducible etale Galois
covering S' I S so that we have a member A : A x s S' -+ tAx s S' in "X.
The map Gal(S'IS) :3 a M AI-a E 0(17)+ is a homomorphism of the finite
group Gal(S' I S) into the torsion-free module 0(17)+ (on which Gal(S' IS)
4.1 Hilbert~Blumenthal Moduli 101
acts trivially). Thus Al~a = 1, and A is defined over S. By this descent

argument, we can always find a member A E "X that is defined globally over
S if "X is defined over S. Thus our definition of S-integrality is equivalent to
having a member A defined over S in the class "X (which is the definition of
integrality Kottwitz used in [Koj Section 5). As already remarked, the isogeny
rt
2A is induced globally by a line bundle; so, if 2 E, we have A in "X globally
associated with an ample line bundle over S.
Take a geometric point S'E S of the base scheme of an AVRM A/ s . We
now study the Tate module T(A) = Ts(A) = ~NA[N](k(s)) for the residue
field k(s) of s (which is algebraically closed). By ~, T(A) is an O-module. We
put T(A)p = T(A) 0 0 Op (Op = ~nOjpno) for a rational prime p. An
abelian scheme A/ s of relative dimension 9 is called ordinary (at p) if we have
an embedding /L~ Y A[Pj of finite flat group schemes etale locally.
Proposition 4.1 Suppose that k(s) is of characteristic p. Let As/k(s) be the
fiber of A at s, and suppose that As is ordinary; that is, As[Pj ~ /L~ x (ZjpZ)9
for 9 = dim(As). Then we have T(As)p ~ Op as O-modules.
Proof. Since As and the connected component As[PjO ~ /L~ have the tan-
gent space of dimension 9 over k(s), they share the tangent space Lie(As) at
the origin. As O-modules, they are free of rank 1 over 0 0z k(s) by (rm4).
Write As[PjO = Spec(R) for a k(s)-bialgebra R (e.g., [GMEj 1.6.3 for bial-
gebras). Then for its unique maximal ideal meR, we have Lie(As) =
HOmk(s) (mjm 2 , k(s)). By Cartier duality between A[Pj and tA[Pj (e.g., [GMEj
1.7 and Theorem 4.1.17 (2)), we have
tAs[Pjet ~ Homcp-sch(As[PjO,/Lp) Y HomscH(As[PjO,/Lp) ~

HOmk(s)~aI9(k(s)[tl/(tP), R) -+> HOmk(s)~aI9(k(s)[tl/(tP), Rjm 2 ) ~ mjm 2 .
Since A[PjO ~ /L~ over k(s) for 9 = dimAs, it is easy to see that the above
morphism induces tAs[Pjet 0lF p k(s) ~ HO(As, DAB/k(s)). By duality and po-
larization, we get As [pjet 0lF p k(s) ~ Lie(As). This shows that
Lie(As) ~ T(As)p 0z k(s) as 0 0z k(s)-modules. (4.1)
Then by Nakayama's lemma ([CRTj Theorem 2.2), we conclude from (rm4)

the desired assertion. 0
The fact (4.1) shows that Lie(A)/s and Os 0 T(A[POOjet) for the locally
constant sheaf T(A[poojet)/s = ~nA[pnjet have the same stalk everywhere
if Aj S is ordinary at every point over Sand p is locally nilpotent over S, and
hence they are isomorphic.
Corollary 4.2 If p is locally nilpotent over S and an abelian scheme A/ s is
ordinary, we have a canonical isomorphism Lie(A)/s ~ Os 0z T(A[poojet).
If furthermore, pn = 0 over S, we have Lie(A)/s ~ Os 0z A[pnjet, where
A[pnjet is the maximal etale quotient of A[pnj.
Proposition 4.3 Suppose that k( s) is of characteristic O. Then we have an

isomorphism T(As)p ~ O~ as O-modules.
Proof. Since pn E End(As) acts on DAs/k(s) by multiplication by pn, pn E
End(As) induces an automorphism of the cotangent space; so, p : As --+ As
is etale of degree p2 g; so, As[pn] ~ (Z/pnz)2g, and hence T(As)p ~ z~g on
which 0 acts as linear endomorphisms. Thus it is an Op-module for Op =
o 0z Zp. Since the characteristic polynomial of the action of a E 0 is given
by the square of the characteristic polynomial of the regular representation of
a acting on the <Ql-vector space F (cf. [GME] Theorem 4.1.20), we find that
T(As)p ~ O~. 0
By the two propositions, we have
Corollary 4.4 For an AVRM A over S, the kernel A[N] of the multiplication
by a positive integer N is a locally free group scheme of rank N 2g. If N is
invertible on S, A[N] ~ (O/NO)2 etale locally. If the reduced part sred is of
characteristic p and A is ordinary over S, we have A[pn]o ~ (l-l 0z /-Lpn and
A[pn]et ~ O/pnO etale locally. The exact sequence 0 --+ A[pn]o --+ A[pn] --+
A[pn]et --+ 0 is split over each geometric point of S.
Proof. The multiplication by N : A --+ A is a finite flat morphism, because
it is quasi-finite and A is is smooth proper and surjective (cf. [GME] 2.8.1).
By a result of Mumford, we show in Sections 4.1.2 and 4.1.6 that the coarse
moduli scheme of an AVRM contains a characteristic 0 geometric point and
an ordinary characteristic p point. The above two propositions then show
that A[gn] for all primes g is locally free of rank g2gn over S. Since A[N] =
TIR A[ge(R)] for the prime decomposition N = TIR ge(R), A[N] is locally free of
rank N 2 g.
Suppose that sred is of characteristic p. We have an exact sequence of
locally free group schemes, 0 --+ A[pn]o --+ A[pn] --+ A[pn]et --+ 0, which is split
over any perfect field because of A [pnjTed ~ A[pn]et (see [ABV] Section 14). In
particular, it splits over an algebraically closed field and over geometric points.
We know that A[pn]/~ ~ (O/pnO)ls etale locally by Proposition 4.1. The
Cartier dual of A[pn]o is t A[pn]et ~ (O/pnO)IS (e.g., [GME] Theorem 4.1.17),
and the dual of O/NO = (Z/NZ) 0z 0 is given by (l-l 0z /-LN by the pairing
induced by the canonical isomorphism Homcp_sch(/-LN, IG m ) = Z/NZ combined
with the trace pairing (l-l x 0 --+ Z. 0
4.1.2 Moduli Problems with Level Structure
We consider the fibered category of quadruples
over a Z[~]-SCH. Here the triple (A,A,~: 0,--+ End(A)) is an object of

A F , the level r(N)-structure ¢N is supposed to be O-linear, and the duality
pairing eN : A[N] x t A[N] -+ /IN (the so-called Weil pairing; see (PI) in 8.2.3)
composed with the polarization A : A -+ t A agrees via ¢ with the natural
pairing \-, .) N : CO- 1 ®7/,/lN )ffi(O®7/,N- 1 ZjZ) -+ /IN given by (a ® (, b ® m) r-+
e (TrF/IQ!(ab)) em, where e(x) = exp(2nix). We often simply write (A, A) for
(A, A, ~ : 0 Y End(A)), dropping ~ as we agree to have ~ always in this
chapter. A morphism f : (A, A, ¢) / S -+ (A', A', ¢') / S is induced by a morphism
f : (A, A) -+ (A', A') in AF with the additional requirement that ¢' = f 0 ¢.
We fix a polarization ideal c, and assume that A is a c-polarization. Then
for a given c, the Hilbert-Blumenthal scheme 9J1( c, r(N)) /7/,[11J represents the
following functor Cr(N) = Cc,r(N) : Z[k]-SCH -+ SETS given by
Cr(N)(S) = [(A,A,¢N)/s with c-polarization A], (r(N))
where as before H is the set of isomorphism classes of the objects * inside the
brackets. When N 2: 3, as we show in Remark 6.21, this functor is rep-
resentable by a quasi-projective scheme over Z[k] (which is smooth over
Z[C, Nd(~')C] for 0 < CEQ with (C) = Q n c, as we show later in Sec-
tion 7.1), and for small N ::; 2, we have at least a coarse moduli scheme (cf.
[GME] 2.3.2).
Let 91 be a nonzero ideal of O. Abusing the notation, we write /l'Yl/7/, for
the locally free group scheme (of finite rank) given by the 91-torsion points
of G m ®7/, 0- 1 : /l'Yl(R) = {x E Gm(R) ®7/, ()-II91x = a}. If 91 is generated by
a positive integer N, we have /l(N) = /IN ®7/, ()-1. We consider the moduli
problem classifying test objects (A, A, i : /l'Yl Y A[N]) / s of level rl (91) in
place of the level r(N)-structure. The classifying functor cr{('Yl) = cc,r{('Yl) :
Z[k]-SCH -+ SETS for an integer N with 91 n Z = (N) is given by
Cr,J('Yl)(S) = [(A,A,i'Yl)/s with C(A) = c],
where i'Yl : /l'Yl Y A[N] is the inclusion of locally free group schemes. If
each test object (A, A, i'Yl) / s is rigid without nontrivial automorphism, the
functor is representable by a quasi-projective scheme 9J1( c, rl (91)) /7/,[ 11 J' which
is smooth over Z[Nd(F)]. Thus the functor cr,J('Yl) is representable if 91 is deep
enough. The coarse moduli 9J1( c, rl (91)) /7/,[ 11 J exist for all 91.
Let O~ = r;n Ox. We may ease the polarization condition, that is,
replace a polarization A by a polarization class 0 ~ A = {EA lEE 0 ~ }. Easing
the polarization requirement allows us to change morphisms of our category
from O-linear isomorphisms to O-linear isomorphisms up to unit factors (we
later study categories up to a larger class of isogenies in Section 4.2). For a
given O~A of A, we say that O~A is defined over S if we find an ample line
bundle, locally under Zariski topology on S, giving rise to an element in O~A
(cf. [ABV] Section 8). Consider the following functor Crd'Yl) : SCH/7/,[I/NJ -+
SETS (of level rl(91)) for an integer N with 91 n Z = (N) given by
Cr,('Yl)(S) = [(A,O~A,i'Yl)/s with c-polarization class O~A],

where (A,O~A,i) ~ (A',O~A',i') if we have an isomorphism f : A ~ A' of

abelian schemes with f 0 i = i' and O~A = O~j* A'.
Since E E O~ = Ox n F~ with E == 1 mod I)( gives an automorphism
of (A,O~A,i'J1)/s, the automorphism group Aut((A,O~A,i'J1)/s) is not triv-
ial; so, the functor Er , ('J1) is not representable in the strict sense that we
have a moduli scheme 9J1(c,l)() and a universal couple (X,O~A,i'J1) such
that for any triple (A,O~A,i'J1)/s E Er ,r'J1) (S), we have a unique mor-
phism f : S ~ 9J1( c, I)() such that j* (X, O~A, i'J1) is isomorphic in a unique
way to (A, 0 ~ A, i'J1) / s. But if I)( is sufficiently deep, the uniqueness is valid
up to a multiple of 0 ~. The coarse moduli scheme 9J1( c, I)() with functo-
rial isomorphisms 9J1(c, I)()(k) ~ Er,c'J1)(k) (for algebraically closed fields k)
exists always. This fact can be seen as follows. Letting E E O~ act on
Ec,rl('J1) by (A,A,i) H (A,EA,i), we find that En ('J1) has its coarse mod-
uli if Erl ('J1) has, and the coarse moduli scheme 9J1( c, I)() /Z[ k 1 is given by
9J1( c, I)() = O~ \9J1( c, rl (I)()). As is clear from this construction, for each mem-
ber (A, O~A, i'J1)/s E Er, ('J1) (S), we have a unique morphism f : S ~ 9J1(c, I)()
with j*(X, O~A, i'J1) ~ (A, O~A, i'J1)/s, but this last isomorphism is unique
modulo O~ even if (X, O~A, i'J1)/'JJ1(c,'J1) exists (e.g., if E = 10 2 with 10 E Ox as
long as E == 1 mod I)( for E E 0 ~). This is a typical phenomenon we encounter
if we consider the moduli problem with ambiguity of polarization (which is
a moduli problem of GL(2)-type). To justify this point, we need to change
the structure of the category AF declaring some class of isogenies as isomor-
phisms. We study this point later in Section 4.2. We record what we have
described.
Theorem 4.5 Let the notation and the assumption be as above. Let r be one
of the level structures of type r(N), rl (I)() , and r 1 (I)(). Then we have
(1) The coarse moduli scheme 9J1( c, r) representing the functor Er always
exists over Z [1:t ] for the minimal positive integer N E 1)(.
(2) Take c 1 cO, and let C be the minimal positive integer in C~l. The
scheme 9J1(c, r) for r = r(N) and rl(l)() is smooth over Z[d(F1Ncl
if I)( is sufficiently deep that test objects of type r have no nontrivial
automorphisms (e.g., if N 2: 3 for r = r(N)).
(3) The scheme 9J1(c, r) for r = r(N) and rl(l)() represents the functor Er
over Z [1:t ] if I)( is sufficiently deep that test objects of type r have no
nontrivial automorphisms.
4.1.3 Complex Analytic Hilbert Modular Forms
We describe the Archimedean theory of Hilbert modular forms, limiting our-

selves to what we need later. We accommodate the r 1 (I)()-type moduli prob-
lem for an integral ideal I)( (as well as that of r(N)-type for an integer N).
When we deal with r(I)()-structure, the ideal I)( is assumed to be generated
by the integer N. We use the symbol ¢ (resp. i) to indicate a level r(N)-

structure (resp. a r 1(1)1)-structure). Over (C, by the theory of theta functions
due to Riemann-Poincare-Lefschetz (see [ABV] Chapter I), the category of
Hilbert modular test objects (A, A, ¢, (resp. i), w) is equivalent to the category
of triples (.L:, A, ¢, (resp. i)) made of the following data: .L: is an O-lattice in
o i3)z (C = (CI, an alternating pairing A : .L: 1\0 .L: ~ c* , and a level N -structure
¢ or i. A level r 1 (1)1)-structure i is an inclusion i : 1)1* /0* '---+ F.L:/.L:, and a
level r(N)-structure ¢ is an inclusion ¢: (1)1*/0* x 0/1)1) '---+ F.L:/.L:, where
1)1* = 1)1-1()-1 for the absolute different () of F.
The alternating form A is supposed to be positive in the sense that the
symmetric bilinear form (u, v) H I~\~~~) is totally positive definite and that
the pairing (-, .) N matches the pairing
via ¢ when we deal with level r(N)-structure. The differential w can be recov-
ered by L: A(<C) = (CI/.L: so that w = L*du, where u = (UU)UEI is the variable
on (CI. Conversely, for a given test object (A, A) in AF with c-polarization A,
is a lattice in (CI, and the polarization A : t A ~ A i3) c induces .L: 1\ .L: ~ c* (see
(Pl-3) in Section 8.2.3).
Using this equivalence, we can relate our geometric definition of Hilbert
modular forms with the classical analytic definition. Define 3 by the product
of I copies of the upper half complex plane SJ. By (~~) H(~: :7-f~~:) u'
we have G 1(JR.)/Ci ~ 3 for n
= S02(F i3)Q JR.) and i = (0, ... ,0) E 3
(which is the origin 0 of 3). We regard 3 C F i3)Q (C = (CI made up of
z = (zu )UEI with a totally positive imaginary part.
We identify /1N with N- 1'lL/'lL by exp(21l'im) +-t (m mod'lL). This iden-
tification induces /1(N) = /1N i3) ()-1 ~ (N- 1 'lL/'lL) i3) ()-1 = (N)* /0*, which
further induces /1m ~ 1)1*/0*. Choose two ideals a and b prime to 1)1 with
ab- 1 = c. For each z E 3, we define
.L: z = 21l'H(bz - a*), Az(21l'H(az - b), 21l'H(ez - d)) = (ad - be) E c*
with iz : /1m = 1)1* /0* --t (CI /.L: z given by iz(a mod 0*) = (-21l'0a +.L:z)
and ¢m,z : 1)1* /0* X 1)1-1/0 --t (CI / C given by, for J 1 = (~ (/ ),
¢m,z(a, b) == 21l'H( -a + bz) == (a, b)J1 . t(z, 1) mod .L: z .
Taking association (a, b) H -a + bz instead of (a, b) H az + b looks strange

(in order to put a complex structure on Lffi. = Li3) z JR. for L = a* EEl b). However,
we later interpret our moduli scheme in Deligne's language in Section 4.2, and
then 3 is identified with the set X+ of all conjugates under G(JR.)+ of the group
homomorphism ho : C x 3 a + bi r-t C!\ ~) E G (JR.) +. The identification is
given by 3 3 g(i) r-t 9 . hog- l E X+ (i = (R, ... , R) E 3). For each
h E X+ (corresponding to Z E 3), LJlI. 3 v r-t h( a + bi)v gives the complex
structure we want. For the alternating pairing (x, y) = tx . Jly of V = F2
(column vectors), Hh(X, y) = (x, h(i)y) gives a positive definite Hermitian
form on LJlI. = VJlI. so that (-,.) gives rise to the Riemann form on LJlI./ L =
C 1 / L z . The natural action v r-t av of a E G(JR.)+ on LJlI. induces the right
action: z r-t a-l(z) on 3. Indeed, we have
(ax,h(i)ay) = (x,a'h(i)ay) == (x,a-Ih(i)ay) mod T(JR.)+
for the adjoint involution i, since a' = III.taJI = det(a)a- l . In other words,
the natural left action of G on V induces the right action z r-t a' (z) = a ~ I (z)
on 3. This is the explanation of why we need to have (a, b) r-t -a + bz.
If a = (~~) E G(Qlh, the multiplication by (-cz + a) E Fr!j induces an
isogeny
I xc-+( -cz+a)x I
(C /La.-1(z),A,¢'Jl,cc1(z)) ) (C /C,det(a)A,¢'Jl,z oa). (4.2)
This shows that the action of a E G(QI)+ sends the complex point of the
projective limit SJJ1(C) = 0.!l N SJJ1(c,T(N))(C) represented by z E 3 to that of
a-l(z). The above isogeny is an isomorphism if and only if a E r(m; a, b) for
the arithmetic subgroup r(m; a, b) of G(QI) defined below.
For four fractional ideals a2 ] C F, we write symbolically (~;; ~;;) for the 0-
lattice in M2 (F) made up of matrices whose (i, j)-entry is in a2 ] for i, j = 1,2.
Define three types of congruence subgroups of G(QI)+ by
rl(m'ab)={(ab)E(
, , c d 0
'Jlabi1 (abl*)lad-bcEO
0
x
+, a-1Em},
rll(m; a, b) = rl(m; a, b) n SL 2 (F), (4.3)
r(m; a, b) = {( ~~) E rl(m; a, b* E m(ab)*}.
Since det(rl(m; a, b)) = 1, the condition a-I Em for rl(m; a, b) automati-
cally implies d - 1 E m.
For the moment, we deal with rl(m)-structures. We let 9 = (g,,) E G(JR.)+
act on 3 by linear fractional transformation of g" on each component z". It
is easy to verify by (4.2), similarly to Theorem 2.39, that
(Ism) (Lz, Az , iz) ~ (Lw, Aw, iw) -¢=? W = r(z) for r E rl(m; a, b).
Thus the complex points of the coarse moduli scheme SJJ1( c; rl (m)) are given
by rl(m; a, b)\3. However, it looks odd that we have an extra dependence on
the pair (a, b) not just on c and m. Actually, as long as ab- l = c, the complex
manifolds rl (m; a, b) \3 are isomorphic for all choices of (a, b). To see this
fact, let us analyze the equivalence classes of cusps of r(O; 0, C~l). Cusps of
3 = SjI are given by pI (F) = F U {oo}. Recall that G(Q)+ is the subgroup
of G(Q) made up of matrices with totally positive determinants.
The groups G 1 (Q) and G(Q)+ act on Pl(F) by linear fractional trans-
formation (or equivalently by matrix multiplication on homogeneous coor-
dinates). The stabilizer of the infinity cusp is the subgroup BeG of up-
per triangular matrices. Thus Pl(F) ~ G 1 (Q)/B 1 (Q) = G(Q)+/B(Q)+
for B(Q)+ = B(Q) n G(Q)+ and Bl = B n G 1 . For an arithmetic sub-
group r c G 1 (Q), its closure l' in G 1(A,.(00)) is an open compact sub-
group, and r = l' n G 1 (Q). Two cusps sand s' are equivalent under r
if ,(s) = s' for some, E r. Thus, by the strong approximation theorem,
G 1 (Q) is dense in G 1 (A(00)), and the equivalent classes of cusps are in bi-
jection with r\G 1 (Q)/B 1 (Q) = 1'\G 1 (A(00))/B 1 (Q). We now make this set
explicit for r = r(O; 0, C 1 ). For 9 E G 1 (Q), we write £(g) for the left
column vector of g. Then writing £(g) = t(a, c), define ilc(g) = (cell + aO)
as a fractional ideal of F. Then £(g)b = £(gb) for b = (g:) E B 1 (Q). If
we choose another base, ilc(g) = c'ell + a'O, we find, E r(O; 0, e- 1 ) with
, . £(g) = t(a', c') = £hg). Thus by 9 H ilc(g), we find an isomorphism
r(0;0,c 1 )\G 1 (Q)/B 1 (Q) ~ elF for the class group elF of F. In other
words, the cusps of r( 0; 0, e- 1 ) are in bijection with the class group of F. We
can extend ilc to G 1 (A (00)) by ilc (g) = (cev + aO) n F, which gives rise to the
identity 1'(0; 0, e- 1 )\G 1 (A (00))/ Bl (Q) ~ elF. In particular, ilc (g)o = ilc (ga)
for a E BdA(oo)) with riO n F = o. By the Iwasawa decomposition (e.g.,
[EPEl Lemma 9.2), we have G(A(oo)) = 1'(0; 0, e- 1 )B 1 (A(00)); so, each cusp
of r(0;0,e- 1 ) is represented by b E B 1 (A(00)). Thus we have an association
of cusps s to ideals of F given by choosing b E B 1 (A (00)) representing sand
sending s to the ideal class of il c (b).
There is another description of cusps of r( 0; 0, e- 1 ). The set of pairs
(0, b) with ob- 1 = e is in bijection with the set of cusps of r( 0; 0, e- 1 ) in the
following way. Since ob- 1 is fixed to be e, we find (0, b) = (0, oe- 1) = 0(0, c 1 ).
Thus a standard choice of a cusp is (0, C 1 ), which we call the infinity cusp
of 9)1(e, 1), and 0 corresponds to a cusp s if ilc(b) = 0 for b E B 1 (A(00))
representing s.
For one r of these level structures rl(91), n(91), and r(N), the moduli
space 9)1 ( e, r) has a finite number of cusps (depending on the choice of a level
r-structure on the Tate AVRM Tatea,b(q) we define in Section 4.1.5) over
a given cusp s represented by (0, b). When rl (91) e r e r 1 (91), we have
a canonical level structure i can ,')1 as we specify in the following subsection.
When r = r(N), we have a canonical level structure ¢can,N over an extension
ring Z[d(i)N]((N- 1 0b)) of the canonical ring of definition Z[d(i)Nl((ob)) of
Tatea,b(q). We simply write (0, b) for the cusp of 9)1(e, r) with this canonical
choice of the level structure. Write
108 4 Hilhert Modular Varieties
For 'Y E S L2 (F) with 'Y ( abo ) = (~'1 ), 'Y brings the cusp corresponding to the
pair (0, b) to the standard cusp (0, c- 1). In particular, we have
'Yr11(C, sn),,(-l = r11(sn; 0, b), 'Yr1(C, sn),,(-l = r 1(sn; 0, b), and

'Yr(c, sn),,(-l = r(sn; 0, b). (4.4)
Thus we find rl (c, sn) \3 ~ rl (sn; 0, b) \3 for the above choice of 'Y. For each
'Y
ideal 0, (0, oc- 1) gives another cusp. The two cusps (0, oe 1) and (5, sc- 1)
°
are equivalent under r1 (c, sn) if = as for an element a E F X with a == 1
mod sn in F~, where F'Jl = TIiI'Jl FI for prime factors [ of sn.
Returning to the coarse moduli 9J1(c,rl(sn)) over Z[d(f)N] (snnz = (N))
for the functor [c,rf('Jl) : Z[d(f)N ]-SCH --7 SETS,
[c,rf('Jl)(S) = [(A,).,i)js with c().) = c],

we have 9J1(c, rl(sn))(C) = [c,rl('Jl)(C) ~ rl(c, sn)\3 from (Ism). In ex-
actly the same way, we get 9J1(c, r(N))(C) ~ r(c, N)\3 and 9J1(c, sn)(C) ~
r 1 (c, sn) \3, canonically.
Let G = Res ojz GL(2), and set Z = TIC:prime Zc. Define four open compact
subgroups of G(A(=)) by
ro(sn) = {(~~) E G(Z)le E snO},
r 1 (sn) = {(~~) E Fa(sn)la == 1 mod snO},

(4.5)
r(sn) = {(~~) E r 1(sn)lb E sno, a == 1 mod snO},
rf(sn) = {x E r 1(sn) I det(x) E 1 + snO}.
Put
S* (sn) = (g ?) -1 r* (sn) ( g?) for *= 0, 1, sUsn) = (g ~ )-1 rf (sn)( g ~ )

and S(sn)=(g~)-lf(sn)(g?) (4.6)
for an idele d with dO = () and d(V) = l. Then taking an idele e with cO = C

and etc) = 1, we see that
r 1 (c,sn) = (8~)Sl(sn)(8~)-1 nG(Q)+,

r(c,sn)= (8nS(sn)(8n-1nG1(Q) and
r 11(c, sn) = (8 ~) sUsn) (8 ~) -1 n G1(Q).
Choosing a complete representative set {e} C FJ:

for the strict ray class group
Cit of F, we find by the approximation theorem that
CEClt CECltCl)'l)
where G(JR)+ is the identity-connected component of the Lie group G(JR),

Clt (In) is the strict ray class group modulo >J1, and S is either Si(>J1) or
S(>J1). Similarly, by the strong approximation theorem ([LFE] Chapter 9), we
have G 1(A) = G 1(Q)K . G(JR)+, where K is given either by
Gl(ACoo))n(o~)st(>J1)(on-l or Gl(ACoo))n(o~)S(>J1)(o~)-l.
This shows
G(Q)\G(A)jSl(>J1)CO ~ U VJ1(c, >J1) (<C) ,
CEClt
u VJ1( c, >J1) (<C),
(4.7)
G(Q)\G(A)jSt(>J1)Z(JR)Co = U VJ1(c,rl(>J1))(<C),
CECltCl)'l)
G(Q)\G(A)jS(N)Z(JR)Co = U VJ1(c,r(N))(<C),
CECltCl)'l)
where Z is the center of G and Co is the stabilizer of i = (.;=1, ... ,.;=1) E 3

in G1(JR) (which we regard as the origin 0 of 3).
Recall the identification X(T) = HOmalg_gp(T, G m ) with Z[I] so that
k(x) = TI" a(x)ku. Let ~k over VJ1(c, >J1), VJ1(c,rl(>J1)), and VJ1(c, r(N)) be
the automorphic vector bundle of weight k E X(T) = Z[I] (whose precise
definition will be given in Section 4.1.6). Assuming that F =1= Q (since the
case where F = Q is already treated in Chapter 3), define
Gk(C, >J1; R) = HO(VJ1(c, >J1)/R'~/R)'

Gk(C, rl(>J1); R) = HO(VJ1(c, rl(>J1))/R'~/R)'
and Gdc, r(N); R) = HO(VJ1(c, r(N))/R'~/R) (4.8)
for Z[dCf)N ]-algebras R. Regarding J E Gk( c, >J1; <C) as a holomorphic function
of z E 3 by J(z) = J(C z , O~Az, i z ), it satisfies the following automorphic
property by (4.2) (cf. [GME] 2.4.3),
Jb(z)) = J(z)(cz + d)k with (cz + d)k = IT (c" z" + d")ku (4.9)
"
for all 'Y = (~~) E r 1 (c, >J1). We leave it to the reader to formulate the
corresponding facts for J E Gk (c,rl(>J1);<C) and J E Gdc,r(N);<C). The
function J has the Fourier expansion J(z) = I:~Eab a(~)eF(~z) at the cusp
corresponding to (a,b). Here eF(~z) = exp(27r.;=1I:,,~"z,,). This Fourier
expansion gives the q-expansion Ja,b(q) substituting q~ for eF(~z) as we show
in the following subsection (see also [DAV] Proposition V.1.5).
4.1.4 Toroidal Compactification
We now give an oversimplified description of how to make the toroidal

compactification of the Hilbert-Blumenthal moduli space 9Jt(c, r(N)) over
Z[~, f.lN], because the main purpose of this chapter is to present a short de-
scription of the structure of our theory applied to Hilbert modular varieties
and in any case, the process of the toroidal compactification has been well cov-
ered in book form by [DAV] and [CSM]. We can formulate the result also for
the moduli spaces 9Jt( c, rl (1)1)) and 9Jt( c, 1)1) with minor modification which
is left to the reader, because in any case these moduli over Ware (geometric)
quotients of 9Jt( c, r(N)) by a finite constant group action as long as N is
prime to p. Main references in addition to the two books are [C], [K3], and
[Ra] (and [HT]).
Let (0, b) be a cusp of 9Jt(c, r(I)) for two fractional ideals and b (so, °
c = ob- 1 ). Since the cusp is identified with the infinity cusp of r( 0; 0, b), it
is fixed by r 00(0; 0, b) = r(O; 0, b) n B(Q), where B is the upper triangular
Borel subgroup of G. The group roo (0; 0, b) is the stabilizer of the exact
sequence 0 -+ 0* -+ (0* EB b) -+ b -+ 0 in the sense that "( E r(O; 0, b) acts on
the column vector t (a, b) E (0* EB b) via left multiplication t (a, b) f--t "( . t (a, b)
and r00(0; 0, b) is the stabilizer of (0* EB 0). The O-module H = (0* EB b)
(column vectors) is identified with .c z = bz + 0* c Fc by sending t(a, b) to
(a, b)J1 . t(z, 1) = a - bz for J 1 = (~ r} ), and the left multiplication by ,,(-1
gives rise to z f--t "((z) by ((a, b) . t"(-I) J 1 . t (z, 1) = (a, b) J 1 . t ("(( z), 1) (cz + d)
if"( = ( ~ d)'
Let N be a positive integer prime to p. Since 9Jt(c, r(N)) classifies AVRMs
with level N-structure cPN : (~r EB ~ ~ (1':,~)* EB N~'b = N~~.cz, a cusp of
9Jt(c, r(N)) is determined by a triple (0, b, cPN) with ob- 1 = c modulo the ac-
tion of roo (( N); 0, b). We fix a triple (0, b, cP N) as above and study the toroidal
compactification of 9Jt(c, r(N)) at the cusp (0, b, cPN), which heavily depends
on a chosen cone decomposition of C = {~ E F 00 I~O' > 0 for all u : F '--+ lR}
(the cone of totally positive numbers in Foo = F ®IQJlR = lR!).
A (simplicial) cone u in C of dimension m ::; [F : Q] = 9 is an open
span u = lR+ VI + lR+ V2 + ... + lR+ Vm inside C for linearly independent
VI, ... ,Vm E C, and a cone decomposition C of C is a decomposition of C into
a disjoint union UO'EC u of open cones u of dimension::; g. Choose a cone
decomposition C = C(o, b, cPN) of C = UO'EC u such that
(PCl) u is an open simplicial cone;
(PC2) The cones in C are permuted under multiplication by E2 for E E
T(Z)(N), where T(Z)(N) = {E E OX IE == 1 mod N}. There are
finitely many cones modulo T(Z)(N), and E(U) n u -=I- 0 implies that
E = 1;
(PC3) u is smooth (i.e., generated by a part of a Z-base of (ob)*);
(PC4) {u} is sufficiently fine so that the toroidal compactification is projec-
tive (see [DAV] IV.2.4 for an exact condition for projectivity).
The existence of such a cone decomposition was remarked by T. Shintani in

1976 (for a proof, see [LAP] IV.5.2 or [LFE] Theorem 2.7.1) and previously
also by algebraic geometers studying toroidal compactification ([TEB]). We
make a choice once and for all of such a cone decomposition C(a,b) for N = 1
and use the same decomposition for the cusp s = (a, b, ¢N) for every ¢N
and all N > 1. Then we have an action of r(O; a, b) on {Cs}s (indexed by
equivalence classes s of cusps), and r(O; a, b) permutes them by ¢N M ,,(o¢N.
If a and b are prime to N, we have a canonical isomorphism
r(o; a, b)/ r(N; a, b) ~ r(O; 0, c- 1 )/ r(N; 0, c- 1 ),

which in turn is isomorphic to SL 2 (0/(N)) (possibly noncanonically) by con-
jugation inside G(A(oo)). Thus we may regard the action on the cusp as an
action of SL 2 (0/(N)).
Each Hilbert modular form f has a q-expansion
f(q) = a(O, f) + a(~; f)qt.
at the cusp (a, b, ¢N). The minimal compactification M* (c, r(N)) of the open
modulus space fJJt(c,r(N)) if it exists is characterized by the property that
it is covered by any smooth compactification of fJJt( c, r(N)) (having a divisor
of normal crossing at the cusps) so that the covering map induces an isomor-
phism in the interior fJJt(c, r(N)). Thus we convince ourselves that the formal
stalk of the minimal compactification at the cusp s = (a, b, ¢N) is given by
HO(T(71.)(N), Rs(N)), where
Rs(N) = {ao + ~
t.EN-1abnc
at.qt.lat. E 7l.[~,J1,N]}
on which f. E T(71.)(N) acts by
ao + ~at.qt. MaO + ~at.qE2t..

t. t.
Here the series ao + Lt. at.qt. is a formal series whose coefficients at. could be
nonzero for all ~ E (C n N- 1 ab)
U {O}. Thus Rs (N) is the completion of the
monoid ring Rs(N) of the semi-group (N- 1 ab) n C under the adic topology
of the augmentation ideal.
Let a V be the dual cone a V = {x E FooITrF/IQi(xa) :::: O}. Then C =
ncr a v. To make our conviction feasible, we try to describe the complicated
ring Rs(N) using more reasonable rings. We consider the monoid ring Rcr(N)
of a V n N- 1 ab. Thus
Ra(N) ~ {ao+ ~a"z'la, E:/:[ ~'MN1, v

,E' n N-'ab} ,
but the sum ao + I;~ a~q~ is supposed to be finite (so, a~ = 0 for almost
all O. For each (J as above, by (PC3), each cone (J is generated by a part
of a Z-basis of N (ab) *, and hence (J v is generated by a part of a Z- basis
of N- I ab. Thus we may assume that (N (ab) *) n (J is generated over Z by
t I , ... , tr (0 < T :; [F: Q] = g). We have a base 6, ... ,c,g of N-Iab so that
Tr(t,c,)) = 6,) for 1 :; i :; j :; T and Tr(t,c,r+k) = 0 for k > 0, and each
C, EN-lab n (Jv can be uniquely written as C, = I;, mic', with mj E Z and
m) :::: 0 if j :; T. Writing T) = q~J, we find
with Spec(R,,(N)) = G~ x G'ir,-r.

The additive group G a = Spec(Z[ -b,
/-IN ][T]) is a "local" compactification
of G m at the origin filling the point "0" onto the hole at the origin of G m :
G a = G m U {O}, and therefore Spec(R,,(N)) is a partial "local" compactifica-
tion of Spec(Rr(N)) ~ G~-I x G'ir,-r+I for each face T of (J. Thus we can glue
{Spec(R"(N))},, over the monoid ring Rs(N) to get the toroidal compactifi-
cation X of {Spec(R"(N))},, on which T(Z)(N) acts by translation.
If one adds the origin 0 to C, OV is the total space F oo , and the correspond-
ing ring is given by the group ring Ro(N) of N-Iab isomorphic to
Z [~, /-IN] [TI , ... , Tg][TI-l, ... , Tg- Il,
and Spec(Ro(N)) ~ G m ® (N-Iab) (on which T(N)(Z) acts naturally by

q~ >--t qE2~). We have the inclusion Spec(Ro(N)) C X equivariant under the
action of T(N)(Z). The divisor 00 = X - Spec(Ro(N)) is stable under the
action of T(Z)(N). Write X for the formal completion of X along 00, and let
Spf(R,,(N)) be the formal completion of Spec(R" (N)) C X along 00.
We can make a quotient :£(N) = X jT(Z)(N). By the work of Mumford
[Mull, :£ carries a semi-abelian scheme Tates(q) having real multiplication
(which gives rise to an AVRM over :£O(N) := :£(N) - Ds for the image Ds
of 00 in :£(N)) and equipped with a level r(N)-structure and a polarization
determined by the cusp s (see Section 6.1.6 for the definition of semi-abelian
schemes). Indeed, the tautological homomorphism q : ab -+ Gm(Ro(l)) send-
ing C, to q~ induces by duality q : b -+ (Gm ® a*)(Ro(l)). We write qb for
its image. As was done by Mumford [Mul] (reproduced in [DAV] in Ap-
pendix), we have a semi-abelian scheme Tates(q) over :£(N) that coincides
with an appropriate "quotient": (G m ® a*)jqb on :£(N). The quotient is in
turn isomorphic to the pullback to :£O(N) of the universal abelian scheme
over 9Ji.(c, r(N)). Then by the universality of 9Ji.(c, r(N)), Ds gives a toroidal
boundary of 9Ji.(c, r(N)) at s. Performing this process for all cusps s, we
obtain the smooth toroidal compactification M(c, r(N)) = Mc(c, r(N)) of
9Ji.(c, r(N)).
This Tate AVRM Tates(q) is revisited in the following subsection. Thus by

this construction, we have a semi-abelian scheme 9 = gc over Me( c, r(N))
extending the universal abelian scheme over 9J1( c, r(N)) which induces the
Tate AVRM Tates(q) at the cusp s = (a, b, <PN). Here is a reproduction of a
statement in [C].
Theorem 4.6 Let C = {C s } s be the collection of the cone decompositions sat-
isfying (PCl-4) for each equivalence class s of cusps of9J1(c, r(l)). Then there
exists a smooth projective toroidal compactijication Me( c, r(N)) dependent on
C such that
(1) The semi-abelian scheme gc extends the universal abelian scheme over
9J1( c, r(N)) which degenerates into G m ® a* over the cuspidal divisor Ds
for s = (a, b, <PN) and coincides with Tates(q) on the formal completion
Ms(c, r(N)) of Me(c, r(N)) along Ds·
(2) The formal completion Ms(c,r(N)) along Ds is isomorphic to ?£(N) over
Z[-hJLNj.
We write 9J1 for one of the moduli spaces 9J1( c, r(N)), 9J1( c, rl (1)1)) and
9J1(c, 1)1) defined over Z[-h]. From the data C, we get a unique toroidal com-
pactification M j71I -k'!-'NJ of 9J1 j71I -k,!-'NJ' which carries a semi-AVRM 9 with a
level structure (see [C] 3.5.1) canonically determined by the toroidal compact-
ification data.
Once the smooth toroidal compactification M and the universal semi-
abelian scheme 9 ~ M with respect to C are constructed, we get a vector
bundle ~ = 7r*fl9jM which is locally free of rank g = [F : Q] over M. The
Koecher principle (which can be proven in exactly the same way for 9J1 C M
as in [DAV] Proposition V.1.5) tells us
(4.10)
for all j E Z if F -I- Q (the case of F = Q is already treated). Therefore

HO(M, det(~)®k) is independent of the compactification M; so, M covers
M* = Proj(OM) for OM = EBk>oHO(M,det(~)®k). Since the theta con-
stants give the modular functions classifying the abelian part 9J1 (cf. [CSM]
and [DAV] Theorem V.2.3), the scheme M* = Proj(OM) gives the minimal
compactification of 9J1 adding one point at each cusp, M* = 9J1 U {cusps}. In
particular, M* has the desired formal stalk Rs(N) at each cusp s.
We now study the geometric inertia group of the cuspidal prime divisor.
Since characteristic 0 theory is easier, we describe a characteristic p tame
inertia group. The prime p is assumed to be unramified in F /Q; so, it is prime
to the discriminant d(F) of F. We assume that W is the strict Henselization
of Z(p) with respect to the fixed embedding ip : ij Y ijp. Thus W is the
pullback by ip of the valuation ring of the maximal subfield of Qp unramified
over Qlp. Write IF for the residue field of W, which is an algebraic closure of
lF p . Hereafter in this subsection, MN for N prime to p denotes the special
modulo m fiber M(c, r(N)) 0Z[J.LN] IF' for the maximal ideal meW. By the
projectivity and smoothness of MN over Z [Nd(F) ' f.1N], MN/F is irreducible
(Zariski's connectedness theorem [ALG] III.11.3).
We use the same cone decomposition C for all N. Then we get a finite
covering, MNI := Mc(c, r(N')) -» Mc(c, r(N)) for all multiples N' of N. As
long as N' is prime to p, this covering is tamely ramified at the cuspidal divi-
sors I:s Ds (over Z(p)), and the closure f(o; a, b)(p) in G1(A(poo)) naturally
acts on MN'. Taking the limit M2P) = ~pfNIMNI (N' running over multiples
of N prime to p), we have Gal(M(p) /MN / F ) = feN; a, b)(p). However, the full
group G1(A,(poo)) may not act on M(p).
Let O~)+ = O~) n F':, where O(p) = 0 0z Z(p). Among simplicial cones
in C/T(Z)(N), r-dimensional faces are in bijection with generic points of the
(g - r)th stratum. This is due to the following reason (see [DAV] IV.2.5). If
a cone a is spanned by n linearly independent elements WI, ... , wn in F':,
its dual cone a V is given by QVl + ... + QVg - n + Q+Wl + ... + Q+wn with
TrF/Q(v,w J ) = O. Note that
N (ab ) * = { ex E F I (6 ~) E f( N; a, b) } .
By (PC3), the choices of VI,"" v g - n and WI, ... , Wn can be made to be
generators over Z of N-l(ab). This point that w J is a part of a Z-base of
N-l(ab) assures us of the smoothness of M N . By this fact, w J is prime to p
if pf N; in other words, Wj E O~)+'
We have Ru(N) =Z [Nd(F),f.1N] [T"SJ][T,-l] with T, = qV, and Sj = qW J •
Thus if we shrink f( N; a, b) sending N --+ 00 (keeping p t N), the tower

of rings {Ru(N)} N tamely ramifies maximally along the irreducible closed
subscheme defined by the set of n-equations {Sl = 0, ... , Sn = O}. We find
Ru(N) 0z W = W[T" SJ][T,-lkJ; so, this gives rise to an isomorphism
G~ c Spec(Ru(N)) ~ G~-n x G~ c G~ with 9 = [F : Q].
In the process of compactification, etale locally, the hole at 0 in G~ is filled
by a closed (g - n )-dimensional subscheme G~-n x 0 to get the "local"
compactification Spec(Ru(N)). By the natural morphism Spec(Ru1(N)) '---t
Spec(Ru(N)) for a face a' of a, we glue schemes {Spec(Ru(N))}uEC and then
make a formal quotient under the action of T(Z)(N). The formal quotient
is then algebraized. The algebraization requires us to impose the conditions
(PCl-4) on C ([DAV] IV.2). Thus g-dimensional cone a gives a closed point
in the highest stratum, and one-dimensional simplicial cones in the closure
of a are in bijection with prime divisors containing the point. By construc-
tion, these prime divisors cross each other normally giving the divisor Ds of
normal crossing (see [TFG] 1.8 for divisors with normal crossing). Thus the
intersection where n - 1 distinct prime divisors in Ds meet is of dimension
9 - n which forms the nth stratum of Ds.
When 0" is one-dimensional, the corresponding divisor is defined by 51 =

qWl = 0 etale locally; so, its maximal tamely ramified extension is generated
by 5~/nl = qwI/nl (etale locally) for all n1 prime to p.
Take a generic point TJ of the cuspidal divisor DSl which therefore corre-
sponds to a half line CQl~~1] c F;
with ~1] E O~)+, where O~)+ = O(~) n F;
and O(p) = 0 (>9z Z(p). As seen above, M(p) /MN infinitely ramifies over
MN at a generic point TJ of an irreducible component {TJ} of the cuspidal
divisor, and by the q-expansion principle, the inertia group I N ,1] of TJ in
Gal(M(p) / M N ) = f(N; a, b )(p) is given by
I N ,1] = {O" = (~U~") E f(N; a, b)lu E A(POO)}.
More generally, a generic point TJn of the nth stratum is determined by a

n-dimensional simplicial cone O"n. This cone determines a unique submodule
N(ab)~ of rank n inside N(ab)* generated by O"n n N(ab)* over Z, and the
inertia group in Gal(M(p) /MN) for TJn is given by
The inertia group is a module under conjugation over DN,1]n/IN,1]n for the
decomposition subgroup D N,1]n C Gal(M(p) /M N ) of TJn. If we incorporate the
data of this Galois action, we have
Theorem 4.7 Let the notation and the assumption be as above. In partic-
ular, p is a prime unramified in F /CQl. Let TJn be a generic point of the nth
stratum of the cuspidal divisor Ds of M N/ fF • Then the p-tame inertia group
I N,1]n C Gal(M(p) / M N ) of TJn is given by I N,1]n ~ N( ab)~ (>9z Z(p) (1)) as mod-
ules over the decomposition group D N,1]n) where Z(p)(l) = ~pfNJ..lN as a
Galois module.
The Galois module structure of the inertia group as in the theorem follows
from Kummer's theory described in [TFGl Section 2.
4.1.5 Tate Semi-Abelian Schemes with Real Multiplication
Consider the ring R[[(ab)+ll ((ab)+ = ab n F.:) made up of all formal. series:
aO + L a~q~ with a~ E R
~E(ab)+
for a given base ring R. Take a subset qub+ = {q~l~ E (ab)+} of R[[(ab)+ll
stable under multiplication, and. write R{ ab} for the localization of R[[( ab)+ II
by the set qub+. Then Z[-b,J..lN]{N- 1 ab} is the localization of the formal
stalk Rs (N) at the cusp 5 = (a, b, ¢ N) of the minimal compactification Miv.
The formal completion of the semi-abelian scheme Qc in Theorem 4.6 along

the cuspidal divisors corresponding to the cusp (a, b) is given by the formal
quotient (G m 0 a*)/qb (a* = a-I,,-I), where the inclusion ~ M qf. of the
*"'
additive group ab into the multiplicative group Gm (Z[ /LN]{ ab}) induces
*"'
qb = {qf.l~ E b} '---+ (G m 0 a*)(Z[ /LN]{ ab}) by the isomorphisms:
1
Hom(ab, Gm(Z[ N' /LN]{ab}))
1
~ Homo(b, Hom(a, Gm(Z[N' /LN]{ab} )))
~ Homo(b, (Gm 0 a*)(Z[~'/LN]{ab}))

via the identification Hom(a, Z) = a* under the trace pairing on F. This
formal quotient can be algebraized into a semi-abelian scheme by a work of
Mumford [Mul] (which is included in [DAV] as Appendix). Strictly speaking,
by the inclusion Rs(N) C R(J(N) for the rings introduced in the previous
subsection, over each Spf(R(J (ab)), we construct the quotient (G m 0 c:...*)/qb
and then glue them together to get the formal semi-abelian scheme Q over
X(N). We denote the algebraization of this formal quotient (G m 0 a*)/qb
by Tatea,b(q) (which is defined over Z[*",/LN]{ab}) and call it the Tate semi-
abelian scheme at the cusp (a, b). The scheme Spec(Z[*" , /LN]{ab}) is obtained
by removing the cusp s from its formal neighborhood Spf(Rs(N)) of s E
MIv. The semi-AVRM coincides with A Xvn Spec(Z[*",/LN]{ab}) which is the
universal abelian scheme A over 9Jl restricted to Spec(Z[*",/LN]{ab}), and
over the cuspidal divisor D s, the connected component of Q is isomorphic to
G m 0a*.
If w E ° is prime to a, we have a canonical exact sequence 0 -+ a-I ~
a-I -+ O/(w) -+ O. Tensoring this sequence with G m 0,,-1 over 0, we get
another exact sequence:
since G m 0 ,,-I is divisible and G m 0,,-1 00 a-I = G m 0 a*. Since

Dedekind domain, we have Torp(G m 0,,-I,0/(w)) ~ O/w, and the above
° is a
exact sequence identifies G m 0 a*[w] with /L(w). If 1)1 is prime to a, taking

w E 1)1 prime to a, we find /L'Yt C /L(w) C G m 0 a*. This inclusion combined
with the projection Gm 0 a* ---» Tatea,b(q) gives us a canonical level T I (I)1)-
structure ican,'Yt : /L'Yt '---+ Tatea,b(q) as long as a is prime to 1)1. If further,
b is prime to 1)1 = (N), N-Ib/b is canonically isomorphic to N-I/O; so,
we get a level r(N)-structure sending ((,~) E (/LN 0 0*) X (N)-I/O to
ican,(N)(()qf. E Tatea,b(q) which we write as
cPcan,N : (/LN 00*) X (N)-I /0 ~ Tatea,b(q)[N]

4.1 Hilbert Blumenthal Moduli 117
as long as N is prime to a and b. This level structure is defined over Rs(N),

although Tate a,b(q) itself is defined over the smaller ring Rs (1).
For a positive integer n, we have a perfect pairing of group schemes
Tatea,b(q)[n] x Tateb,a(q)[n]-+ J-tn
given as follows: Taking x E (Gm 0 a* and y E G m 0 b* such that xn = qt.

for ~ E band yn = q'rJ for TJ E a and writing [x] E Tatea,b(q)[n] and [y] E
Tateb,a(q)[n] for the points these x and y represent, we define
([x], [Y])n = x'rJ jyt. E J-tn,
because (x'rJ jyt.)n = q'rJt. jqt.'rJ = 1. Since we may identify the complex
points of the Tate AVRM Tatea,b (q)(q Iq=exp(Tr(z)) with (CX)I je(Lz) (L z =
21fH(bz + a*)) for the variable z E 3, where e : C I -+ (CX)I given by
(u.,.).,. H (exp(u.,.)), the polarization Az induces the above pairing for any z.
Thus the polarization of the universal AVRM over m(c, r(N)) induces an
isomorphism ({J : t Tatea,b(q) ~ Tateb,a(q) so that the Cartier duality pairing
t Tatea,b(q)[M] x Tatea,b (q)[M] -+ J-tM (in (Pl-3) in the proof of Theorem 8.9)
composed with ({J coincides with the above pairing (-, .) M for all M. Then writ-
ing c = ab- 1 , the natural isomorphism
composed with ({J-l gives rise to a canonical c-polarization Aean on Tatea,b(q).

Thus by definition, the level structure <Pean,N is compatible with the Weil
pairing on Tatea,b (q) [N] as required for level r(N)-structures. Similarly, the
polarization Aean combined with i ean ,<J1 also gives a rl(91)-structure (still
denoted by i ean , <J1) of Tate a, b(q).
4.1.6 Hasse Invariant and Sheaves of Cusp Forms
Hereafter we often make the base change of our moduli spaces m, M and M*
of type r = r(N), rl (91), and r 1 (91) to the valuation ring Wand regard these
schemes as defined over W. Assume that W 0 0 = WI by w 0 a H wO"(a);
so, W contains all conjugates of 0 over Z, and p is unramified in 0 jZ. Further
assume that the polarization ideal c is prime to the residual characteristic p
of W. We fix a cone decomposition C as in (PCl-4) once and for all and write
g/M for the universal semi-AVRM with respect to C.
A little more generally, we start with a W-scheme S carrying a semi-
abelian scheme 9 with real multiplication by O. Thus over an open dense
subscheme So c S, 9 Xs So is an object of A F , and over the closed subscheme
Z = S - So, 9 is a smooth multiplicative group scheme of finite type whose
connected component is given by (Gm 0 0 (which is canonically isomorphic to
a more standard (Gm 0,,-1 because of unramifiedness of pin FjQ). Writing
the structure homomorphism of 9 as 1f : 9 -+ S, we have a vector bundle
fo!. = 7r*[2gIS over S. The pullback action of the endomorphism 0 makes !,;Z.
into a schematic TiS-module for T = Reso/zGm. Since p is unramified in
F/«J, identifying X(T) with Z[JJ, we find fo!. = EBaEI!,;Z.a, where fo!.a is the
a-eigenspace of T. Then we define fo!.k = 18J(fo!.a)®ku for k E Z[JJ, which is an
invertible sheaf over S. The sheaf EBkEZ[I] fo!.k of graded Os-algebras is the
affine ring of the scheme M gIs = Specs(S) (see [GME] 1.5.4) representing
the following functor PgI S : S-SCH --+ SETS given by
because having S-morphism S' -4 M is equivalent to having an isomorphism

wa ~ Os' for each a E J, which is equivalent to having a generator w E !,;Z.
over 0 18Jz as,. We note that det(fo!.)®J = I\a(fo!.a)®j ~ !,;Z.JI.
We now construct the Hasse invariant relative to (JIs, Let R be a W-
algebra of characteristic p and (A,w) E PgIs(S') be a pair of a semi-abelian
scheme A over S' = Spec(R) --+ S of relative dimension 9 and a base
W = {WI, ... , w 9 } of HO (A, [2 AI s') over R. We have the absolute F'robenius
endomorphism Pabs : AIR --+ AIR- Let TAls' be the relative tangent bundle;
so, HO(A, TAIs') is spanned by the dual base TJ = TJ(w). For each derivation
D of a A,a, by the Leibniz formula, we have
Thus DP is again a derivation. The association D r-+ DP induces an Fabs-

linear endomorphism F* ofTAls', Then we define H(A,w) E R by F* 119TJ =
H(A, W) 11 9 TJ. Since TJ(AW) = A-ITJ(W) for A E GL 9(R), we see
H(A, AW) 11 9 TJ(AW) = F*(det(A)-1 11 9 TJ(W)) = det(A)-P F* 11 9 TJ(W)

= det(A)-P H(A, w) 11 9 TJ(W) = det(A)I-p H(A, W) 11 9 TJ(AW).
Thus we get H(A, AW) = det(A)I-p H(A, w). We call a semi-abelian scheme A
ordinary if J-l~ can be embedded into AlP] etale locally (i.e., after a faithfully
flat etale base change, we achieve the closed immersion of group schemes,
14 y AlP])· In the same manner as in the elliptic curve case (see Section 3.2.6
and [GME] 2.9.1), we know
H(A,w) =0 ~ A is not ordinary.
Now we return to (Jis given by (JIM for the modulus 9Jt and its toroidal
compactification Mover Z[Nd(F) , J-lN]' By a result of Moret-Bailly ([DAV]
Proposition V.2.1), the graded algebra OM = EBJ2:oHO(M,fo!.JI) is of finite
type over Z[Nd(F) ' J-lN]' Recall the Koecher principle (4.10): HO(M,!,;Z.JI) =
HO (9Jt, fo!.J I) for all j E Z if F -I- «J (the case of F = «J is already treated). Thus
OM = EBJ~OHOU:m,~JI), and ~l is ample on M* = Proj(OM) (however,

~l is not ample on M). Then for a sufficiently large integer a> 0, we have
HO(M*, ~/~-1)1) ®w IF = HO(Mjw, ~/ir-1)1) for the residue field IF of W (by
a theorem of Serre; see [ALG] III Theorem 5.2), and we have a lifting
E E HO(9J1 /w,_ Wa(p-1)1) (4~O) HO(M/w,_
wa(p-1)1) = HO(M*/w,- Wa(P-1)1)
(E)
of Ha; that is, Ha = (E mod p). Since the Hasse invariant is a nontrivial
section of ~(p-1)1 = det(~)P-1 for I = 2::,,- 0', the scheme S* = M*[1;] is
affine and irreducible (because E is a section of an ample line bundle). Then
S* c M* is defined by S* = Spec(OMj(E-1)) for OM = EBJ~o HO(M,~JI).
Define S;,., = S* ®w W m , which is affine, and S:x, = ~mS;'" is an affine
formal scheme. Similarly, we put S = M[~], and define Sm = S ®w W m . We
have a formal scheme Soo = ~mSm' These (formal) schemes Sm, Soo, and
S are not necessarily affine if F =I- Q.
Let Moo be the formal completion of M along the modulo p special fiber
M1 = M ®w W 1 for W 1 = W jpW. Then Soo C Moo is the ordinary locus
of Moo; that is, Soo is the maximal formal subscheme of Moo on which the
connected component Q[p]O of Q[P] is isomorphic to j..l~ etale locally. Let
Tm,n/wrn = Isomo(O ® j..lpn,Q[pnn ~ Isomo(~, OjpnO)

~~ - U ~[n]. (4.11)
Then Tm,njSm is an etale covering with Galois group T(Zjpnz) = (Ojpn)x

for T = Reso /zG m (called the Hilbert modular Igusa tower over Sm). By a re-
sult of K. A. Ribet [Ri] , T m,n is irreducible (see Theorem 4.21 for another proof
of this fact). By the irreducibility, we get the following q-expansion principle
for any p-adically complete W-algebra R = ~mRjpm R (see Corollary 4.23).
(q-exp) f(q) = 0 in R[[q~]]~ ~ f = 0 over Tm,n/R for a section f of the
--
line bundle over T m,n/ R defined at 00.
The sheaf ~oo/soo = OSoo ®zp ~nQ[Pn]o is isomorphic to the dual ~/M of
7r*Lie(Q/M) for 7r : Q --+ M, because Lie(Q[Pn]sJ = Lie(Q/srn) (see Corol-
lary 4.2). In other words, ~/M = Hom(7r*Lie(Q/M), OM) = 7rJ1Q/M is the
algebraization of the formal sheaf ~oo on Soo (which is uniquely determined
by Sl independently of the choice of E). We define ~k C ~k by the invertible
subsheaf of ~k made up of sections of ~k vanishing over the cuspidal divisor
'Dcusp = 7r- 1(M - 9J1) on M. We call ~k the sheaf of cusp forms of weight k.
4.1.7 p-Adic Hilbert Modular Forms of Level r(N)

Fix a sufficiently fine simplicial cone decomposition C of C for cusps of
9J1( c, r(1)) so that we have a smooth projective toroidal compactification
M = Mc(c, r(N)) for all N > 0 (using this C) and the minimal compact-
ification M* = M* (c, r(N)) of 9Jl( c,r(N)).
Fix a positive integer N prime to p. We put S = M[il and S* = M*[il
for a lift E of the Hasse invariant as in (E). Recall Mm = M 0w Wm, Sm =
S0wWm, and S:n = S*0wWm for Wm = Wjpmw = Wjpmw. The proof of
the vertical control theorem we have given for elliptic modular forms extends
to our Hilbert modular case almost intact except for the following hypothesis
(see the second identity in (3.2)),
(Hp1)
for the sheaf ~k C ~k of cusp forms of weight k. In the elliptic modular case,
we have proven this resorting to the affineness of the open curve Sm there. In
the present case of F =1= Q, the scheme Sm is not affine; so, we need to verify
this. This is a subtle point, and this fact is only valid for the cuspidal sheaf
~k not for ~k, as we show.
Write 1f: S -+ S*. Since by definition HO(S,£) = H O(S*,1f*(£)) for any
sheaf £ / sand S* is affine, we need to verify
If this holds, we have
HO(S'~k) 0w Wm = HO(S*,1f*(~k/S)) 0 Wm ~ HO(S*,1f*(~k/S) 0 Wm)
(j HO(S*, 1f*(~k/S 0w Wm)) = HO(Sm, ~k 0w Wm)

as desired, where the identity (1) holds by the affine ness of S* (i.e., the am-
pleness of ~I on M*).
We prove (*) stalk by stalk. Outside the cusps, the two sheaves are the
same; so, nothing to prove. Recall that for an element t; E F, we write t; » 0
if O"(t;) > 0 for all 0" E I. We have for each cusp x associated with the ideals
ab- 1 = c:
-- 1
1f*(~/R)x ~ HO(T(Z)(N),R[[N(abhoJ])
= { L a(t;)ql; E R[[~(abhollla(f2t;) = fka(t;) Iff E T(Z)(N)} ,

I;E~(ab)20
where (ab»o = {t; E ablt;» O}U{O} = Cnab, and f T(Z)(N) in (PC2) acts
E
on R[[tJ(abholl by f· (LI;E(abho a(t;)ql;) = LI;E(ab)2 o c ka(f2t;)ql;. We define

~k by requiring that its stalk at every cusp be given by those q-expansions
vanishing at the cusp. The group cohomology H°(U,7) for U = T(Z)(N)2
commutes with taking the tensor product 0w Wm if Hl (U, 7) = O. The van-
ishing holds if 7 is isomorphic to a product of copies of Hom(Z[U], R) = R U
(e.g., [CGP] III.6.6). This is the case for any R[U]-submodule of R[[*"(abho]l
made up of q-expansions without the constant term. Then we get
(4.12)
Remark 4.8 If k is a multiple weight congruent modulo I(OjpO) x I to a

parallel weight koI, the character k and koI is identical on T(lF p ), and hence
we often have a nontrivial section E E HO(MI,c:!i) congruent modulo p to an
Eisenstein series Eo E HO(M/w,l,,;!..koI) with nontrivial constant term modulo
p. Thus HO(MI,l,,;!..k) is bigger than HO(MI,l,,;!..k)' However, by the Koecher
principle, HO(M/w,l,,;!..k) ®z Q = HO(M/w,l,,;!..k) ®z Q. Thus we conclude that
HO(M/ w , I,,;!..k) ®w WI =I- HO(M I , I,,;!..k) because of W-flatness of 1,,;!..7M'
We put
Vm,n = HO(Tm,n, OT""J, V = ~m Vm,OO,

V = ~mVm,oo, R£ = EBHO(M,l,,;!..k),
V~~~p = HO(Tm,n,OT""n(-Deusp)), Veusp = ~mV~~:', (4.13)

Veusp = ~mV~~:', R~uSP = EBHO(M,l,,;!..k)'
k>£
D£ = ;J(R£)[~]
p
nV, D~uSP = ;J(R~USP)[~]
p
nVcusp,
where Deusp = (M - 9Jl) XM Tm,n is the cuspidal divisor and ;JCLk fk)
2:k ;Jk(ik) for ik E HO(M, I,,;!..k) with the morphism ;Jk sending classical mod-
ular forms of weight k to p-adic modular forms:
;Jk(fk) (A, A, <pp) = fk(A, A, <Pp,*wean )

for the canonical differential Wean induced by the ~®1 on Gm®(')-I. Here k > f
means that ka > fa for all (J" E I. As we show in Section 8.1.4 (Corollary 8.4)
through a formal argument, the subspece D~uSP C veusp is dense under the
p-adic topology of veusp.
In this GL(2)-case, we define in Section 4.1.10 two Hecke operators U(p)
(acting on cusp forms of level divisible by p) and T(p) = T(l,p) (acting on
cusp forms of a level prime to p) preserving integrality of I,,;!..k' The operator
U(p) has its effect on q-expansion a(~, fIU(p)) = a(~p, f) and decreases the
level to pn-I from pn for n > 1, and if k > 21, then T(p) == U(p) mod p. Let
e = limn -4oo U(p)n! (resp. eO = lim n -4oo T(p)n!) be the idempotent attached
to U(p) (resp. T(p)). We attach a subscript or superscript ord to the object
after applying the idempotent e or eO (depending on the setting). From this,
we conclude
Theorem 4.9 Let the notation be as above,- in particular, M = M(c, r(N))
and S = M[~] for a lift E of the Hasse invariant as in (E). Suppose that
N 2: 3 and that p is prime to N d(F)c. Then we have the following facts.
(1) The submodule D~uSP is dense in Vcusp;

(2) The Pontryagin dual vz~1p*(= Homw(Vc':s~, W)) of VZ~1p is a projective
W[[T(Zp)]]-module of finite type;
(3) Vz~1p* 0W[[T(/lp)J],k W ~ Homw (H~rd(S, ~k)' W) if k ::::: 31 (I = Lo- (J),
where the first tensor product is with respect to the algebra homomorphism
W[[T(Zp)]] --+ W induced from the character k sending x to ITo- CJ(x)k u ;
(4) If k ::::: 31, e induces an isomorphism
H~rd(S'~k) ~ H~rd(M'~k)'
where H~rd(S, 7) = eHO(S, 7) and H~rd(M, 7) = eO HO(M, 7).
Assertion (3) in the theorem is (the Pontryagin dual version of) the "precise
control" described in Section 1.2.
Proof. As we remarked, assertion (1) follows from a more general result which
is given in Chapter 8 (see Corollary 8.4) through the theory of false modular
forms. We do not prove the assertion (1) here. It is known that
dimH~rd(M'~k) depends only on k[T(lFp) if k::::: 31. (4.14)
From this, the assertions (2), (3) and (4) follow by the same argument as in
the elliptic modular case (see Section 3.3). The argument giving this exact
bound (4.14) is rather involved; so, we allocate the exposition to its own sec-
tion, Section 4.3 (see Theorem 4.37). Here we just prove the boundedness of
dimH~rd(M'~k) (independent of k) assuming the existence of a Hecke equiv-
ariant inclusion map G k (c, T; q '-t Hg (T, L( k*; q) (the generalized Eichler-
Shimura map for k* = k - 21; see Theorems 4.36 and 5.13) for the arithmetic
subgroup T = T(c, N) of SL 2(F) = GI(Q) giving 9)1(c, T(N))(q = T\3. By
the definition (4.3), T c SL2(O(p)) for O(p) = Op n F. Here g = [F : Q],
but actually, the following argument is valid for any integer g ::::: O. Note that
Gk(C, T; W) 0w C = Gk(c,T; q and Gk(c, T; W) = Gk(C, T; W) 0w W (flat
base change; see [GME] Lemma 1.10.2), because these are spaces of global
sections of the invertible sheaf ~k. By the long exact sequence attached to the
short one: L(k*;W) -4 L(k*;W) ---7> L(k*;lF) for IF = W/pW ([MFG] Corol-
lary 4.28), we find that Hg(T, L(k*; W)) 0w IF injects into Hg(T, L(k*; IF));
so, we need to bound dimIFH~rd(T,L(k*;lF)). We may identify L(k*;R) with
the space of R-integral polynomials in (Xo-, Yo- )o-EI homogeneous of degree
k; = ko- - 2 for each pair (Xo-, Yo-). The action of l' E M2(Op) on L(k*; W) is
given by <P((Xo-, Yo-)t(J(1')'), where xx' = det(x). Let
T' = {a E T[a mod pM2(Op) E U(lFp)}
for the upper triangular unipotent subgroup U c G I . Then [T, r'] =

[GI(lFp)/U(lFp)[ = [PI(O/pO) x (O/pOV [ which is prime to p. Since the ac-
tion of r' on L(k*; IF) factors through U(lFp), the evaluation f(X, Y) at (1,0)
gives a morphism of r'-modules L(k*; IF) ~ IF, where IF is the constant r'-
module with coefficients in IF. Similarly, IF ~ L(k*;lF) given by IF '3 art ayk*
gives a morphism of r' -modules. We get two morphisms of cohomology groups:
which are induced by i and j, respectively. Choose T E M2 (F) which is suf-

ficiently close to (~ c/ ) p-adically and sufficiently close to (6 g)
outside p so
that T normalizes r'. This is possible by the strong approximation theorem
(e.g., [LFE] Theorem 3.2 for F = Q whose proof is valid even for F i= Q). By
a simple computation, the endomorphism j 0 i of the r'-module L(k*; IF) is
the action of T. By the definition of the normalized Hecke operator T(p) by
the following commutative diagram,
T
Hg(r',L(k*;lF)) -----+ Hg(r',L(k*;lF))
ReSr/r f r lTrr/rf
Hg(r, L(k*; IF)) -----+ Hg(r, L(k*; IF))

T(p)
for the restriction Resr/T' and the transfer map Trr/T' (cf. [MFG] 4.3.1),
we find that Trr/T' 0 IT 0 t 0 Resr/T' = T(p). This shows that to ReSr/rf is
injective over H;'rd(r, L(k*; IF)). Since Hg(r', IF) is finite-dimensional (because
9J1(c, r(N)) is homologically equivalent to a finite simplicial complex; e.g.,
[H79] Section 3), the dimension of H;'rd(r',L(k*;lF)) is bounded by that of
Hg(r',lF) independently of k. 0
4.1.8 Moduli Problem of rl(m)-Type
Recall open compact subgroups I\(91), rl(91), and r(91) in G(Z) = GL 2 (O)
defined in (4.5). As long as the moduli problem of type rl (91) is repre-
sentable by a quasi-projective smooth scheme 9J1( c, rl (91)) /Z(p) , for its smooth
projective toroidal compactification M (c, rl (91)) /Z(p) , we get the same as-
sertions as in Theorem 4.9 replacing the pair (9J1(c, r(N)), M(c, r(N))) by
(9J1(c, rl(91)), M(c, rl(91))) for weights k with k E (71.,·1+271.,[1]) (i.e., ku == kT
mod 2 for all (1, T : F '--+ Q). This parity condition is necessary to define
f f-t fk/2 for totally positive units f E T(Z) (since fI = 1 for such units).
We only need the representability of the rlU m 91)-moduli problem for a suf-
ficiently large m and a prime ( to have a result similar to Theorem 4.9 as we
describe this fact now.
To formulate the result, we consider the moduli problem of type r for r =
rl(91) and r 1 (91), and accordingly, when representable, we define D~uSP, V,
V, V~~~p*, and V~~~p with the level r-structure. Even if the modulus problem
is not representable, we have the coarse moduli scheme 9J1( c, r) /W by making
a geometric quotient (cf. [GME] Proposition 1.8.4) of 9J1( c, r(N)) (over W)
by a finite constant group f / f(N) for a suitable N ~ 3 as in the proof of the

following theorem. Then we define the above objects by the submodules (of the
corresponding module of level r(N)) fixed by f / T(N). When it is necessary
to indicate their dependence on (c,r), we write D~USP(c,r), V(c,r), V(c,r),
ord *( C, r) ,an d vord
V cusp cusp (c, r)'In pace
I 0 f DCusP
rd * an d vocusp'
e , V , V , vocusp' rd
Theorem 4.10 Let 91 be an integral ideal prime to p with IJ1 n Z = (N).

Suppose that p is prime to N d(F)c and that p ~ 5 if N = 1. Then we have
for T = Reso/zlGm,
(1) The submodule D~USP(c, rl(IJ1)) is dense in the p-adically complete space
Vcusp (c,rl (91));
(2) The dual V~~~p*(c, rl(IJ1)) = Homw(Vc':s~(c, rl(IJ1)), W) of the module
V~~~p(c, rl(IJ1)) is a projective W[[T(Zp)]]-module of finite type;
(3) V~~~p*(c, rl(IJ1)) ®W[[T(Zp)]],k W ~ Homw(H~rd(M(c, rl(IJ1))[-,€;]'f!d.k)' W)
if k ~ 31;
(4) If k ~ 31, the projector e induces an isomorphism
Proof. We deduce the theorem from Theorem 4.9. First suppose that W con-
tains a primitive Nth root of unity (. For the differental idele d E FJ: with
d(D) = 1 and dO = D, we identify ()-l/N()-l = ()-l ® Z/NZ with O/NO by
x H dx. By m H (m, we have an isomorphism O/NO ~ ()-l®f.LN well-defined
over W. Then 9J1(c, r(N)) over W is the moduli scheme classifying test objects
(A, A, ¢: (O/NO)2 ~ A[N])/R with (¢(l, 0), ¢(O, l))N = (for W-algebras R.
Thus we have a natural action of, E SL 2(O/NO) on 9J1(c,r(N)) induced by
the functorial action on the test objects (A, A, ¢) H (A, A, ¢ 0 I)' The coarse
moduli scheme 9J1(c,rl(N))/w of rhN)-type is given by the geometric quo-
tient 9J1( c, r(N)) /U (cf. [GME] Proposition 1.8.4) by the constant subgroup of
order prime to p, U = {(51) lu EO/NO} c SL 2(O/NO). By the definition
of the quotient, we have
X(c,rl(N)) = HO(U,X) for X = D~uSP, V, V, V~~~p*, and V~~~P'
where X(c, rl(N)) is the object corresponding to X defined relative to the

rl(N)-moduli problems. Note that X(c, rl(N)) is a direct summand of X
(with projection 7r : X -7 X(c, rl(N)) given by 7r(x) = IUI- 1 LuEU u(x)),
because lUI is prime to p. Thus each assertion of the theorem follows from
the corresponding assertion of Theorem 4.9 when IJ1 = (N) with N ~ 3. For
a general W that may not contain (, the object we studied over W[(] is the
scalar extension to W[(] over W; so, by faithfully flat descent, we recover the
desired result for W when IJ1 = (N).
We now treat the case where IJ1 is not necessarily generated by an inte-
ger N ~ 3. The rl (1J1)-moduli problem is representable if each test object
of level rl (1)1) does not have any automorphism other than the identity, be-
cause then the quotient process making 9J1( c, rl (1)1)) out of 9J1( c, r(N)) /w at
the functor side and the moduli side matches. Suppose 1)1 C;; 0, and choose a
prime factor [of 1)1. By the positivity of Rosati involution ([ABV] Section 21),
the automorphism group of a test object (A,),) is finite. By classification of
the endomorphism algebra of a given abelian variety (cf. [ABV] Application
I in Section 21), the size 1Aut(A, )')1 is bounded independently of the test
object. Thus for a sufficiently large power [m, 9J1( c, rl ([ml)1)) is representable
over W, because then each test object of level rl ([ml)1) does not have any
automorphism other than the identity map. Thus the theorem is valid for
level rl ([ml)1). Note that 9J1( c, rl ([ml)1)) is a faithfully fiat finite covering of
9J1( c, rl (1)1)) of degree d prime to p (actually d is a power of N ([)). Since
the restriction composed with the trace map is the multiplication by the de-
gree d, we can realize all objects D~USP(c, rl(I)1)), V(c, rl(I)1)), V(c, rl(I)1)),
vg~~p*(c, rl(I)1)), and vg~~p(c, rl(I)1)) as a W[[T(Zp)]]-direct summand of the
corresponding objects of level rl ([ml)1). By this, we conclude the desired as-
sertions if 1)1 C;; O.
We now suppose that 1)1 = O. Since p is unramified in F/Q and p 2: 5,
Gal(F[p,p]/ F) S=: (Z/pZ) x by sending the Frobenius element at a prime [t p to
N ([) mod p (class field theory). Thus we find a prime [ of F such that N ([)
mod p generates the cyclic group (Z/pZY. In particular p t (N([) ± 1). Then
the modulus variety 9J1( c, r([)) for f is a faithfully fiat finite covering of degree
prime to p over 9J1( c, r(l) ) (the degree is a factor of N ([) (N ([) -1) (N([)2 -1)).
Then by the same argument as above, we get the result for 1)1 = O. D
4.1.9 p-Adic Modular Forms on PGL(2)
We now interpret the result obtained in the previous section in terms of auto-
morphic forms on PC = Reso/zPCL(2). For simplicity, we hereafter assume
that k is even, since the general case has been explained in [SGL] Chapter 2.
We consider a test object (A,O~)"i)/R of level r 1 (1)1). For any ideal 3
prime to Np ((N) = I)1nZ), we make a quotient AI = A/A[3] = A@03- 1 ;
thus, A = A I @3. Then AI[I)1] S=: A[I)1] canonically; so, i induces a level r 1 (1)1)-
structure i l into AI. Let P(A) = {A E Hom(A, tA)lt), = ),} and P+(A) C P(A)
be the subset made of polarizations. Then we have an isomorphism), : P S=: c
of O-modules taking P+ onto the subset c+ of totally positive elements of c.
Dualizing the exact sequence 0 --+ A[3] --+ A --+ AI --+ 0, we get another exact
sequence 0 --+ tAl [3] --+ tAl --+ t A --+ 0, because tAl [3] is the Cartier dual of A[3]
(see the proof of Theorem 8.9). This shows tAl S=: t A @3 S=: A @ C3 S=: AI @ C3 2 ;
so, ), induces ),1 : P(AI)+ S=: (q 2 h. Thus (A, O~)" i) f--+ (AI, O~),I, i/) induces
(3) : V(C3 2,rl(I)1)) S=: V(c,r1 (1)1)) by fl(3)(A,0~)"i) = f(AI,O~),I,i/). The
operator (3) is called the diamond operator. We identify V (c, r 1 (1)1)) and
V(C3 2,rl(I)1)) (resp. V(C,rl(I)1)) and V(C3 2 ,rl(I)1))) by (3), and write it as
Vc (resp. V c) whose ordinary and cuspidal part is written as Vc~s~,c (resp.
V~~:p,J. Thus these spaces only depend on the strict ideal class of c modulo
square ideal classes. We then define
(4.15)
where c runs over strict equivalence classes of ideals modulo square classes;
thus, it runs over the group cltl(Clt)2, where Clt is the strict ideal class
group. Recall that
u
CEC l il(C l i)2
where c runs over a complete representative set for cltl( Clt)2 in FI; Foo =
F 0Q1 JR.; PGLt(Foo ) is the identity-connected component of PGL 2(Foo ),
and 1\(1)1) is the image of i\(I)1) in PGL 2(Fl oo )). Thus we may regard
V~~:p*(i\(I)1)) as the W-dual of the space ofp-adic cusp forms oflevel i\(I)1)
on PGL 2(FA.). We now repeat the vertical control theorem in the setting of
PGL(2).
Theorem 4.11 Let 1)1 be an integral ideal prime to p with 1)1 n Z = (N). Sup-
pose that p is prime to N d(F)c and p 2: 5. Then we have for T = Reso/zGm,
(1) The dualV~~:p*(i\(I)1)) ofv~~:p(h(I)1)) is a projective W[[T(Zp)]]-module
of finite type;
(2) If k 2: 31, V~~:p*(i\(I)1)) 0W[[T(Zp)]],k W is isomorphic to the direct sum
EBcEc l il(C l i)2 Homw(H~rd(M(c, r 1 (1)1))'~k)' W),
where M (c, r 1 (1)1)) is the smooth projective toroidal compactification of the

moduli scheme 9J1( c, r 1 (1)1)).
Proof. Choose a prime [ f pN with p f (N(!)2 - 1). We consider the mixed
moduli problem rl(!m)nn(I)1), whose test objects are (A, A, ¢[, i<Jl)IR with c-
polarization A, a level r 1 (1)1)-structure i<Jl and a level rl ([m )-structure ¢[. This
problem is representable by a moduli scheme 9J1 = 9J1(c, rl(!m) n r 1 (1)1)) for
sufficiently large m. Then 9J1 is a faithfully flat finite covering of the coarse
modulus 9J1( c, r 1 (1)1)) of degree prime to p. Since the result for each piece
V~~:p:c is valid for 9J1, the result can be proven as in the proof of Theorem 4.10.
o
A subgroup C of an abelian scheme AI s is called etale cyclic of order 1)1 if C
is isomorphic to the constant group (0/1)1) IS etale locally. In place of level
r 1 (1)1)-structure, we can think of a quadruple
(A, A, ip : /Jp= 0,,-1 '---+ A[poo], C)
for an etale cyclic subgroup C of order 1)1. The level structure C is called the
level r o(I)1)-structure. Considering this type of level structure, we may con-
struct Vcusp (Fo(I)1)) and Vcusp (fo(I)1)) in the same manner as above, replacing
level r 1 (1)1)-structures by those of level ro (1)1).
More generally, identifying /l-N 129 0* with O/(N) (for N prime to p), we
can let g E GL 2(0/(N)) act on level r(N)-structure ¢N : (0/(N))2 ~ A[N]
by ¢N I--t ¢N 0 g. Thus for a given subgroup r c GL 2(0), we can think of the
orbit ¢ = ¢N 0 r and can think of the moduli problem classifying (A, A, ¢).
We call this level structure ¢ of type r. We write 9J1(c, r) for the coarse
moduli scheme for this modulus problem of type r with the c-polarization
requirement. We again define the module V~~~p (c, r) of p-adic modular forms
for this modulus problem as before.
Since the stabilizer of the standard subgroup (0/91 EB 0) of (0/91)2 is
given by the subgroup fa (91) of G L2 (0), the level ro (91)-structure introduced
above can be considered to be the level structure of type ro(91). Any level
r o(91)-structure (even if we require isomorphisms to keep polarization) has
automorphisms ±1; so, it cannot be representable without adding an extra
structure. Nevertheless, we obtain the following theorem in exactly the same
manner as in the proof of Theorem 4.1l.
Theorem 4.12 Let the notation be as above. Suppose that 91 is an integral
ideal prime to p with 91 n Z = (N) and that p 2': 5 is prime to Nd(F)c. Then
we have for T = Reso/zGm,
(1) The dual V~~~p*(jh(91)) ofV~~~p(.fo(91)) is a projective W[[T(Zp)]]-module
of finite type;
(2) If k 2': 31, V~~~p*(fo(91)) ®W[[T(Zp)]],k W is isomorphic to the direct sum
EBcEc 1tl(C 1t)2 Homw(H~rd(M(c, ro(91))'~k)' W), where M(c, ro(91)) is
the smooth toroidal compactification of 9J1( c, ro (91)).
4.1.10 Heeke Operators on Geometric Modular Forms
We call a ring R a p-adic ring if R = ~ nR/ pn R. We now assume that

the base algebra R is a W-algebra when we consider classical modular forms.
When we consider p-adic modular forms, we further suppose that R is a p-adic
ring; so, it is a W-algebra automatically for W = ~ n W /pnw.
For a prime ideal q of F, we define Hecke operators U(qr) and T(l, qr).
The operator T(l, q) is often written as T(q), but T(l, qr) i= T(qr) if r > l.
If [ and q are distinct prime ideals, we see easily from the definition that the
operators defined for [ and q commute. The operator we need (to define the
projectors e and eO) is U(p) = Il p1p U(p) and T(p) = Il p1p T(p), respectively.
We call a subgroup C of an abelian scheme A is cyclic of order qr if either
C ~ 0/ qr or C ~ /l-qr over an etale faithfully fiat extension of S. We call C
etale cyclic if C ~ O/qr etale locally. We think of test objects (A,A,i,w)ls
for (A,A,i)ls giving a point of 9J1(c,rl(91))(S), where w is a generator of
H O(A, nAI s) over Os ®z O. Assume that S is an R-scheme for a base ring R.
When qlp91, we have additional information C of a cyclic subgroup scheme
C c A of order q given by i(/l-q) if q t p and the unique connected subgroup
C = i(J-Lq) of A[q] if qlp as long as the base ring R is a p-adic ring. Here
"connectedness" is defined as a scheme relative to Spec(R); so, a connected

scheme over Spec(R) has a connected topological space over each connected
component of Spec(R). Since A/R is an O-module, we may consider A[a] =
{x E Alax = O} for an O-ideal a =I- O. For the integer a with an Z = (a),
A[a] C A[a]. The scheme A [a] = Ker(a : A ~ A) is a locally free group
scheme of finite rank, because A is smooth and proper over R (see [GME]
2.8.1). Choosing a generator a of the ideal a/aO, the group scheme A[a] is
given by the kernel of a in A[a], and hence A[a] is locally free of finite rank.
We start with describing the definition of Hecke operators for classical
modular forms, and later we adjust the definition to p-adic modular forms.
We define Hecke operators T(l, qT) (if q t 1)1) and U(qT) (if qll)1) on the space
of modular forms Gk(c, rf(I)1); R) first under the assumption that N(q) is
invertible in the base ring R. For an etale cyclic subgroup G' of A/ s of order
qT, we can make the quotient abelian scheme A' = A/G' with the projection
7r : A ~ A' (cf. [ABV] Section 12 and [GME] Proposition 1.8.4). The c-
polarization A induces a cqT-polarization 7r*A. This can be checked as follows.

Tensoring the exact sequence qT '---+ 0 -» 0/ qT with t A ~ A ® c, we have
another exact sequence A ® cqT[qT] '---+ A ® cqT -» A ® c. Taking the dual of
7r : A ~ A', we have one more exact sequence {} '---+ t A' -» A for the Cartier
........ ......... A,ot7r
dual G' of G', which gives rise to a short exact sequence G' '---+ t A' [qT] -----*
G' ® c. The kernel of the composite (7r ® id) 0 A 0 t7r : t A' ~ A ® c is the
entire qT-torsion subgroup tA'[qT]. Since tA'/tA'[qT] = tA' ® q-T, we have
constructed an isomorphism (7r ® id) 0 A 0 t7r : t A' ® q-T ~ A ® c. Tensoring
qT with this isomorphism, we get the desired A' : t A' ~ A' ® cqT.
Since the projection 7r : A ~ A/G' is etale ({:? 7r* : [l(A/C')/S ~ [lA/S), the
differential w gives a differential (7r*)-lw on A/G'. If qll)1 and the schematic
intersection G' n G = G' XA G is reduced to the identity {O} (in this case, we
say that G' and G are disjoint), 7r 0 i gives rise to the level rf(I)1)-structure
on A/G'. Then we define for f E Gk(cqT, rf(I)1); R),
fIU(qT)(A,A,i,w) = N(l T) Lf(A/G',7r*A,7r o i,(7r*)-lW), (4.16)

q c'
where G' runs over all etale cyclic subgroups of order qT disjoint from G.
Since 7r*A = 7r 0 A 0 t7r is a cqT -polarization, the modular form f has to
be defined for abelian varieties with cqT-polarization. When q t 1)1 and N(q)
is invertible in R, we define the action of the Heeke operator T(l, qT) on
f E Gk(cqT, rf(I)1); R) by
fIT(l,qT)(A,A,i,w) = N(l T) Lf(A/G',7r*A,7r o i,(7r*)-lw), (4.17)

q c'
where G' runs over all etale cyclic subgroups of order qT. We can check that
fIU(qT) and fIT(l, qT) stay in Gk(C, rf(I)1); R) if f E Gk(cqT, rf(I)1); R).
Now we allow the ring R in which N(q) may not be invertible. First assume
that R [N(q)] is flat over R. Then by the flat base change theorem ([ALG]
Proposition III.9.3), we have
The operators are well-defined over R [N(q)]. Thus if it preserves the R-

integral structure, it is well-defined over R. The operator U(qT) is always
integral and T(l, qT) is integral if the weight k is sufficiently positive. We
show in Section 8.3.1 the R-integrality of these operators using the Serre-
Tate deformation theory when R is a p-adic W-algebra. The integrality can
be also checked using the q-expansion principle (which shows that the sum
in (4.16) is actually divisible by N(qT) and T(l, p) == U(p) mod p as long as
k ;::: 21; see Section 4.2.9).
If ~kI is very ample (which holds true if k » 0), the sheaf is generated by
global sections; so, HO (9J1, ~'Jf;p'" R) = HO (9J1, ~~k) ® R/pm R, and
Gk(c,r;R/pmR) = HO (9J1(c,r)'~~kjP"'R)
2:' HO (9J1(c, r)'~~k) ®R R/pm R = Gk(c, r; R) ®R R/pm R
p'"
follows from the long exact sequence from the short one R -=----t R - R/pm R.
Thus if the operators U(qT) and T(l, qT) are R-integral, they are well-defined
over R/pm R. Later we give a definition of these operators as an algebraic cor-
respondence, which works well over a more general ring R (see Section 4.2.5).
We now deal with p-adic modular forms defined over a p-adic ring R clas-
sifying test objects (A,>..,i p : J-Lpoo ®()-l y A[poo],¢)jR' (for R-algebras R')
with prime-to-p level structure ¢ of type r. When q is prime to p, N(q) is
invertible in a p-adic ring; so, we define fIU(qT) (for qllJl) and fIT(l, qT) (for
q f 1Jl) replacing the differential (1T*)-lw by the level pOO-structure 1T 0 ip in
the above definitions (4.16) and (4.17).
To define U(p) (pip) acting on p-adic modular forms, let SoojW be the
formal completion of 9J1(c, r)[~] along its modulo p fiber for the lift of the
Hasse invariant as in (E). Let (X,.x, ¢)j'JJt(c,r) be the universal test object
(supposing the classification functor of prime-to-p level ris representable).
We pick an etale cyclic subgroup C' c X[p]jSoo of order p. As we show in
Section 8.3.1 (particularly, the argument below (8.25)), C' can be defined
only over a locally free covering S'oo/ Soo ofrank N(p), and S'oo/ Soo is radiciel
(purely inseparable over the generic point of Soo). By the universality, we have
a unique morphism Spec( R) Y:.t Soo c 9J1( c, r) with cpo (X, .x, ¢) 2:' (A, >.., ¢).
Then Spec(R') = cp*S,=, and R' is an R-algebra (locally free of rank N(p))
with trace map Tr : R' --+ R (see [K] 3.11), and G' = cp*C' gives an etale cyclic
subgroup of AIR' of order p. Here the trace Tr(x) is given by the coefficient
(times -1) of the characteristic polynomial over R of multiplication by x on
R'. Since A/G' is defined over R', the operator N(p)U(p) for a prime ideal
pip can be defined by
fiN(p)U(p)(A, A, i p , ¢) = Tr(f(A/G', 7f*A, 7f 0 i p , 7f 0 ¢)). (4.18)
If R[~] is flat over R, the trace Tr(f(A/G', 7f*A, 7f 0 ip, 7f 0 ¢)) E R is just
the sum I: c ' f(A/G', 7f*A, 7f 0 i p , 7f 0 ¢) over all cyclic subgroups G' of order
p generically different from the connected component A[P]jR of A[P]IR; so,
(4.18) is compatible with (4.16). The divisibility of the operator N(p)U(p)
by N(p) can be checked by computing either the q-expansion of fiN(p)U(p)
(in Section 4.2.9) or its expansion at a Serre-Tate deformation space (as we
do later in 8.3.1). The representability of the level r modulus problem is not
essential in defining U(p), because we can define it at any level r if we once
have the operator well-defined for a representable r n rl ([m) moduli problem.
There is a shortcut to get the operator U (I') acting on p-adic modular
forms if R = W. By the earlier argument for classical modular forms, we have
a well-defined W-integral operator U(p) on D~uSP C v cusp in (4.13) (for any
e
weight > 0) given by (4.16). Then by the p-adic density of D t SP in vcusp
(Theorems 4.9 and 4.10), the operator U(p) extends to p-adic modular forms,
which of course coincides with the one given by (4.18).
We have from the definition U(qr) = U(qt. If n = (~) for a ~ E P';, then
A H ~-l A gives a bijection between c-polarizations and nc-polarizations; so,
it induces an identification Gk(c, rl(91); R) ~ Gk(nc, rl(91); R) by the asso-
ciation (A,A) H (A,~-lA). Thus we may regard U(O = TIq U(qe(q») for the
prime decomposition n = TIqqe(q) as an endomorphism of Gk(c,rl(91);R).
This really depends on the choice of ~ (because the identification is given
by A H ~-l A), though U(~) is well-defined, up to a unit scalar multi-
ple, independently of the choice of ~. Similarly, we can define T(l,~) by
T(l,~) = TIq T(l, qe(q»). Thus we have a well-defined U(p) as an endomor-
phism of Gk(c, rl(91); R). Similarly, we have T(p) acting on Gk(C, rl(91); R).
In the above definition, we can replace rl (91) by r 1 (91). Since the polar-
ization of a level r 1 (91) test object is specified only modulo multiplication by
O~, U(~) and T(l,~) are well-defined independently of the generator ~ of n;
thus, we may write U(n) and T(l, n) for U(~) and T(1,~), respectively.
Note that the space of modular forms Gk(c, 91; R) depends only on the
strict ideal class of c. So we define
Gk(I\(91);R) = EB Gk(c,91;R), V(i\(91);R) = EB V(c,91;R) (4.19)
for the space of p-adic modular forms V(c, 91; R) over R. The Hecke op-
erators U(q) and T(l, qr) permute the components Gk(C, 91; R) and give
4.2 Hilbert Modular Shimura Varieties 131
endomorph isms of G k (T1('01); R) (or V(T1 ('01); R)). The diamond operator
(3): G k (c,'01;R) -t G k (C3 2,'01;R) (resp. V(c,'01;R) -t V(q2,'01;R)) intro-
duced in the previous section for ideals 3 prime to p'01 also acts on the above
spaces (permuting ideal classes c -t C3 2 ). This is the easiest way of extending
the definition of Hecke operators to GL(2) and is based on the first identity
of (4.7). There is another way of extending the definition of Hecke operators
to G L(2), using the open compact subgroup st ('01) in (4.6) in place of Sl ('01).
Then the approximation theorem yields
U 9Jt(c,r~('01))(q = G(Q)\G(A)lsi('01)Z(~)Co.
CECZt ('Jl)
The above naive definition of the Hecke operator acting on Gdc, rl('01); R)
gives rise to the Hecke operator on
Gk(Tl('01); R) = EB Gk(c, r11('01); R)

CECZt('Jl)
and V (TIl ('01); R) = EB V (C,r11 ('01); R)

CECZt ('Jl)
as an endomorphism. Again A r-+ A Q9 3 induces an operator (3) acting on the

above spaces as long as 3 is prime to p'01.
The natural diagonal embeddings
c'rve c'rvC
are equivariant under the Hecke operator action. Here in the above sum, c'
runs over the classes in Cit ('01) equivalent to c in Cit.
Since Clt('01) = (Fl oo )) x IF!; det(Tl('01)), we may define a quotient Cl r
of Cit ('01) by det(r) for the more general subgroup r c GL 2(O) containing
Tl ('01). Then we define for a p-adic W-algebra R,
v(r; R) = EB V(c, r; R), (4.20)

cECZ r
on which Hecke operators act as we show later.
4.2 Hilbert Modular Shimura Varieties

To translate Hecke operators defined in a geometric manner into group-
theoretic operators (in a more automorphic way), we introduce here the
Hilbert modular Shimura varieties. We give an exposition of more general
Shimura varieties later in Chapters 6 and 7. Let ho : § = ReslC/IFtG m -t G 11Ft be
the homomorphism of real algebraic groups sending a + bA E §(JR) = ex

to the matrix (~b ~). We write X for the conjugacy class of ho under G(JR)
(with origin 0 = ho ). The group G(JR) acts on X from the left by conju-
gation. Since the centralizer of ho is the product of the maximal compact
subgroup of the identity-connected component G(JR)+ of the real Lie group
G(JR) and its center Z(JR), the identity-connected component X+ containing
o = ho is isomorphic by g(O) t--+ g(i) (i = (A, ... , A)) to the product
3 = f:/. Here the action of (ga )aEI E G(JR) with ga = (~: ~: ) on 3 is given by
Z = (za) t--+ (aq Zq
CaZ a
:~<7
0'
). Thus X is a finite disjoint union of the Hermitian sym-
metric domain isomorphic to 3, and for an arithmetic subgroup r c G(Q),
r\X is a finite disjoint union of connected Hilbert modular varieties.
The pair (G, X) satisfies Deligne's axioms for admitting Shimura varieties
in [D2] 2.1.1 (see (Dl-4), (SC), and (CT) in 7.2.1 in the text). The Shimura
variety is a proalgebraic variety (i.e., a projective limit of algebraic varieties)
defined over a canonical field of definition (called the reflex field of (G, X))
equipped with a scheme-theoretic action of G(A (00)) which characterizes the
variety. The complex (pro- ) analytic space of complex points of the Shimura
variety is given by a projective limit under the inclusion relation of open
compact subgroups of G(A(oo)):
Sh(C) = Sh(G,X)(C) = ~KG(Q)\ (X x G(A(OO))) jK
Cj G(Q)\ (X x G(A(OO))) jZ(Q) (~) G(Q)+\ (X+ x G(A(OO))) jZ(Q),

(4.21)
where (')', u) E G(Q) x K acts on (x, g) E X x G(A(oo)) by 'Y(x, g)u =

(')'oo(x),'Y(oo)gu), and Z(Q) is the topological closure of the center Z(Q)
in G (A (00)) (see [EPE] Chapter II for a detailed discussion of the topology
on adele groups). We leave for the reader the proof of the above identity
(*) in (4.21), which is not difficult (see [D2] 2.1.10 or [Mi] page 324 and
Lemma 10.1). The identity (**) follows from the fact that G(Q) acts tran-
sitively on the set 7To(X) of connected components of X (and G(Q)+ is the
stabilizer of X+). On Sh(G,X)(C), the group G(A(oo)) acts by right mul-
tiplication x t--+ xg. This proalgebraic variety has a unique canonical model
Sh(G, X) for G = ResF/QGL(2) defined over the reflex field Q of (G, X), as
we show later. In this section, assuming the represent ability of the classifica-
tion functors introduced in Section 4.1.2, we review the construction of the
model, emphasizing its automorphism group £(G, X) (fixing a geometrically
irreducible component). Strictly speaking, the group £(G,X) we study is a
subgroup of finite index in the full automorphism group, and the full automor-
phism group is a semi-direct product of £ (G, X) with the field automorphism
group Aut(F). As is clear from Shimura's original construction of canonical
models [Sh5] (see also [AAF] Chapter II) and the later treatment of Deligne
[D2], the full knowledge of '£(G, X) is almost equivalent to the existence of

the canonical model itself.
4.2.1 Abelian Varieties up to Isogenies
Let V = F2 be a column vector space, and put V(A(CXl») = VA(oo) := V @IQ>

A(CXl). We often write FA(oo) for F @IQ> A(CXl), which is the finite part of the
adele ring FA = F@IQ>A. Then V(A(CXl») is an FA(oo)-free module ofrank 2. We
consider the fibered category A~ over Q-SCH defined by the following data:
(Object) abelian schemes with real multiplication by 0;
(Morphism) Hom~(A, A') = Homo(A, A') @z Q.
For an object A/ s , we take a geometric point s E S, consider the Tate module
'T(A) = Ts(A) = ~NA[N](k(s)), and define V(A) = Vs(A) = 'T(A)@zA(CXl).
The module V(A) is an FA(oo)-free module of rank 2 and has an O-stable lattice
'T(A), where 0 = 0 @z Z = I1e:prime Oe.
A full level structure on A is an isomorphism 'T/ : V (A (CXl») ~ Vs (A) of
FA(oo)-modules, picking a geometric point s in each connected component of
S. For a closed subgroup K c G(A(CXl»), a level K-structure is the K-orbit
'fj = 'T/K of'T/ for the right action 'T/ H 'T/ 0 U (u E K). For many instances,
we assume K to be open compact. Since A[N]/s is an etale finite group
scheme, the algebrogeometric fundamental group 11"1 (S, s) with base point s
(see Section 4.4) acts on A[N](k(s)) and hence on Vs(A). The level K-structure
is defined over S if ao'fj = 'fj for each a E 11"1 (S, s). If the compatibility ao'fj = 'fj
is valid at one geometric point s for each connected component of S, it is valid
for all s E S (see Section 6.4.1 for the proof of this fact given for Siegel modular
varieties) .
Two polarizations ,X,,XI : A ~ t A are said to be equivalent (written as
,X rv N) if ,X = a'x' = 'x' 0 a for a totally positive a E F. Here a is any
fraction in F'; (not just an integer in 0). Without introducing the category
A~ up to isogeny, our notion of polarization classes does not make sense. The
equivalence class of a polarization ,X defined over S is written as :X. As we have
:x
seen, if the class is defined over S, we can find a polarization ,X E really :x
defined over S. We recall the following condition in Section 4.1.1 imposed on
all triples (A,:X,'fj)/s which is equivalent to (rm4) there.
(det) The characteristic polynomial of each a E 0 on Lie(A) over Os is
given by the image of n,.
(t - a( a)) E Z[t] in Os [t], where a runs over
all embeddings of F into Q.
We consider the following functor from SCH/IQ> into SETS,
£~)(S) = [(A,:X,'fj)/s with (det) I 'fj is a level K-structure].

Here [ ] = { }/ ~ indicates the set of isomorphism classes of the objects
defined over S in the brackets. An F-linear morphism ¢ E Hom~(A, A') is an
isomorphism between triples (A,"X,TJ)/s and (A',"X','i'j')/s if it is compatible

with all data; that is,
¢ 0 'i'j = 'i'j' and t¢ o"X = "X' 0 ¢.
Equip V = F2 with an alternating form A : V I\F V ~ F given by (x, y) =

t xJ1 Y for J 1 = (~ r} ). We define a Q-alternating pairing (-, .) : V x V --+ Q by
TrF/IQI 0 A. Suppose the point s E S is a complex point s E S(C); so, we have
the Betti homology group H 1(A, Q) := H1 (A(k(s)), Q). Then the polarization
A : A --+ t A induces a nondegenerate F-Hermitian alternating pairing E).. :
/\ 2 H1 (A, Q) --+ Q (the Riemann form; see [ABV] Sections 1 and 20 and
also the proof of Theorem 8.9 in the text). Here the word: "F-Hermitian"
means E)..(O'.x, y) = E)..(x, O'.y) for all 0'. E F. We write e).. : V I\F V ~ F for a
unique alternating form satisfying 'frF/lQloe).. = E)... The Hodge decomposition:
H 1(A, C) = HO(A(k(s)), [l'A'irrJ EB HO(A(k(s)), [l~/d induces, by Poincare
duality, an embedding h = hA : C X = §(lR) --+ AutF(H1 (A,lR)) such that
1. h(z)w = zw for all w E Homc(HO(A(k(s)), [lA/d, C) (and h(z)w = zw);
2. E)..(x,h(A)y) is a positive definite Hermitian form on VIR = V 01Q1lR
under the complex structure given by h.
In the above definition of Er:), a condition usually required is missing:
(pol) There exists an F-linear isomorphism 1 : V ~ H1 (A, Q) such that
1- 1 0 hA 01 is a conjugate of ho under G(lR), 1 == TJ mod K under
the canonical isomorphism Vs(A) ~ H 1 (A,A(oo)) = H 1 (A,Q)0IQ1A(oo)
and e)..(f(x), 1(Y)) = 0'.. A(x 1\ y) for some 0'. E FX.
For every nondegenerate F-Hermitian alternating form E on V and an open
compact subgroup K c G(A(oo)), we can find 0'. E G(Q) such that E(x, y) ==
(O'.x,O'.Y) mod K (the strong approximation theorem). Since G(lR) is the full
group of FIR-linear automorphisms of VIR, 1- 1 0 hA 01 is always conjugate to
h o. Thus this condition is redundant; so, we ignore it.
By [Sh3] and [D1] 4.16-21, the canonical model Sh(G, X)/IQI represents
the functor Ei0) over Q for the trivial group 1 made of the identity element.
Through the action of G(A.(oo)) on F1<co)l 9 E G(A.(oo)) acts on the level
structure by TJ f--t TJ 0 9 and hence on the variety Sh( G, X) from the right. If
K is open and sufficiently small (so that Aut((A,"X,TJ)/s) = {I} for all test
objects (A, "X, TJ)/s), ShK(G, X) := (Sh(G, X)! K)/IQI (whose complex points
are given by G(Q)\(X x G(A.(oo)))/ K) represents Er:) over Q.
We now give a brief outline of the proof of the represent ability, reducing
it to the representability of the functors similar to EJ'(N) in Section 4.1.2. By
shrinking K, we may assume that det(K) n O~ c (K n Z(Z))2. This is to
guarantee that the images of gKg- 1 nG 1 (Q) and gKg- 1 nG(Q)+ in PG(Q)
are equal; so, ShK(C) can be embedded into ShK, (C) for K1 = G 1 (A.(oo))nK,
because the moduli problem with respect to K1 is neat without having any
nontrivial automorphisms. Let LeV be an O-lattice. We may assume that

L = n* EB b for a pair (n, b) of two fractional ideals. We define the polarization
ideal c by F :J c* = A( L 1\ L). For each point hz EX, we have a unique point
z E (C - Il~Y fixed by hz (C X) (in this way, we identify 3 with the connected
component X+ of X containing ho ). By changing the identification V = F2,
we may assume that Z E X+ = 3. The action of hz(C X ) on VIR = V 0Q IR
gives a structure of a complex vector space of dimension 9 = [F : Q] on VIR;
that is, VIR = C I via (a,b) H -a+bz = (a,b)J1·t(z, 1) for J 1 = (? (}). Then
L C VIR gives rise to the lattice L z , and A induces the c-polarization Az , and
we recover the data (Lz, Az ) introduced in Section 4.1.3. Set L = L 0z Z C
V~DO) = V 0Q AJoo), and define an abelian variety A z / IC by Az(C) = C I I L z .
Then we have T(A z ) = L, which induces TJz : VA,(oo) ~ V(Az) and gives rise
to a level N-structure ¢N: N-1LIL ~ Az[N] for any N > O.
Let Cl(K) = F:(oo) I det(K)F~ , which is a finite ideal class group. We fix
a complete representative set for Cl(K) in fractional ideals c of F. We choose
an O-lattice Le = n* EB be V as above with A(Le 1\ Le) = c* ({::} nb- 1 = c).
We pick an element (A,~, 1])/s E E~) (S). Let Z(l) = ~ N/LN and identify
it with Z by choosing a primitive Nth root (N = e( -b) in Q. The duality
pairing e : T(A) I\z T(tA) ---» Z(l) = Z (as in (Pl-3) of Section 8.2.3) is 0-
Hermitian (because 0 C End(A/s) is invariant under the Rosati involution).
Thus we have a unique O-linear alternating form eo(A) : T(A) 1\ T(tA) ~
0* such that e = Tr p / Q 0 eo(A). If one makes a choice of a polarization
A E ~, the duality pairing eo composed with A : T(A) --+ T(tA) (and pulled
back by 71) induces an alternating pairing AA : TJ-l(T(A)) 1\ TJ-l(T(A)) ~
C" C Floo ) with AA == A mod det(K). This means that A = AA 0 9 for
9 E G(AJoo)) with det(g) E det(K). Under this condition, the fractional ideal
c is uniquely determined in the complete representative set for Cl(K) we have
chosen, because ~ is a F~-orbit of A and 'ii is the K-orbit of 71.
Case 1: We first assume that TJ(Le) contains T(A). We can naturally iden-
tify C = TJ(Le)IT(A) with a subgroup of A. For each geometric point s E S
and rr E 7rl(S, s), we have rr 0 71 = 71 0 p(rr) for p(rr) E K, and for x = TJ(v)
mod T(A) with v E Le, we have rr(x) = rr(TJ(v)) = TJ(p(rr)v) and p(rr)v E L"
because Le is stable under K. This implies that the subgroup C is stable un-
der 7rl (S, s) for all geometric points s E S, and hence C is a subgroup scheme
of A defined over S. We make the quotient A' = AIC which is an abelian
scheme over S (cf. [ABV] Section 12). For the quotient map 7r : A --+ A', we
AI
define AA' = t 7r - l 0 A 0 7r- 1 . By definition, s has 7rl (S, s )-equivariant exact
sequence 0 --+ T(A) --+ T(A') --+ C --+ 0 and hence V(A') = V(A), which
induces 71 : V~DO) ~ V(A) = V(A') with the exact identity TJ(Le) = T(A').
The polarization of A induces an equivalent polarization AN on A', and we
have i : F y End~(A) = End~(A'). Thus as elements in E~) (S), (A,~, 'ii) /S
and (A', ~A" 1])/s are the same.
Case 2: When T/(Le) C T(A), by the canonical identification of the alge-

braic fundamental group 7f1 (A, 0) = T(A) (see [ABV] Section 18), we have a
unique abelian variety A;s with an F-linear etale isogeny 7f : A' ---» A with
T(A') = T/(Le). Then in the same manner as above, we create (A', XA" r;)ls
giving rise to the same class of (A, X, r;). In particular, AA' = 4rr 0 A 0 7f.
Case 3: If T/eLc) does not have either inclusion relation with T(A), we take
O-lattices L' with L' = T/-I(T(A)) + Le and L" with L" = T/-I(T(A)) n Le.
-II -
Starting from A, first create (A", A ,'ij") IS in the class of (A, A, 'ij) with
T(A") = T/(i'). Then from (A", X", 'ij")ls, we create (A', XA" r;)ls with
T(A') = T/(Le).
By the above argument, in each class of (A, X, r;) IS, we can find a unique
(A', XA" r;) IS with T/(Le) = T(A'). By our choice of Le and c: A(Le 1\ Le) = c*,
actually only Case 3 occurs, because A and AA' are proportional up to units
in det(K). So we may assume that A = AA', because the level structure is
defined modulo K.
In any case, XA, still contains a unique c-polarization AA' made from A.
Thus once we have chosen a c-polarization in Xfor each member (A, X, r;), we
have a unique triple (A', AA', r;) IS with c-polarization AA" If two such choices
are isogenous, the isogeny between them has to be an isomorphism keeping the
polarization. We have proven E~)(S) ~ E~(S) := UeEC1(K) Eke(S), where c
runs over the ideal classes in Cl(K) = FiCO) x / F': det(K), and
Eke(S) = {(A, A, r;)ls with (det) I T/(Le) = T(A) and C(A) = c} / ~ .
Here ~ is the isomorphism (not an isogeny) for a chosen polarization integral

over the fixed lattice Le in the class of X (in other words, A induces a fixed
alternating form on the V integral over Le up to units in det(K)).
We still need to worry that one might not be able to choose A in Xfuncto-
rially (depending on S). In other words, the above bijection E~) (S) ~ Ek (S)
may not be induced from an isomorphism of functors. Under the condition
det(K) n O~ c (K n Z(Z))2, this ambiguity of the choice of AA' does not
matter. Indeed, since AA' == A mod det(K), any c-polarization A' in XA,
inducing A' == A mod det(K) is given by EAA' for E E det(K) n O~. By
det(K) n O~ c (K n Z(Z))2, take c; E K n Z(Z) with C;2 = E. Writing
i : 0 '---+ End(A;s) for the O-multiplication, we have
in E~(S), since we have c;'ij = 'ij because c; E K. Thus the choice of A also does
not matter in the target E~, and the above bijection is indeed induced by an
isomorphism of functors if det(K) n O~ c (K n Z(Z))2.
If we take L = i)-I EEl 0 C V, then LI\L ~ 0* by (a, b) 1\ (a', b' ) H a'b - ab' .
For this choice, the polarization ideal is trivial; that is, c = O. If further, we
identify f-LN with 71.1 N71. by choosing a primitive Nth root ( = (N of unity in
Q[f-LN], having Tj is equivalent to having
Take the open compact subgroup S(N) in (4.6), a conjugate of r(N), stabi-
lizing the lattice L = ()-l EEl O. Over Q[f-LN], the component of ShS(N)(G, X)
corresponding to [~(N),O (which corresponds in turn to G(Q)S(N)G(JR)+ in
(4.7)) represents the now familiar functor:
[r(N) over Q[f-LN]-SCH if L = ()-l EEl 0 c V with c = O. (4.22)
The choice (gives rise to the identification Q[f-LN] = Q[Xl!(<PN(X)) with Q[(]
for the cyclotomic polynomial <PN(X) in 2.2.1, and an automorphism 0' E
Gal(Q[(]/Q) changes the identification by ( H (CT, whose action is induced by
¢N H ¢NO( 8 ?) for a unit c E ZX such that (0' = (c. In other words, the action
of (8n E G(A,(oo)) on Sh(G,X)/Q; brings 9J1(O,r(N))/Q;[(] c ShS(N)(G,X)
to its O'-conjugate 9J1(O,r(N));Q;[(] in ShS(N)(G,X)/Q;'
To accommodate all other components G (Q) ( 8?) S (N) c G (A (00)) is to
change the lattice L to Lc (i. (8 ?)-l) n V. Thus we have
=
[~(N) = U [c,r(N) over Q[f-LN]-SCH, (4.23)

CEC1~(N)
which implies
ShS(N)(G,X)/Q;[j.LN] = U 9J1(c,r(N)) over Q[f-LN]. (4.24)

CEC1~(N)
Note that the identity over Q is wrong, because, over Q, the group scheme
(f-LN ® ()-l) X (01 NO) is not isomorphic to the constant group (()-l IN()-l) x
(OINO). In any case, by the Galois action on 9J1(c,r(N))/Q;[I-'N] described
just below (4.22), we can descend the right-hand side to the base field Q
to obtain the model ShS(N) (G, X) /Q;, because 9J1( c, r(N)) is quasi-projective
as we already observed (see [GME] Example 1.11.1 for the descent). Thus if
F = Q, Sh S(N)(GL(2),X) is irreducible and is isomorphic to the scheme
9J1(1, r(N)) ®Q; Q[f-LN 1 regarded as defined over Q, but not geometrically
irreducible (whose geometrically irreducible components are isomorphic to
Y(N) = 9J1(1, r(N)) over ij). For F i= Q, ShS(N)(G, X) may not even be ir-
reducible over Q, and the picture is not very straightforward (since it involves
an explicit description of the action of G(A(oo)) on the set 7ro(Sh(G, X) /Iji) of
connected components).
Take a sufficiently large integer N so that S(N) C K. Then the coarse
moduli scheme representing [~ is given by the quotient ShS(N)(G,X)IK.
138 4 I-Iilbert Modular Varieties
Here k E K acts through its finite quotient K / S(N) by fj r-t fj 0 k. As before,

if K is sufficiently small so that (A, A, fj) is rigid for all test objects, the
scheme represents the functor. Although E~ can be representable over the
integral base Z [-tv 1' the process bringing EfJ) isomorphically to E~ fully uses
the fact that the group scheme C has order invertible in CQl; so, the functor
EfJ) is representable over CQl but not on the integral base. We write the scheme
representing EfJ) as ShK(G, X) and Sh(G, X) = ¥!!1
KShK(G, X), which is
defined over CQl. Then we recover (4.21) from (4.7).
To construct integral models of Shimura varieties, we use the following
variant (due to Kottwitz [Ko]) of the functor EfJ). We fix a proper subset E
of rational primes. This concerns an open-compact subgroup K maximal at
E (i.e., K = G(ZE) x K(E)), where ZE = I1 pEL' Zp and OE = a 0z ZE.
As usual, we have written K(E) = {x E Klxp = 1 for all pEE}. Recall
A(Eoo) = {x E Alx v = 0 for all v E Eu {oo}}. We identify the multiplicative
group A(ECX!) x with {x E AX Ixv = 1 for all vEE u {oo}}.
We consider the following fibered category A};') over Z(E)-schemes:
(Object) abelian schemes with real multiplication by 0;
(Morphism) We define Hom A(E) (A, A') = HomA F (A, A') 0z Z(E), where
F
This means that to classify test objects, we now allow only isogenies with
degree prime to all primes in E (called "prime-to- E isogenies" ), and the degree
of the polarization A is supposed to be also prime to E. Two polarizations are
equivalent if A = aA' = ,AI 0 a for a totally positive a prime to E.
Fix a prime pEE and an a-lattice LeV = F2 with A(L /\ L) = c*, and
assume self Op-duality of Lp = L 0z Zp under the a-Hermitian alternating
pairing A : V /\ V ~ F. Consider test objects (A,);",fj(E))/S' Here 7](E) :
V(A(1.7°o)) = V 0Ql A(ECX!) ~ V(E)(A) = T(A) 0z A(ECXJ) and A E );" are
supposed to satisfy the following requirement, T(E)(A) /\ T(E)(A) -"-2.t F1ECX!)
is proportional to A : V /\ V ~ F up to scalars in (F 0 A(ECXJ))X. We write
the K(E)-orbit of 7](E) as 'Fj(E). Then we consider the following functor from
Z(l:)-schemes into SETS.
(4.25)
Since all nondegenerate FE-linear alternating forms on VE = V 0Ql CQlE are

isomorphic (see Section 7.l.5) and there is only one isomorphism class of
OE-Iattices in VE , if S is a Z(p)-scheme, we can extend 7](E) : V(A(ECX!)) r-t
V(E)(A) to 7](p) : V(A(PCXJ)) r-t V(p)(A) (keeping rationality over S) so that
77(P) sends the alternation form A : V /\ V ~ F to e).. up to scalars in
(FlpCXJ)y and 7]E-{p}(L E -{p}) = TE-{p}(A). Indeed, defining p: 11"l(S, s) r-t
G(([JlE-{p}) by 0" 0 TJ~_{p} = TJ~_{p} 0 p(O") for a chosen identification rl~'_{p} :

VtZiE-{p}([JlE ~ TE-{p} (A)tZiZ([JlE-{p}, the image of p is in a maximal compact
subgroup of G(([JlE-{p}) , which is a conjugate of G(ZE_{p})' By conjugating
back, we may assume that TJE-{p} is defined over S. By the argument simi-
lar to Steps 1-3, we may further assume TJE-{p}(LE-{p}) = TE_{p}(A). This
extension TJ(p) is unique modulo G(ZE_{p})' In other words, as long as K is
maximal at E, for each test object (A, >:, 17(2;))/S for a Z(p)-scheme S, we have
a unique test object (A, >:, 17(p))/s, where 17(p) is the K(p)-orbit ofl](p), and we
have the following expression on Z(p)-SCH.
(4.26)
Let 0(E)+ = 0;:; n F;. As long as E is finite and K is maximal at E,

we can identify Cl(K) = Fl
oo ) x /F;
det(K) with F1
Eoo ) x /OtE)+ det(K(2;)).
Thus we may choose the representative set {c} prime to p (and hence we may
assume the self-duality of L at p). Assuming that E is finite, by the same
process as bringing Er:) isomorphically to E;"/rQJ' the functor is equivalent
to E;"/z(p) defined over Z(p)-SCH; so, it is representable over Z(p), giving
a canonical model shlf)(G,X)/z(P) of ShK(G,X)/rQJ over Z(p). The functor
E;"/"""-'(p) is a disjoint union of functors E;" , c indexed by c E Cl(K), where
By c(),) = c, the proportionality of A and e A and A(Lp 1\ Lp) = 0P' we must

have A(L 1\ L) = c*
We now move around p inside E. We have reached the functor E;"/z (p)
defined over Z(p)-SCH. Indeed, the subgroup C in the argument has order
prime to p for any given pEE because we removed the E-components from
V(A); so, the process making the quotient can be performed even over Z(E)
(as long as lEI < (0). Since E;" is well-defined over Z(E), moving around p,
we find EkE) is isomorphic to E;" over Z(E), and EkE) is represented by a
scheme shlf) over Z(E)' If E is not finite, we take an increasing sequence of
finite subsets Ei with E = Ui Ell and we have shlf)z(E) = lli!?,sh~'i(El); so,
finiteness of E is not necessary to have an integral model over Z(E) as long
as K is maximal at E.
A subtle point is to relate sh)i;E) to Sh/rQJ' Since the above process of
identifying functors E~~~) and Eg~) ends with the functor E~(N) independent
of E (though their domain Z(E)-SCH depends on E) as long as N is prime
to E, we conclude that Sh(E) tZiZ(E) ([Jl ~ (Sh/G(ZE))' Since 7/ ~ L for
any O-lattice L' c V, it is essential to allow all O-isomorphism classes of
O-lattices L' to define E~(N)' because in the definition of E(E), only DE) is
specified (which does not determine the isomorphism class of L if the class
group of F is nontrivial). This problem is more acute at p because over Z(p),
Tp(A) does not determine Lp. Indeed the p-adic Tate module of an abelian
scheme of characteristic p has less rank than its characteristic 0 counterpart.
The self-duality at p of L has to be imposed to overcome this point (see the
argument just above Remark 7.4). Also we need the density of the derived
group G 1 (Q) in G 1 (A.c oo )) (the strong approximation theorem) in order to
know that geometrically irreducible components of She:) are indexed by the
class group Cl(K): 7ro(ShC:}ij) ~ Cl(K). For Shimura varieties of a more
general pair (G, X), these points have to be studied carefully (see Remark 7.4).
If pEE is unramified in F/Q (and K(E) is sufficiently small), She:) is
smooth over Z(p) by the infinitesimal criterion of smoothness (e.g., [NMDl
Proposition 2.2.6); that is, we can show that any characteristic p test object
lifts to characteristic 0 infinitesimally. To explain this, let R be a Z(p)-algebra
with a nilpotent ideal I C R containing a power of p. Put Ro = R/ I. We
fi6
want to show the existence of a lifting of a test object (Ao, );"0, E ))/Ro to R.
The abelian variety Ao lifts to an abelian scheme A/R (with A @R Ro ~ Ao)
by the deformation theory of Grothendieck-Mumford (cf. [CBTl V.1.6, [GITl
Section 6.3, [DAVlI.3, and also Theorem 8.8 in the text and the remark after
the theorem). Since the degree of polarization is prime to p (here we use the
fact that we can choose a representative c prime to p in a given class in Cl(K)),
A also lifts because we may assume that Ao : Ao ---+ tAo is etale (and hence
t A = A/C for an etale subgroup C C A lifting Ker(Ao); see [ECHlI.3.12). As
for the level structure ry6 E ), it is prime to p and hence etale over Ro. Then it
extends uniquely to a level structure ry(E) : v1 E ) ~ VeE) (A) over R. By the
deformation theory of Barsotti-Tate groups (see [CBTl V.1.6), using (det),
we can find a deformation A/R of AO/Ro with an embedding 0 '---+ End(A/R)
compatible with 0 '---+ End(Ao/Ro) (see Theorem 8.8 and the remark after the
theorem).
When E consists of a single prime p, we write t:f) as t:JI), She:) as Sh~),
and so on.
4.2.2 Global Reciprocity Law
Traditionally a reciprocity law in class field theory is an explicit description

of a field automorphism group (see examples in Chapter 2). The global reci-
procity law of a Shimura variety is such a description of automorphisms of a
geometrically connected component of the Shimura variety as a scheme over
Q. An interesting fact (particular to Shimura varieties) is that all generic auto-
morphisms (of the function field) extend to global automorphisms (see [Mt]);
so, in such a case, the description of scheme automorphisms and field auto-
morphisms of the function field are equivalent. Here we describe Shimura's
global reciprocity law for the Hilbert modular variety.
Shimura's proof of the reciprocity law is based on the analysis of the

topological fundamental group of the variety as a subgroup of the analytic
automorphism group of the bounded symmetric domain; in our setting here,
the domain is the product 3 = f/, and the analytic automorphism group is
given by Aut(3) = 61 I>< PSL 2 CIT?.)f, where 6 1 is the symmetric group of I
permuting the components .fj of 3. See [rAT] 6.6 for a typical example of his
proof of Theorem 2.43 in the elliptic modular case, which can be extended
to more general Shimura varieties (see [Sh5] II and [MiS]). Here we give a
(partially) new proof (once suggested by Ching-Li Chai; see [H03b]) based
on the analysis of the algebraic fundamental group of the variety and the
following theorem of Faltings and Zarhin.
Theorem 4.13 (Faltings and Zarhin) Let K, be a field finitely generated
over the prime subfield of K,. Fix a separable closure K, of K,. Let A/I< and
A/I< be abelian varieties defined over K" and write 7£(A) and 7£(A') for the
e-adic Tate modules of A/;;; and A/;;; as Gal(K,/ K,)-modules, where e is a prime
invertible in K,. Then the natural injection induces the following isomorphism,
Homalg.gp(A/I<, A/I<) ®z 7/.,£ ~ HomGal(;;;/I<) (7£(A), 7£(A')),
and 7£(A) ®z Q is a semi-simple module over Gal(K,/K,).

When K, is a finite field, this is a classical result of Tate ([Tl]) , and Tate
conjectured the above statement to be true; indeed, one can formulate the
conjecture more generally for motives over a field of finite type using e-adic
etale cohomology in place of the Tate modules (see [T2]). The case where
the base field has odd characteristic p > 0 is proved by Zarhin ([Z] and [Zl])
(and later the odd prime condition for p is removed; see [DAV] V.4.7). The
characteristic 0 case is due to Faltings (see [RTP] VI.3). The proof of this
theorem is not given here, since it is beyond the scope of this book.
Take an algebraically closed field k of characteristic O. An irreducible com-
ponent of Sh(G,X) xQ k is called a geometrically irreducible component of
Sh(G, X). Fix an embedding ij '--t k. For each open compact subgroup K,
ShK(G, X) xQ Q has finitely many irreducible components, and irreducibility
is intact after extension of the scalar to k over Q. Thus the notion of a geo-
metrically irreducible component of Sh( G, X) is independent of the choice of
k.
Let us make explicit some geometrically irreducible components of the
Shimura variety Sh(G, X). The functor £hN) : Q-SCH -+ SETS given by
£hN)(k) = [(A'),'¢~)/k] with ¢~ : (j-l/ND- 1 X O/NO ~ A[N] and an
O-polarization ), is representable over Q by 9Jt'(0, r(N))/Q if N ~ 3. Let
9Jt' = ~N9Jt'(O,r(N)). The functorial morphism £hN)(k) 3 (A,),,¢~) f--t
(¢~(l,O),¢~(O,l))N E /-IN(k) (after taking the limit with respect to N) in-
duces a morphism L : 9Jt' -+ Spec(QCYC) for QCYc = UN Q[/-lN] making the
following diagram commutative.
9)1' ~ Sh(G,X)
·1 1n
Spec(QCYC) ------+ Spec(Q) .
Giving ¢'tv becomes equivalent to having a level r(N)-structure over Q[(N]

under the choice of a primitive root (N = (¢'tv(1, 0), ¢'tv(0, l))N (which gives
an identification: P,N ® i)-I with i)-I / Ni)-I). Thus we find that
9)1'(0, r(N)) xQ Q[P,N] ~ U9)1(O, r(N)) xQ Q[P,N], (4.27)

(N
and hence 9)1'(0, r(N))/Q is isomorphic to 9)1(0, r(N)) XQQ[P,N] regarded as

defined over Q. Note that 9)1'(0, r(N)) Y ShS(N)(G, X) by our construction
of the Shimura variety. The variety 9)1(0, r(N)) /Q is geometrically irreducible,
because of the connectedness over <C: 9)1(0, r(N))(C) = r(O, N)\3. Since
{S(N)} N gives a system of open neighborhoods of the identity of G(A (00)),
for a compatible system of Nth roots of unity so that (m = (;:/m (if min) gives
one geometrically irreducible component, which is the image of X+ x 1 in the
expression (4.21) over <C. This component 9)1 is called the neutral component
of the Shimura variety ([D2] Definition 2.1.5).
Let c = (8 ~) for c E F;c oo ). Since r( c, N) = (cS(N)c- I ) n G I (Q) if
cO n F = c, we find that
(4.28)
is the image of 9)1 under the right action of c; so, 9)1c gives another geomet-
rically irreducible component of Sh( G, X) /Qcyc. In other words, QCYc is the
field of definition of the component 9)1c in the sense that QCYc is the algebraic
closure of Q in the function field of 9)1c over Q.
By the description (4.21) of complex points of the Shimura variety, the
group G(A(oo)) acts transitively on 7ro(Sh(G, X) /ij). Thus, up to conjugation
by G(A(oo)), the automorphism group of a geometrically irreducible compo-
nent is independent of the choice of the component.
We now start writing down explicitly the group that gives an open sub-
group of Aut(9)1/Q) with finite index. As before we let 9 E G(A(oo)) act on
Sh(G, X)/Q by
(A,::\,1]) H (A,::\, 1] 0 g),
which gives a right action of G(A(oo)) on Sh(G, X). We write 7(g) for the
automorphism of Sh(G,X)/Q induced by 9 E G(A(oo)). For each point x =
(A,::\,1]) of the Shimura variety Sh(G, X), we can associate a lattice L =
1]-1 (T(A)) C F1c oo )l and a choice of a base W = (WI, W2) of this lattice
gives the identification TJ : 8 2 ~ T( A) by (a, b) H aTJ( wd + bTJ( W2). This

is actually how we related the two functors £f:) and £k-. In other words,
the automorphism x H 7(g)(X) is interpreted into the base change: tw H
g-l . tw because (TJ 0 g)-l(T(A)) = g-lTJ-1(T(A)) = g-lL. Over C, test
object (A,>.,w) is classified by lattices L C C I and its base w. Thus the
action of 7(a) for a E G(Q)+ n G(Z) induces Z H a- 1(z) (see also (4.2)).
Since in Theorem 2.43 we normalized the action using the action on lattices,
it is given by 7(g)-1 and hence GL 2(A) acts on the elliptic modular function
field j{ on the right. In the corresponding theorem, Theorem 4.14 in the Hilbert
modular case, the action is on the left given by the pullback action of 7(g).
Here is the description by Shimura of a group that gives almost all auto-
morphisms of the neutral component 9J1/IQI:
Q(G,X) = {g E G(A)I det(g) E AX FXF~+},

and define £ = £(G,X) = Q(G,X)/Z(Q)G(JR)+ (see [Sh5] II, [Sh6], and
[AAF] Section 8). Here F~+ is the subgroup of totally positive elements in
Foo = F ®IQI R This group £(G, X) gives a subgroup of finite index in the
automorphism group in question.
For a proper subset E of rational primes, we define G(Z17) = ITpE17 G(Zp),
which is a closed subgroup of G(A(oo)). We have the quotient variety
(E)
We write 9J1~~) for the image of 9J1 in Sh(17) (G, X). We define the localization
of Z at E by Z(17) = Q n Z17 inside A (00) for Z17 = ITpE17 Zp. As a convention,
we write Z(0) = Q and Z0 = A(oo). We define
where A(17) X = {x E AX Ixp = 1 if pEE} and 0(17) = ®z Z(17)

put
° C F. We
£(17) = £(17)(G,X) = Q(17)(G,X)/Z(Z(17))G(JR)+,
which gives almost all automorphisms of the component 9J1~~) of Sh(17).

We recall here another definition of £ due to Deligne [D2] (see also
[MiS]), although in the present Hilbert modular case, the definition given
here looks (and indeed is) more complicated than the one already given.
However, this new definition works well for any algebraic group G that ad-
mits Shimura varieties. So we introduce general notation for reductive group
G/IQI, forgetting for the moment about our specific group ResF/IQIGL(2). Re-
call the identity-connected component G(JR)+ of the Lie group G(JR) and
G(Q)+ = G(Q) n G(JR)+. Write Gad = PG for the adjoint group of G (thUS
cad is the algebraic group quotient of C modulo the center Z, but cad(k)
may not equal C(k)/Z(k) depending on the field k). For the quotient map
ad: C -+ cad, we define C(QI)+ = ad-1(cad(QI)+). Often C(QI)+ = C(QI)+,
but there are examples where C(QI)+ ;> C(QI)+ (like some unitary groups).
Here we use the symbol cad in place of PC in order to facilitate an easy
reference to the papers quoted above ([D2] and [MiS]).
We write Z(QI) for the topological closure of the center Z(QI) of C(QI) in
C(A(oo)). Let the adjoint group cad(QI) act on C(A(oo))/Z(QI) by conjugation
and consider the semi-direct product
on which the product law is given by (g,h)(g',h') = (g. ad(h)(g'), hh'), where
ad(h)(g') = hg'h- 1. We now bring the group C(QI)+ into the semi-direct
G(A(oo») d
product Z(IQl) )<I ca (QI)+ by"( H h-1,adh)). Then the image is a nor-
G(A(oo») d
mal subgroup of Z(IQl) )<I ca (QI)+. We define the amalgamated product
G(A(oo») d G(A(oo») d
Z(IQl) *G(IQl)+ ca (QI)+ to be the quotient of Z(IQl) )<I ca (QI)+ by the im-
. . G(A(oo») d +
age of C(QI)+. We wnte [g, adh)] for the class m Z(IQl) *G(IQl)+ ca (QI)
G(A(oo») d
represented by (g,adh)) E Z(IQl) )<I ca (QI)+. Since
-,
is well-defined, we define a new group [ (C, X) by the inverse image of the
closed subgroup A x FX F:"+/ px F:"+ of F; / FX F:"+ in the amalgamated
G(A(oo») d
product Z(IQl) *G(IQl)+ ca (QI)+.
We may axiomatize the above construction as follows (cf. [D2] 2.0.1). Take
three groups Q, r, and L1. We assume to have a homomorphism r 3 "( H -;y
onto a subgroup r
of Q and to have the following commutative diagram of
group homomorphisms.
(4.29)
Q -----+ Aut(Q).
ad
Here Aut(Q) is the automorphism group of the group Q and ad(g)(x) =gxg- 1
for 9 E Q. We often write 'Ph) for 'P(-;Y) and assume that ,(0) (0 E L1) preserves
r as a whole. We suppose the following two compatibility conditions,
(a) '('Ph)) = ad(-;Y) for all "( E r (commutativity of (4.29));
(b) cp(r(bWm = ad(b)(cph)) for all b E ..:1 and, E r, where ad(x)(y) =

xyx- 1.
We consider the semi-direct product: 9><1..:1 whose multiplication law is given
by (g, b)(h, E) = (g. (r(J)(h)), bE), and we have (g, b)-1 = (r(b- 1)(g-I), b- 1).
By computation, we have
(g, b) (,},-1 , cp(,)) (g, b) -1 = (g. r( b) (,},-1 )r( b· cp(,) )(r( b- 1)(g-I)), ad( b) (cp(,))).
Then again by computation,
g. r(b)('}'-I)r(b· cph))(r(b- 1)(g-I))

= g. r(b)('}'-I)r(b)('}'(r(b- 1)(g-I))'}'-I) = g. r(b)((r(b- 1)(g-I))'}'-I)
= g. r(b)(r(b- 1)(g-1 ))(r(b)('}'-I)) = r(b)('}')-I.
This shows that r = {(,},-1, cph)) I, E r} is a normal subgroup of the semi-
direct product 9 ><I ..:1. We then define
(4.30)
We have the following commutative diagram with exact rows.

c -
Ker(cp) -----+ r ~
onto
..:1 -----+ Coker(cp)
111 nl n1 61--+[1,6] 111

Ker(cp) -----+ 9 ) 9 *r..:1 -----+ Coker(cp).
c gl--+[g,l] onto
Then by the suitably applied snake lemma, we get a canonical isomorphism
7'\9 ~ ..:1\(9 *r ..:1). (4.31)
By this isomorphism, the amalgamated product 9 *r..:1 acts on 7'\9, and the
action of [g,b] E 9 *r..:1 on the class [x] in 7'\9 (which is sent to [x, 1] E
..:1\(9 *r ..:1)) is given by
[x] . [g, b] = [x, l][g, b] = [xg, b] = [1, b][r(b)-I(xg), 1] = [r(b)-I(xg), 1].
(4.32)
Suppose that 9 is a topological group with continuous action r of..:1 on 9 under

the discrete topology on ..:1 and that 7' is a closed subgroup of 9. Giving the
group 9 *r ..:1 the quotient topology of the product topology on G ><I ..:1, 9 *r ..:1
becomes a topological group, and 9/ Ker( cp) C 9 *r ..:1 is an open subgroup.
Then (4.31) is a homeomorphism, which is left to the reader to verify.
Let us now return to G = Reso/zGL(2) and G 1 = Reso/zSL(2). Since
Z(A. (00)) n G(Q) = Z(Q) n G(Q) = Z(Q),
we see that Gad(Q) = G(Q)/Z(Q) is a discrete subgroup of C~~)). Taking
with 'P given by "( H adh) and "( H "i given by "( H h mod Z(Q)), we find
G(A (00)) ad + G(A (00)) ad +

-==c-'- *C(Q)+ G (Q) = *cad(Q)+ G (Q) .
Z(Q) Z(Q)
From this it is obvious that the above group is C~~)), and [( G, X) coincides
~ -
with £ (G, X). Thus the two definitions of £ are consistent. We record
(4.33)
by g +-t [g,l].
Take the geometrically irreducible component 9]1/QCYC c Sh( G, X) /Qcyc
containing the image of X+ x 1, and define Aut(9]1/Q) by the scheme auto-
morphism group. The following fact has been shown in [Sh5] II 6.5 and [Mt]
Theorem 2 (see also [MiS] 4.6 and 4.13).
Theorem 4.14 Let E be a proper subset of rational primes (E could be
empty). The stabilizer in G(A(Eoo))/Z(7L(E)) of each geometrically irreducible
component of Sh(E)(G,X)/Q is given by [(E)(G,X). When E = 0, the
right action of (g, adh)) E [(G, X) h E G(Q)+) on [x, g'l is given by
[X,g'] H b-1(x), (glg)ad h )], where (glg)ad h ) = ,,(-l(g'gh. Moreover for the
function field jt(E) of 9]1)~), Aut(9]1)~)) = Aut(jt(E)), and we have a canon-
ically splitting exact sequence of topological groups:
1 ---+ [(E) (G, X) ~ Aut(jt(E)) ---+ Aut(F) ---+ 1,
where Aut(F) is the group of field automorphisms of F. The splitting of the

above projection Aut(jt(E)) -» Aut(F) is induced by the action of a E Aut(F)
on test objects given by (A,:X, i : 0,---+ End(X), 1]) H (A,:X, i 0 a, 1] 0 a).
The neutral component 9]1(0) in the theorem is the tower of canonical models
in the sense of Shimura; so, the above result includes an interpretation in
Deligne's language of a result of Shimura in [Sh5] II 6.5. When we regard
g E [(G,X) as an automorphism of OSh or Sh(G,X)/Q, we write it as T(g).
The topology of the Galois groups is as introduced in Section 2.3; in par-
ticular, Aut(F) has the discrete topology. Note that, a priori, Aut(9]1)~)) is
a subgroup of Aut(jt(E)) for the function field jt(E) of 9]1(E), and we know, a
posteriori, the equality of the two. In any case, the topology of Aut(9]1)~)) is
induced by that of Aut(jt(E)) given in Section 2.3.
The homomorphism T in the theorem is normalized with respect to

G(A/oo)), and the present T(X) is actually T(X)-1 in Theorem 2.43, because
the action on the level structure 7] H 7] 0 0'(00) corresponds to z H a-I (z) by
(4.2). This comes from the fact that we took the column vector space V = p2
with left G(Ql)-action in order to define the level structures 7]. Because of this,
we embedded G(Ql)+ into G(A(oo)) XI Gad(Ql) by a H (a- 1 ,ad(a)).
Over C, by (4.21), we have [z,g] = [,(z),l'g] (<=? [,-I(z),g] = [z,l'g]) for
(z,g) E 3 x G(A(CXJ)) and I' E G(Ql)+, taking the expression
and noting X+ = 3. Each [z, g] carries the triple (Az, Az, 7]z 0 g), where
Az(C) = CI/(O* + Oz) and 7]z (b) = bz - a identifying T(A z) = 0* + Oz.
To see this, we note an easy fact that the map: [z,g] H (A z,AZ)7]z og) sends
(3 x G(A(oo))) surjectively onto [(0)(C). Thus we need to check
(A,-l(z),;\,-l(z),7],-l(Z) og) ~ (A z,;\z,7]z 0l'(oo)g) in A~) for I' E G(Ql)+
which is equivalent to [,-I(z),g] = [z,I'(=)g]. This is exactly (4.2), because

a,07],-l(z) = 7]z 01'(=) for the isogeny a, : A,-l(Z) -t A z given via the
multiplication by (-cz + a) on C 1 (writing I' = (~~)). This gives a sketch
of the verification of Sh(G,X)(C) ~ [i0)(C) which actually has to be done
for each open compact subgroup K c G(A(oo)) (then taking the limit of
isomorphisms with respect to K).
Proof of Theorem 4.14. We now give a proof of the theorem, which also
supplies a proof of the elliptic reciprocity law formulated in Theorem 2.43.
By the description of the complex points Sh(C) ((4.21)) combined with the
identity Sh(G,X)(C) ~ [i0)(C), G(A) acts transitively on the geometri-
cally irreducible components 7fo(Sh(G, X)). Since £(E) (G, X) is normalized
by G(A(Eoo)), we only need to prove the theorem for the neutral component
9]1(E). The beginning of our proof follows the argument given in [IAT] Chap-
ter 6, and then we follow a different path (given in [H03b]).
Our proof of Theorem 4.14 goes through several steps.
Step 1: We first remark that the description of the action of £ (G, X) (defined
by Deligne's amalgamated product) on the variety follows from (4.32), if the
functorial action (A,;\,7]) H (A,;\,7]og) of 9 E G(A,(oo)) induces x H xg
on the complex points of the Shimura variety. This point has already been
verified (see (4.2) and the above argument).
Step 2: £( G, X) gives the stabilizer of the geometrically irreducible component
9]1(E) (and hence also of 9]1~E) for all c) in G(A(ECXJ))/ Z(7L(E)).
By definition, we have Sh(G,X) = ~sShs(G,X) for open compact
subgroups S of G(A,(=)) and, hence, the action of G(A(oo)) on Sh(G,X) is
continuous. Pick a E Z(Ql) = px. Then the isogeny a : A -t A sends
(A,)':, 7]) to (A,~=oo2A,007]=7]ooJoo)).

So, the two test objects (A,~, 7]) and (A,~, 007] = 7]00(00)) give rise to the same
point of Sh(G, X). Thus Z(Q) in G(A/oo)) acts trivially by continuity, and we
obtain a homomorphism G(A(oo))IZ(Q) ~ Aut(Sh(G, X)). The injectivity
of T follows either from the expression (4.21) or from the fact EndlQJ (X) = F
for the universal abelian scheme X/ 9Jl (E) (because X specializes to infinitely
many nonisogenous abelian varieties with complex multiplication by distinct
quadratic extensions of F).
For a given lattice LeV and g(oo) E G(A(oo)), we note g(oo)L 1\ g(oo)L =
det(g(oo))(LI\L); so, if g(oo) preserves 9J1 c , det(g(oo))cn F is strictly equivalent
to c n F. Thus det(g(oo)) = uoo(oo) for a E F"; and u E Ox. By choosing
,E G(Q)+ so that deth) = 00- 1, we find ,(oo)g(oo) has determinant in Ox.
Let QP:;) C QCYc be the maximal extension unramified at all pEE. Thus
it is a union of the cyclotomic field Q[f.LN] for N prime to E. Since 9J1~E)
(for c prime to E) is irreducible and also geometrically irreducible over Q(E),
writing its function field as jt(E) = Q(E) (9J1~E)), any automorphism of jt(E)
over Q(E) extends to jt~E) = jt(E) ®IQJ(E) C, which is the function field of 9J1 c /<C;
thus we have the following facts.
1 --+ Aut(jt(E) IQ(E)) --+ Aut(jt(E)) ~ Gal(Q(E) IQ) --+ 1 is exact;

(4.34)
Aut(jt(E) IQ(E)) ~ Aut(jt~) Ie).
Since the minimal compactification M*(c, r(N)) of 9J1(c, r(N)) over Q is

given by Proj(G(c, r(N); Q)) for G(c, r(N); Q) = EBo<::j GJ1(c, r(N); Q), we
have for N prime to E,
M*(c, r(N)) xlQJ Q(E) = Proj(G(c, r(N); Q) ®IQJ Q(E)).
Thus we can let Gal(Q(E) IQ) act on M*(c, r(N))/1QJ through the Q(E)_
factor of the tensor product G(c,r(N);Q) ®IQJ Q(E). By this "Proj" expres-
sion, the arithmetic action of Gal(Q(E) IQ) preserves the open subscheme
9J1( c, r(N)). Thus we have a canonical section s of the restriction map
Aut (9J1/IQJ) ~ Gal(Q(E) IQ). By the q-expansion principle, the action of
s(u) on f(q) = L.; a(Oq'; E GJ1(c, r(N); Q(E)) is induced by L.; a(Oq'; H
L.; a(OO" q';. On the other hand, by (4.27), the action of c = (0 ?) for e E ZX,
is given by ¢'tv(x, y) H ¢'tv (ex, y) for t(x, y) E ()-1 I N()-l X 0 I NO, which after
identifying f.LN ® ()-1 with ()-lIN()-l by choosing (N E f.LN, corresponds to
the automorphism u E Gal(Q[f.LN]/Q) with ((N)O" = (N (note here the ac-
tion of a scheme automorphism on the structure sheaf is the pullback action).
Identifying ZX with Gal(Qcyc IQ) as in (2.2), we find Gal(Q(E) IQ) ~ Z(E) x,
since 12; C ZX gives the inertia group at pEE. Thus for a section
SA : Z(E)x Y G(A(Eoo))IZ(I2(E)) given by sA(e) = C, we have the follow-
ing commutative diagram,
G(1rJEoo»)/Z(Z(E») ~ Aut(Sh(E)(G, X))
SA r r S
---+ Gal(Q(E) /Q) ,
where L is the Artin reciprocity law map given in Theorem 2.12.

Since A(E)x (resp. Z(E)JR~) is a closed subgroup of FlE) x (resp. A(E)X)
and A(E)XO(E)F~+ is topologically generated by A(E)x and O(E)F~+, we
have
X lOx F X ~ A(E)X/ZX JRx ~ Z(E)X
A(E)xOX(E) F00+ (E) 00+ - (E) + - .
This combined with the above commutative diagram shows that
for K = {s E G(A(E»)I det(s) E O(E)F~+ }. Taking the finite part, we find

K/Koo ~ K(oo) = {s E G(A(Eoo»)1 det(s) E O(E)+} '
where O(E)+ = O(E) n F';. Then the strong approximation theorem (e.g.,
[MFG] Theorem 3.2) shows that g(E)(G, X) = G(Z(E»)+ . SA(Z(E)X). As seen
in (4.2), for Z E 3, the test object (£z,det(a)Az,<PN,z 0 a(oo») is prime-to-
E isogenous to (£"'-'(z), A",-,(z), <p"'-'(z») for an element a E G(Z(E»)+ =
G(Z(E») n G(JR)+. Thus a(oo) (a E G(Z(E»)+) sends a point in 9)1(E)(C) rep-
resented by Z E 3 to that represented by a-1(z), and hence the image of
G(Z(E»)+ in '[(G,X) preserves 9)1(E), in particular G(Z(E»)+ c g(E)(G,X)
preserves the component 9)1(E). We conclude from the continuity that r :
,[(E)(G,X) '-+ Aut(9)1)~»). Since G(QE) = B(ZE)G(ZE) for the upper trian-
gular Borel subgroup BeG, it is easy to see that the normalizer of G (ZE)
is itself. Thus any element g E G(A(oo») preserving Sh(G, X)/G(ZE) has its
E-component gE in G(ZE), and ,[(E) (G, X) gives the stabilizer of 9)1(E) in
G(A(Eoo) G(Z XA(Eoo)
( ) = E () • Up to this point, the proof follows Shimura's
Z Z(E) Z(<Q!)nZ(ZExA Eoo )
argument (see, e.g., [AAF] Section 8).
From now on, we argue slightly differently from [Sh5] II, [Mt], and [MiS].
Step 3: Let ~ = (X,~, </» be the universal test object on 9)1(E). If the pullback
O'*X = X x!m,O' 9)1 for 0' E Aut(9)1(E») is induced by an isogeny a : X ---+ O'*X,
we find O'*</> = a 0 </> = </> 0 g for g E G(A(E») and hence r(g) 00'= 1, which
basically finishes the proof.
Towards this goal, we introduce preparatory propositions. We start with
Proposition 4.15 Let 0' E Aut(jt(E»). Let U C 9)1(E) be an open dense
subscheme on which 0' E Aut(jt(E») induces an isomorphism U ~ O'(U). Let
K be an open compact subgroup of [(E) (G, X), and write UK for the image of
U in S h (E) / K. Let T) be the generic point of UK and Xry be the generic fiber of
the universal abelian scheme X K over 9J1 K . Then the two abelian varieties Xry
and Xu(ry) (for the generic point 0"( T)) of O"(UK)) are isogenous over!fJ( T), 0"( T)))
(which is the composite in .R(E) of the two function fields of UK and O"(UK )).
The proof uses many facts from the theory of algebraic fundamental groups
1fl (X,x) of schemes X, which is summarized in Section 4.4. Here we use
freely without explanation the terminology concerning fundamental groups,
referring the reader to Section 4.4.
Proof. Open compact subgroups K of [(E) (G, X) give a fundamental system
of open neighborhoods of the identity element of Aut(.R(E)) under the Krull
topology. Let 9J1 K = 9J1/ K whose function field is the fixed field of K in .R(E).
Then K = Gal (9J1/9J1 K ) for the stabilizer of 9J1K. If we choose K sufficiently
small depending on 0" E Aut(9J1/IQJII-'N])' we have UK = O"KO"-I. This follows
from the Galois theory in Section 2.3.2. We write UK (resp. Uu K) for the
image of U (resp. of O"(U)) in 9J1 K (resp. in 9J1uK).
By the argument preceding this proposition, the action of the group
G(A(E)) on the set 1fo(Sh~~)) of connected components factors through the
determinant map of G, and the image of G l (A(E)) in Aut(Sh(E)) is con-
tained in the stabilizer Aut (9J1j1QJ) and hence in the field automorphism
group Aut(.R(E)). Let G l (A(E)) be the image of G l (A(E)) in Aut(.R(E)) (so
0 1 (A (E)) is isomorphic to the quotient of G l (A (E)) by the center of G l (Z(E))).
We take a sufficiently small open compact subgroup S of geE) (G, X). We
write SI = G l (A(E)) n Sand S\ (resp. B) for the image of SI (resp. S)
in 0 1 (A (E)) (resp. in E( G, X)). Shrinking S if necessary, we may assume
that 9J1/9J1s is etale, that the geometric Galois group Gal(9J1 /ij/9J1 s /ij) ~
Gal(9J1 /IQJ(E) /9J1 S /IQJ(E)) = B 1, that S = TIe Se with Se = S n G(!fJe) for primes
f! Ef. IJ, and that uBI = O"B I O"-1 C G l (A(E)).
The Galois group Gal(!fJ(E) /!fJ) acts on the underlying topological space
of Sh(E) / S. Thus we have a finite extension ks of !fJ inside !fJ(E) such that
Gal(Q/ ks) gives the stabilizer of 9J1 s in Sh(E) / S. Since the action of geE)
on the set of geometrically irreducible components 1fo(Sh~~)) factors through
Gal(!fJ(E) /!fJ) via the determinant map composed with the Artin reciprocity
map (Step 2), the field ks is given by the fixed field of det(S) in !fJ(E). The
variety 9J1 s is defined over k s . Thus K = kS(T),O"(T))) is a field of finite type
over !fJ; so, we can apply Theorem 4.13 to Xry and Xu(ry) defined over K. We
fix an algebraic closure K of K and take a geometric point Yj E 9J1 s (K) over T).
Let Xs be the universal abelian scheme over Us. Write X7] = Xry ®1QJ(ry) K for
the geometric fiber of Xs at Tj.
Note that 1fl (Xr;/K' 07]) for the origin 07] of the abelian variety X7] is canon-
ically isomorphic to the Tate module T(Xr;/K) (see [ABV] Section 18) and
that f.Ox = 09]1, because X l-t 9J1 is geometrically connected (by the
construction of 9R over C). We have the following exact sequence by The-

orem 4.43, T(X'i)) 4 1fl(XS/ks'O'i)) -+ 1fl(US/ks,fj) -+ 1. This sequence is
split exact, because of the zero section 0 : Us -+ Xs (and covariant functo-
riality of fundamental groups; see Section 4.4.2). The multiplication by N:
X -+ X is an irreducible etale covering, and we conclude that T(E) (Xx) in-
jects into 1fl(XS/ks, Ox) for any geometric point x E Us. We form the quotient
1ff (X S/ ks , O'i)) = 1fl (X S/ ks , O'i))/i(TE(X'i))), and we get a split exact sequence:
0-+ T(E) (X'i)) -+ 1ff(XS/ks'O'i)) -+ 1fl(US/ks,rj) -+ 1.
By this exact sequence, 1fl(US/ks,'fj) acts by conjugation on T(E) (X'i)). Recall

that we have chosen S sufficiently small so that 9R -» 9Rs is etale. We have
a canonical surjection 1fl(US/ks,'fj) -» Gal(U/Us ) = S. Since X'i)[N] for all
integers N outside E gets trivialized over U, the action of 1fl (US/ ks , fj) on
T(E)(X'i)) factors through 1fl(US/ks,'fj) -» Gal(U/Us ) = S.
We now have another split exact sequence:
0-+ T(E) (XO'(7))) -+ 1ff(XaS/ks,OO'('i))) -+ 1fl(UaS/ks,IJ(fj)) -+ 1.
The action of 1fl (Ua S/ks' d'fj)) on T(E) (XO'(ri)) factors through Gal(U /Ua s) =
aS. We fix a path in Uu S from 1J(fj) to 'fj (as in Remark 4.41) and lift it to a
path from dO'i)) to O'i) in Xas, which induces isomorphisms ([SGA] V.7):
to': 1ff(XuS/ks,OO'(7))) ~ 1ff(XuS/ks,O'i)), to': 1fl(UuS,OO'(7))) ~ 1fl(UuS,'fj)·
We want to have the following commutative diagram.
T(E) (X'i)) '-+ ---;>
---+ 1ff(XS/ks'O'i)) ---+ 1fl(US/ks,fj)
?1 1? 0'·1
'-+ ---;>
T(E) (XO'('i))) ---+ 1ff (Xu S/ks' 00'(7))) ---+ 1fl(Uus/ks,dfj))
"1 "u 1 1"u

T(E) (X'i)) ---+ 1ff (Xu S/ks' 071 ) ---+ 1fl (Uu S/ks' fj).
'-+
By definition, T(E) (XO'('i))) ~ T(E) (X'i)) is an isomorphism of oS-modules.

We ask if we can find a linear endomorphism.c E EndA,(E} (T(E) (X'i)) 0Q)
such that .c(8 . x) = 0'8 . .c(x) for all 8 E S and x E T(E) (X'i)), where 0'8 =
13813- 1 , and 0'8 is the image of to' (13* (8)) in 0' S for any lift 8 E 1fl (US, 'fj) inducing
8 E S. Since Hom( G l (Z,,), G l (Ze)) is a singleton made of the zero-map (taking
the entire G l (Z,,) to the identity of G l (Z",)) if two primes e and £' are large
and distinct (see (S3) in Section 4.4.3 for a proof of this fact), 8 H 0' 8 sends
Sl,,, into 0' Sl,£ for almost all primes e, where Sl,£ = G l (Q,,) n S". By [GAN]
IV.3 (see (Sl) in Section 4.4.3), if we shrink S further for exceptional finitely
many primes €, we achieve that S£ is €-profinite, that 8 H 0- 8 sends SI,£ into

SI,£, and that the isomorphism: 8 H 0- 8 is induced by the conjugation of an
0-
element in G(Q£) up to the automorphism of Aut(F£/Q£): 0- 8£ = g£8Igi 1 for

g£ E G(Q£) and a ring automorphism T of F£ over Q£. By (S) in Section 4.4.3,
the isomorphism 8 H 0- 8 is induced by an element l of the automorphism
group Aut(G 1 (A(Eoo))) fitting into the middle term of the exact sequence:
PG(A(Eoo)) --+ II' (Aut(F£/Q£) ~ PG(QR.)) 7rF) Aut(Fl.Eoo )) , (4.35)

£fiE
where PG(Q£) = PGL 2 (F£). Here [1' is the restricted product with respect to
S, and the semi-direct product Aut(R) ~ PG(R) (for the ring automorphism
group Aut(R)) is taken with respect to the action of Aut(R) on matrices entry
by entry. The element l in Aut(G 1 (A(E))) is in turn induced by an endomor-
phism.c E EndA(E) (T(E) (Xry)0Q). Put g(O") = t - 1 o.c. Then g(O") is an element
of HOID-.s'! (T(E)X TJ , T(E)Xo-(TJ)) invertible in Horns, (T(E)X TJ , T(E)Xo-(TJ)) 0zQ,
and is SI-linear in the sense that g(0")(8X) = 0- 8 , g(O")(x) for all s E 8 1 , As
shown in Proposition 4.47, an automorphism of S£ modulo center is conjuga-
tion by an element of G(Q£) up to automorphism of F£/Q£. Thus this property
extends to 8: g(O")1sx) = ((0- 8 , g.0)(x)) for all s E 8 and (E Z(A(E)) n o-S.
The isomorphism S 3 8 H 0- S E S preserves the determinant map, because
the action of G(A(Eoo)) on 7rO(Sh(E)) factors through the determinant map.
Since 8 H 0- 8 preserves the determinant, we conclude (2 = det( () = 1. Choos-
rt
ing a prime € E and shrinking S£ sufficiently (so 0- S also shrinks), we may
assume that (£ = 1 for the €-component (£ of (. Though .c may depend on
the choice of the path (as in Remark 4.41) from 'if to 0"(fJ), the isomorphism
g(O") (modulo the centralizer of SI) is independent of the choice of the path;
so, we forget about the path hereafter.
Since g(O")£ is 8-linear, it is compatible with the action of Gal(~j /'\,), whose
action on TgX TJ factors through 8. Then by Theorem 4.13, we have
g( 0")£ E HomGal(K:/I<) (TgX TJ , TgXo-(TJ)) = Hom(XTJ/I<' Xo-(TJ)/I<) 0z Q£,

and we find that XTJ and Xo-(TJ) are isogenous over /'\,. By this fact, 0" is induced
by an isogeny ~ : XTJ --+ Xo-(TJ) , and hence, we conclude (= 1 and that g(O") is
8-linear. D
We have a projection 7rF : Aut(G 1 (A(Eoo))) --» Autalg(Fl.Eoo ) jA(Eoo)) (as
in (4.35)) whose kernel is given by PG(A(Eoo)) (as shown in Section 4.4.3).
By the above proof, 0" E Aut(jt(E)) has projection to Autalg(FYoo) jA(Eoo))
which is written as O"F = 1l"F(O")'
Corollary 4.16 If 0" E Aut(jt(E)), we have O"F E Aut(F), where Aut(F) is
diagonally embedded into I1£fiE Aut(F£) = Aut (FYoo) jA(Eoo)).
Proof. By the proof of Proposition 4.15, g(O") (modulo 8) is induced by an

isogeny ~ E Homo(XTJ,Xo-(TJ))' Since O-linear isogenies XTJ --+ Xo-(TJ) (keeping
°
the polarization) give rise to the identity coset of Aut(F) v: PG(A(Eoo)) in
Aut(GI(A(Eoo))) (only depending on i : Y End(Xa(7))) with ambiguity
modulo right multiplication by Aut(F)), we conclude O"F E Aut(F). 0
Here is the last step.
Step 4: Replacing 0" by O"O"F I for O"F E Aut(F) c Aut(J't), we may assume
that O"F = id F for the starting 0" E Aut(R(El). Thus we may assume that
g(O")S = OO7)S for an isogeny 007) : X7) -+ Xa(7)). Writing the level structure
on X7) as ¢~E), we find that 007) 0 ¢~E) = ¢~~~) 0 as-I for as E G(A(E)). In
other words, as-I(O"(TJ)) = 71, which implies that as = 0" on the Zariski open
dense subset Us of 9Jt s and, hence, they are equal on the entire 9Jt s . Thus 0"
comes from a scheme automorphism of 9Jt, and 0" is induced by a = lims as E
G(A(Eoo))/Z(7l,(E)). Since [(E) is the stabilizer in G(A(Eoo))/Z(7l,(E)) of9Jt(E)
by Step 2, we conclude that a E [(El. 0
We now study the automorphism group of the mod p fiber of 9Jt(E). To

ensure that the special fiber is smooth, we assume that p is unramified in
F /Q and that E is a proper subset of rational primes containing the prime p.
Let 7l,i~) be the valuation ring of Q(E) C ij corresponding to the embedding
op: Q Y ijp. Let Sh(E)(G,X)rz(E) = ~KSh}J)(G,X) (for K running open
(p)
compact subgroups maximal at E) be the Kottwitz model of the Shimura

variety representing the functor [~E) (classifying triples (A, -X, TJ(E)) up to
prime-to-E isogenies over 7l,((E))). Let 9Jt(E(E) = ~ pfN9Jt( 0, r(N) )/Z(E). Then
p /z(p) (p)
the generic fiber of 9Jt~7f) gives a geometrically irreducible component of the

generic fiber Sh(E) (G, X) /Q(E) of Sh(E) (G, X)/Z(E). If Y'
(p)
-4. Y is a projective
morphism of Noetherian schemes with f*Oy' = Oy, then by Zariski's connect-
edness theorem (e.g., [ALG] III.11.3), each special fiber f-I(y) is connected.
Taking the smooth projective toroidal compactification M(~~) of 9Jt(E(E) '
/z(p) /z(p)
we find that the generic fiber M;~;E) is geometrically irreducible because

M(E)(C) is the toroidal compactification of r(O, N)\3 (which is connected).
This shows 7r*OM(E) = Q(El for 7r : M;~(E) -+ Spec(Q(E)). Since 7l,i~) is nor-
mal, we find that 7r*OM(E) = 7l,((E)) for 7r : M(~~) -+ Spec(7l,((E))) (see [ALG]
p /z(p) p
III.8.8). Then by Zariski's connectedness theorem, M;:) = M(E) 0 z (E) IF re-

(p)
mains irreducible (for the residue field IF C fp of 7l,i~)), and M;:) is the
toroidal compactification of 9Jt;;) = 9Jt(E) 0 Z (E) IF. This shows that 9Jt;;)
(p)
is irreducible. Thus 9Jt;;l gives a geometrically irreducible component of the

special fiber Sh(E)(G,X)/F of Sh(El(G'X)/Z(E). In this sense, we call9Jt(E(E)
(p) /z(p)
1.54 4 Hilbert Modular Varieties
the geometrically irreducible component of Sh(E) (G, X) /Z(E), which contains

(p)
the image of the connected component X+ x 1 of X x G(A(oo)). Define
V(E)(G, X) = {x E G(A(EOO))I det(x) E (pZ)O(p)F::'+} ,
·
w hose Image .
III
("(E)
G
(G , X)'IS gIven
. by
A mod p version of the above proof of Theorem 4.14 gives (see [H03b] for a
detailed proof):
Theorem 4.17 Let the notation be as above. Suppose that pEE and that p is
unramified in F /Q. Write ~~E) for the function field of 'JJt(E) over the residue
field IF of Z~~). Then T in Theorem 4.14 induces the following isomorphism,
Aut(F) ~ ]5(E) (G, X) ~ Aut('JJt)~;) ~ Aut(~~E)).
In particular, the decomposition group inside Aut(5t(E)) = [(E) (G, X) of the

generic point of the special fiber 'JJt)~; is given by V(E) (G, X) c [(E) (G, X),
and we may normalize the isomorphism so that each element 7r E V(E) (G, X)
with det(7r) = p E F1E) X acts on IF by the pth power map (so that V(E) (G, X)
acts on 'JJt)~; from the right).
4.2.3 Local Reciprocity Law
Since Sh(G,X)(C) = G(Q)\ (X x G(A(oo))) /Z(Q), we write [x,g] E Sh(C)

for the image of (x,g) E X x G(A(oo)). A point [x,g] is called a CM point if
hx restricted to G m x 1 C § /e = G m x G m has an image in a maximal torus
T = Tx c G = ResF/QGL(2) defined over Q. Assume [x, g] to be a eM point.
Since Tx(IR) fixes the point x E X, (Tx/Z)(lR.) is in the maximal compact
subgroup fixing x E X; so, it is an anisotropic torus, and Tx(lR.) n G1(lR.)
is compact. Therefore Tx = ResMx/QG m for a totally imaginary quadratic
extension Mx of F inside ij (a CM field over F). In this way, the morphism
hx : § /IR Y G /IR defined over lR. gives rise to an embedding Px : Tx y G /Q
defined over Q. Identify X+ with 3 and write z E 3 for the point corresponding
to x via the embedding ioo : ij Y CC (we fixed in the introduction). We choose
F and Mx in ij. Since px(a)(z) = z, we find
( a)'(ZZ)=(px(Q)-l(Z)px(Q)-l(z))(CZ+d
PX l I I I
_0
cz+d °
)=(zz)(cz+d _0 )
1 1 cz+d , °
where px(a)' = (~d) E GLi(Foo) = G(lR.)+. This shows that r.p(cz + d) =a
in Mx for r.p = ioo IM x ' that Px is the regular representation given by
Px (a) '(ZZ)~(zz)(aO)
l l ~ 1 10C; ,
and that Mx = F[z]. The abelian variety Xx over x has an analytic uni-
formization Xx(CC) = Fc/.c z (Fre = CI) for an O-lattice £Z = bz + cr* c Mx.
Thus End~(Xx) """) Mx and we may identify V(Xx) with Mx ®1Ql A(CXJ).
For the moment, simply write x for [x, 1]. Then the level structure 7]x of
Xx is given by V(A(CXJ)) '3 (a, b) f---t -a + bz E Mx ®1Ql A(CXJ). In particular,
a 0 7]x = 7]x 0 Px (a). Since det(px (a)) = aa e E F for a E Mx for com-
plex conjugation c generating Gal(Mx/F), we find px(a) E Q(G,X). Let
(X,X,1])ISh(G,X) be the universal triple, and f : Sh(G,X) ---+ Sh(G,X)
be an automorphism defined over Q fixing x. Thus f fixes the geometri-
cally irreducible component containing x E Sh(G, X). By Theorem 4.14,
we find g E Q(G, X) such that f = T(g). By the universality of X, we
~,
have a triple (y, A, 1]') ISh(G,X) and an F-linear isogeny a : Y ---+ j*X giv-
ing an isomorphism (y, X', 7]') ~ (1*X, j*X, j*1]) in the category A~, where
j*X = X xSh(G,X),f Sh(G, X) and j*X and j*1] are the pullback of X and 1].
On the other hand, by the definition of the action of T(g), we know that
(1*X, j*X, j*1]) ~ (X, X, 1] 0 g). Thus the isogeny a induces an endomor-
phism ax E End~(Xx), and ax O1]x = 1]x 0 px(a x ) = 1]x 0 g. This shows
that Mx = End~(Xx) and g = px(a x ). In other words, the stabilizer of [x, 1]
in Q(G,X) is given by the image Tx(Q) of M;.
By the de Rham comparison isomorphism combined with the Hodge de-
composition (e.g., [ABV] Chapter 1), we have an Mx-linear isomorphism:
where HO(Xx, J!xxld is the complex conjugation of HO(X x , J!xxld. Since

HO(XxUC), J!xxld is F ®1Ql C-free of rank 1 (rm4), we find ax + a~ E F. Since
ax preserves the polarization Ax (by the expression of the polarization on Az
for z E 3), we find ax a~ E F::. This gives a proof of the fact that F [ax] is a
CM quadratic extension, and Xx has complex multiplication by Mx.
Conversely, start with a test object (A, A, 7]) Ire with complex multiplication
by a CM field M. We find a lattice.c z = 21fi(c~lZ+O*) in Fre = F®IQlC = C I
such that (A,);) Ire is given by (Fc/.c z , )"z). The action of a E M on the
Lie algebra Lie(A(CC)) = C I = Fre is given by multiplication by a diagonal
matrix diag[a'Pl, ... ,a'P9] for field embeddings CPJ : M'--t C whose restriction
gives all elements of I. The set <P = {cpJ} is called the CM type of A, and
<P U <Pc gives all embeddings of Minto C for the generator c of Gal(M/F).
By changing <P by <Pc if necessary (and permuting CPJ)' we may assume that
iCXJIM = CPl, which we write simply as cp. We write <P(a) for the diagonal matrix
diag[a'Pl, ... , a'Pg]. We extend <P to a representation of M ®1Ql lR = c P = C I ,
where the identification M ®lQllR = c P is given by m ®r ~ (cp(m)r )'PEP. Since
.c z ®z Q is a vector space over M under the multiplication by <P(a), it has
to be dimension 1; so, we find that .c z c (M ® 1) c (M ®lQllR) = C P . Then
M = F[z'P].
1.'56 4 Hilbert Modular Varieties
We define 7]x : V(A(oo)) ~ V(A) = Lx 00 Floo ) = Mloo ) (for x = [z,l])

by (a,b) 1-+ -a + bz. Then we find g E G(A(oo)) so that 7]x 0 g = 7]. We first
assume that g = 1. We have a representation Px : M '--7 M2(F) given by
a7]x = 7]x 0 Px (a); in other words, Px is the regular representation (at) =
px(a)' (z;). By our construction, an endomorphism a : A -+ A (a # 0)
induces (A, A, 7]x) ~ (A, (aae)A, 7]x 0 px(a)), and the point x E Sh(G, X)(C)
corresponding to (A, A, 7]x) is fixed by the action of T(px(a)). Thus we have
an embedding Px : TM = ResMxllQiCGm '--7 G(AOO) uniquely associated with
the abelian scheme (A, A, 7]). When g # 1, the embedding p : TM '--7 G(AOO)
is given by t 1-+ p~(t) = g-l Px(t) . g and p is characterized by the fact that
T(p(MX)) fixes the point of Sh(G,X) associated with (A,A,7]).
Returning to (Xx, "Xx, 17x), we thus find that F[a x ] = Mx and Mx = F[z'P]
for the point z E 3 corresponding to x, since the commutative subalgebra
of End~(Xx) of dimension 2 dim(X x ) is unique (see [ACM] Section 5 and
[ABV] Section 20). When g # 1, (A, >:,7]) is sitting over the CM point [x, g] =
T(g)([X, 1]) because 7] = 7]x 0 g. In this case, we take p = p~ = g-l Px' g in place
of Px, and all the arguments work well for p in place of PX' We have proven
Theorem 4.18 Let [x, g] be a CM point. Then there exist a CM quadratic
extension Mx/ F and hx : § '--7 G defined over lR which has an image in
Px(Tx(lR)) for a unique morphism Px : Tx = ResMxllQiCG m '--7 G defined overQ.
Taking the conjugate p~ = g-l Px . g, the action of p~ (M X) on S h( G, X) fixes
the geometric point [x, g] E Sh(G, X)(C), and the image of p~(MX) in G~;»)
gives the stabilizer of the geometric point [x, g]. Conversely, starting a triple
(A, A, 7]) IC with complex multiplication by a CM field M C End~(A), we find a
unique CM point [x, g] E Sh(G, X)(C) such that the fiber (X[x,gj, "X[x,gj, 17[x,gj)
of the universal triple (X,"X,17)ISh(G,X) at [x,g] is isomorphic to (A,>:,7]). In
this case, we have M = Mx and the image of Tx(Q) in Aut(Sh(G, X)) is
isomorphic to T; = Ker(Tx ~ T), where T = ResFllQiCG m .
The last statement comes from the fact that Tx/Z ~ T; by a -+ a l - c , and
Px(Tx(Q))/Z(Q) ~ Px(Tx(Q))Z(Q)/Z(Q).
We write OM = Ox for the integer ring of M = Mx. We start with
a triple (A, >:, 7]) IC of CM type (M, <1» which is sitting on the CM point
[x, g] E Sh(G, X)(C). We first show that we can descend the abelian vari-
ety A IC to a number field Mil in ij. This fact is first proven by Shimura,
Taniyama, and Weil via their theory of complex multiplication (see [ACM]
8.5). We prove this using the represent ability of E~,r(N) by 9J1'(c, r(N)). Sup-
pose that OM '--7 Endo(A). Then A(C) ~ M0IQilR/(o01) for an ideal 0 eM.
E,
The isomorphism class of A only depends on (0,<1», and for a E Aut(C),
A" = A XC,,,-l C has CM type (M, a<1» because we have D~/c = DAG IC
on which M acts by a<1>. Define the reflex field M' of (M, <1» by the fixed
field of {a E Gal(CQl/CQl)iap ~ p}. Since the composite CQlCM of all CM fields
is characterized by the fact that its complex conjugation c on CQlCM is in
the center of Gal(QcM /Q) and hence is unique (see [IAT] Proposition 5.11
and [ACM] 18.2), we see easily that M' is again a CM field. Thus for any
a E Aut(C/M'), the isomorphism class of AO" over C only depends on a. If
,X : t A 2=' A 0 c, then ,XO" : tAO" 2=' AO" 0 c; so, if (A,'x, ¢N) E £:,r(N) (C), its
conjugate is also in £:,r(N)(C)' Since PicA/K is well-defined over K if A is
defined over a subfield K c C, Aut(C/ K) permutes the component group
NS(A)(K) of PiCA/K' Since the Neron~Severi group NS(A)(C) is in bijec-
tion with the set of Riemann forms inside Homz(a I\z a, Z) (see, e.g., [ABV]
Section 2), NS(A)(C) is a discrete torsion-free group with count ably many
elements. Since A is projective, it is defined over a field K generated over Q by
finitely many elements (giving the coefficients of the homogeneous equations
of A). Thus if x E NS(A)(C) is transcendental over K, its conjugates un-
der Aut(C/K) are uncountable, contradicting the countability of NS(A)(C).
Thus a polarization ,x is always defined over a finite extension of K. Therefore
the number of isomorphism classes of (A,'x, ¢N) over C is countably many,
because that of A is bounded by the class number of M times the number of
CM types. This shows the number of conjugates under Aut(C) of the point
x E ry)1'(c,r(N))(C) carrying (A,'x,¢N) is count ably many; so, x has coor-
dinates in a finite extension Mil / M'. Choosing N sufficiently large so that
£:,l(N) is represented by ry)1'(c,r(N)), we find that (A,'x,¢N) is defined over
Mil; in particular, (A,,X) is defined over Mil.
We let G(Q) act on the column vector space V = F2 through the matrix
multiplication. The action of Tx via p~ makes VA a free module over the adele
ring MA of rank 1. Then the subspace Vx = V 01Ql C on which hx acts by its
restriction /1x = hx IG", x I is preserved by multiplication by M x , yielding an
isomorphism class P of representations of M. In this way, we rediscover the
CM type P, because Vx gives rise to the Lie algebra of A(C) as we put the
complex structure on VIR by h x . Since the isomorphism class P is determined
by its diagonal entries cp, : Mx '---+ C, we may identify P with a formal sum
L, cp, or the set {CPI,"" cPg} as before. Thus the field M' is the minimal
field of definition of /1x : <G m -+ G (the half of h x ). We define a morphism
rx : T~ = ResM' /1Ql<G m -+ Tx by
rx: T x' =
R eSM' /1Ql <G m --'--+
f.1x R
eSM' /1Ql T x ------+
Norm T
x'
Since a morphism of the torus ReSM' /1Ql<G m into another torus of the form
ResK/IQl<G m for a number field K can always be realized by x M det op'(x) for
a representation P' of the algebra M' on a vector space over K, we can write
rx = det oP'. Then P' is a CM type of M', whose explicit form is given in
[ACM] Section 8. The pair (M',p') is called the reflex CM type of (M,p).
For each s E T~(A(oo)) = M'~(OOll we have the Artin reciprocity image
[s,M'] E Gal(M~b/M'), where M~b is the maximal abelian extension of M'.
Since Tx(IR) is the stabilizer of x, we have [x,,),g] = b~I(X),g] = [x,g] for
')' E Tx(Q), and the map: [x,g] M [x,px{rx(s))g] = [x,g . p~(rx(s))] only
depends on the Artin symbol [s, M'] E Gal(M,ab /M') by class field theory.
Hi8 4 Hilbert Modular Varieties
We now recall the main theorem of complex multiplication in [ACM] 18.6.

Theorem 4.19 Let (A, A, ry) jfi of eM type (M, p) be the triple corresponding
to a eM point [x,g] E Sh(G,X)(C). Let ~ : M @iQ lR./(a@ 1) ~ A(C) be a
complex uniformization. Let u E Aut(C/M') for the reflex field M' of (M, p).
Take s E M'~ (with soo = 1) corresponding to u by the Arlin reciprocity law;
so, [s,M'] = u on M'/t,. Then (Au,"Xu,ryU) ~ (A,"X,ryop~(rx(s-l))) in A~.
Formulation of the theorem is slightly different from the one given in [ACM]
18.6, which asserts (Au,"Xu,ryU) ~ (A,"X,rx(s-l)ry). This is equivalent to our
formulation by our definition of p~ : Tx(M;c=») y G(A(oo»). There is a mo-
tivic version of this theorem; see [Mi] 1.5.
Shimura's reciprocity law for a CM point (see [ACM] 18.6, 18.8, and [Mi]
II.5.1) follows from Theorem 4.19, since Sh(G,X) represents the functor £?~).
Corollary 4.20 (Reciprocity at a CM point) Let [x, g] be a eM point of
eM type (M,P) in Sh(G,X)(C). Then, we have [x,g] E Sh(G,X)(M,ab) for
the maximal abelian extension M,ab c C of the reflex field M' of (M, p), and
for any s E T~(A(oo»), we have
Proof. We look at the CM test object (A,"X, ry 0 g) as in Theorem 4.19 sitting

over the CM geometric point [x, g] E Sh(G, X)(C). Then for (J E Aut(C/M')
inducing [s, M'] on M,ab, we have by Theorem 4.19,
The test object at the right-hand side is sitting over [x,gp~(rx(s-l))], and
the test object at the left-hand side is sitting over u([x,g]). Thus the two
geometric points CJ([x, g]) and [x, gp~(rx(s-l ))] = [x, px(rx(s-l ))g] are equal
in Sh(G,X)(C): u([x,g]) = [x,gp~(rx(s-l))] = [x,px(rx(s-l))g]. In particu-
lar, the point [x, g] is fixed by Aut(C/ M,ab) by the above expression, and we
have [x, g] E Sh(G, X)(M,ab). Then we may replace u in the above displayed
formula by [s, M'l, which gives the last identity in the corollary. 0
4.2.4 Hilbert Modular Igusa Towers
Let W c Q be the strict Henselization of Z(p) associated with the embedding

ip : Q Y Qp. We look into the Igusa tower TO'. = T1,O'.jIF p in (4.11) over
the toroidal compactification M of the Hilbert-Blumenthal moduli 9Jt. Write
T~ = TO'. x M 9Jt. We prove the following theorem.
Theorem 4.21 (Ribet) Let p be a prime outside N. Then the Jgusa variety
T~ over 9Jt( c, r) jIFp (and hence TO'.) is irreducible for all CY, where r is any
one of r(N), rl(sn), and rl(sn).
This result was first shown by Ribet [RiJ (and [DR]) using the description (due
to Deligne [D]) of p-ordinary AVRMs and density of ordinary points. We give
a proof assuming that p is un ramified in F jQ. Our proof is a generalization
of the proof given for the elliptic Igusa tower (Theorem 3.3), and uses the
description of the automorphism group of the Hilbert modular automorphic
function field (Theorem 4.14) and the existence of a smooth projective toroidal
compactification of Hilbert modular varieties ([Raj and [C]). Another proof of
irreducibility has been proposed in [C] along the lines of the proof in the Siegel
modular case of Chai-Faltings ([DAV] V.7). The point of Chai's argument is
that the monodromy group at the cusp of the Igusa tower is isomorphic to
the p-adic closure of the totally positive unit group 0 ~, and therefore, if the
closure is full in the norm-I-subgroup Ker(N : 0; --* Z;), the irreducibility
follows. The Leopoldt conjecture predicts that the closure is of finite index in
the norm-I-subgroup, but it is easy to create an example (even for a quadratic
F) such that the index is nontrivial; so, Chai's proof works only in limited cases
(see [DT] Section 12 for more details of this point). This type of phenomenon
always occurs when we deal with G of the form ResFIQG o with F ::J Q for
unitary or symplectic groups Go over F and, hence, the computation of the
cuspidal monodromy does not yield irreducibility of the Igusa tower in general.
Proof. Let N > 0 be an integer prime to p, and write S'N = 9J1(c, r(N))[i-J/w
for a lift E of the Hasse invariant; so, E E HO(M'N,!,!,J..a(p-I)I) for the mini-
mal compactification M'N of 9J1( c, r(N)) with E == Ha mod p for the Hasse
invariant H in Section 4.1.6. Recall that S'N is a geometrically irreducible
quasi-projective scheme over W (Le., S'N ®w K and S'N ®w lF p are both geo-
metrically irreducible). We write S'N = M'N[i-J.
Let TN,a be the Igusa tower over S'lv IF (IF = IFp = WI = W j p W) clas-
sifying quadruples (A,A,<PN,i p) for ip: /Lpn ®()-I y A. Taking a toroidal
compactification MN of 9J1(c, r(N)) and its ordinary locus SN = MN[i-L we
can define the Igusa tower TN, a over SN in exactly the same way as TN,a in
(4.11). Since TN, a is etale over SN, the local structure around the etale neigh-
borhood of a cuspidal divisor of TN, a is the same as that of SN; so, we can
blow down TN,a to get the minimal compactification T'N,a of TN,a over S'N. In
other words, T'N,a is the normalization of S'N in TN,a. We used in (4.11) the
etale dual Q[1Pf to construct TN, a over the toroidal compactification SN. We
could have used the maximal etale quotient X[pa]et of the universal abelian
scheme X over 9J1(c, r(N)) to get TN,a C TN, a over S'N (removing the cuspidal
divisors), since ~ S:' X[pa]et over S'N by the polarization. By Koecher's
principle, we have HO(TN a,I,!:d.JI) = HO(T'N a,I,!:d.JI). Thus we do not really need
T'N,a in the sequel, altho~gh we use the symbol T'N,a in order to have well-
defined cusps on the variety. We write Ta = ~ NTN,a and T~ = ~ NTlv,a
for? = * and 0, where N runs over all integers prime to p. Similarly we define
S = ~NSN and S7 = ~NSJv.
We have a tower of irreducible subvarieties {C", C T~}", with closure C;

in T~ containing the infinity cusp. The variety C'" is an irreducible Galois
covering over So. Thus we only need to prove T~ = CrXJ. By the action on
0;
the level r l (pOO )-structure ip I---t ip 0 a, acts on T~ / So, and T~ /0; = So.
Thus we need to show that Gal(Coo/Co) = 0;. In any case, we can identify
Gal(Coo/Co) with a subgroup H of T(Zp) = 0;.
Over Q, we can form a smooth geometrically irreducible toroidal com-
pactification of 9Jl(c, r(N))/1Q for any level including N divisible by p; so, we
have q-expansion at 00 of sections of (,oLjJ (which determines the sections over
9Jl(c,r(N))), and the minimal compactification M*(c,r(N))/1Q is given by
Proj(<Br(N)), where <Br(N) = ffiJ20GJJ(c,r(N);QCYC). Then any element
f in the function field .Jt = QCYC(9Jl c ) (for the component 9Jl c as in (4.28))
can be written as ~ with n,d E GJJ(c, r(N); QCYC) for j ::::: 2. The minimal
compactification is projective and can be embedded into a projective space
using the theta series, which has integral q-expansion coefficients (cf. [CSM]
and [AAF] Section 6). Thus for a given f E .Jt, we can take nand d such that
prn and pr d (for some r > 0) have W-integral q-expansion coefficients. Indeed
GJJ(c, r(N); QCYC) has a W-integral base for j ::::: 2, which can be proven,
for example, using the duality over Q of the Hecke algebra and the space of
modular forms combined with the existence of the Eichler-Shimura isomor-
phism (e.g., [IAT] Theorem 3.52 and [H88] Sections 5 and 6). Thus, writing
the q-expansion of n as L~ a~(n)q~, the valuation
(4.36)
is a well-defined element in Q U {oo}. Similarly v (d) E Q is also well-defined.

Then we define the valuation v : .Jt ---+ Q U {oo} by v(f) = v(n) - v(d).
By Theorem 4.14, we have a split exact sequence,
1 ---+ £(G, X) ~ Aut(.Jt) ---+ Aut(F) ---+ 1,
where r is an isomorphism of topological groups onto its image. Since the

restriction of r(x) to QCYc is given by the action of
in Theorem 2.12, we find another more geometric exact sequence
We write Vp for the decomposition group in r(£(G, X)) C Aut(.Jt) for this
valuation v. As seen in the proof of Theorem 3.3, Vp is a closed subgroup of
r(£(G,X)), and we need to prove that Vp/lp contains 0; = T(Zp) for the
inertia subgroup Ip.
Since the quadruples (A, >.., ¢N, ip : /l-p'" ® i)-I '--+ A) are classified over W
by Sh(G, X)/(S(N) n Sf(p"')), it appears that we only need to look at the
decomposition group V~ of v in Aut(9Jl u /IQCYc) (9Jlu = 9Jl c /Up /IQCYc) for
Up = n
pfN,a
(S(N) n Si(pa)) = {(b n Iu E Op}.
However, we need to study V~ as well as the full decomposition group Vp.

The quotient G 1(A,(poo))/{±1} acts continuously on SjF and SiF' and as
we remarked, So and S* are geometrically irreducible. Since 9J1 u and T~ both
classify quadruples (A, A, TJ(p) : V(A(poo)) ~ V(p)(A), ip : J1 poo ®()-1 Y A[POO])
over QCYc and over Wp and since C;:, contains the infinity cusp, the function
field of Coo is the residue field of the valuation v restricted to the function
field of 9J1u over QCYc. The action of a E T(Zp) on the function field of Coo
and the function field of 9J1 u is induced by the action on the level p-structure:
ip f-t ip 0 a. Since the action of g E G (A (poo)) is given by the action of the
prime-to-p level structure, the action of g E G(A(poo)) on So extends to T~ in
the same way: TJ(p) f-t TJ(p) 0 g. Since the action of T(Zp) = 0;
for the diagonal
torus T of G 1 commutes with the action of G1(A(poo)) and coincides with the
action of 0; on T~/ So, we have a commutative diagram with exact rows,
Res
Aut(Coo ; SO) --+ Aut(SO)
tu tu
y (G 1(A(poo))/{±1}) X H --» G 1(A(poo))/{±1},
where Aut(Coo ; SO) is the group of automorphisms of Coo that induce au-
tomorphisms of So. This commutative diagram shows that the action of
g E G(A(poo)) preserves the irreducible component Coo c T~.
The stability of Coo under G(A(poo)) can also be proven as follows. By
construction, the image CN,a of Coo in TN for an integer N prime to p is a
subscheme of X[ N pal et (for the universal abelian scheme X / S, ), and the action
of g with gp = 1 only concerns the prime-to-p part that is canonically a factor
of X[Npa]et; so, the irreducible component containing 00 is kept under T(g).
Therefore g E G (A (poo)) sends the cusp 00 to another cusp s on the same
irreducible component Coo. This shows that T(g) for g E G 1(A(poo))/{±1} c
[(G,X) preserves v.
There is one more way to show the stability of Coo under G(A(poo)).
Looking at the Tate abelian variety Tateo,c-' (q) = Gm ® ()-l/qO with
level structure cPcan : (J1N ® ()-l) X (a/NO) ~ Tateo,c-1(q)[N] sending
(( ® a,m) to (( ® a)qm/N, we easily see that T (0 ~)-l with (o~) E SL 2(F)
sends q-expansion L~ a~q~ of a modular form to L~ a~ exp(27riTr(5J))q~a/d;
so, the image of B(Q) n G(Q)+ (for the upper triangular Borel subgroup
B(Q) c G(Q)) in [(G, X) preserves v. For the unipotent radical U of B, we
find T(U(A(oo))) C V p, because Vp is closed and U(Q) is dense in U(A(oo)).
Since Dt SP in Theorem 4.9 for the variety 9J1(c, r(N)) is p-adically dense in
vcusP, we find that vt~p = D~uSP ®w IF. We can extend (using q-expansion of
*
elements in VCUSP) the valuation v to D~uSP and f E j{U(Zp) can be written as
for g, h E D~usp for a sufficiently large integer N prime to p. Since g and h
are defined over 9J1( c, F(N)) for N prime to p, their valuations are unchanged
under the action of Gl(A(poo»); so, the valuation v on jtU(Zp) is preserved by
Gl(A(poo»). As we have already seen, 7(U(Zp)) preserves v on jt, Gl(A(poo»)
preserves v on jt, and hence it preserves Coo.
By the stability of v under 7( G l (A (poo»)), the closure of 7(G l (A (oo»)B(Ql))
in Aut(jt) is in Dp. Note that the projection of GdA(oo»)B(Ql) into Gl(Qlp)
contains B(Ql) C G l (Qlp) , which is dense in B(Qlp) (p-adic density of F in
Fp). Thus 'Dp contains the image under 7 of
Since T(Zp) normalizes Up, the decomposition group 'D~ of v in Aut(9J1u /QlCYc)
contains the image of T(Zp) isomorphic to 0;.
As we already remarked, the action of a E T(Zp) on 9J1 u is induced by
the action on test objects (A, A, TJ(p) , ip) f---t (A, A, TJ(p) , ip 0 a). This shows that
Cal(Coo/Co) = 0;,and hence Tn is irreducible over Fp. 0
In the above proof, we have shown T(Qlp) C Dp which acts by automorphisms

on 9J1 c/QI, but a nonunit x E T(Qlp) n Op induces only an endomorphism of
the characteristic p variety T~ (not an automorphism; e.g., p E Op gives the
relative Frobenius pth power map on T~).
Corollary 4.22 Let'Dp C 7(E(G, X)) be the decomposition group of the val-
uation v defined in (4.36). If p is unramified in F /Ql, the morphism 7 in
Theorem 4.14 induces an isomorphism onto 'Dp from the image in E(G, X) of
Proof. We have already shown that 7(B(Ql)) c'Dp and 7(G l (A(poo»)) C Dp.
Since 7 ( 6~) with tEA x acts through the Artin reciprocity on the q-
expansion coefficients at the infinity cusp, we find that 7 ( 6~) with t E Ql;
is in 'Dp. These elements topologically generate the subgroup indicated in the
corollary. Conversely, if 7(g) E'Dp for 9 E Q(G, X), since 7(g) acts by det(g)
on Qlcyc via the Artin reciprocity, we find that det(g) E Ql; FX F~+. Thus
modifying 9 by an element in B(Qlp), we may assume that gp E GL2(Op).
We consider the universal abelian scheme X/So. We then have the projection
map 7r : X[POO] -+ X[poo]et. Since the irreducible component C n is the gen-
erator of X[pn]et, looking at the infinity cusp taking the geometric fiber of
Tateo,c-l (q)[pOO] over Ql, 9 has to preserve fJpoc 0()-1 in Tateo,c-l (q)[pOO]; so,
gp E B(Qlp). This shows the desired identity. 0
We give a slightly different proof of the fact corresponding to Corollary 4.22

for Siegel modular functions in Theorem 6.28. We later further relate Corol-
lary 4.22 to Theorem 6.28 by embedding the general Shimura variety (in-
cluding Hilbert modular varieties) into the Siegel modular variety (see Theo-
rem 8.15 and Theorem 8.16).
We now state as a corollary to Theorem 4.21 a q-expansion principle. We

keep using the notation introduced in the theorem. Let (a, b) be a pair of frac-
tional ideals prime to plJ1 with ab- 1 = c. We have the semi-AVRM Tatea,b(q)
with canonical T-level structure ican and canonical level pCXJ-structure ican,p
as in Section 4.1.5. Since M(c, T(N))/M(c, T(l)) for N prime to p is a fi-
nite Galois covering by the construction of the toroidal compactification, for
any finite subgroup r c Gal(M(c,T(N))/M(c,r(l))), we can make a quo-
tient M(c,r) = M(c,T(N))/T, where T is a symbol indicating a subgroup
between T(N) and T(l) giving rise to r.
Accordingly, we have the minimal
compactification M*(c,r) = Proj(r8r) for r8r = ffiJ?o HO(M(c, r),!,;!}I). By
the same quotient process, we also have the Igusa tower Ta / M (c, T) and the
result of Theorem 4.21 is valid in this slightly more general setting (by our
construction) .
For a cusp (a, b) of M*(c, T), we write
f(Tatea,b(q), Acan , iean , wean) = aa,b(O,f) + L aa,b(~,f)q~.

~E(ab)+
Corollary 4.23 (q-Expansion Principle) Let the notation be as in Theo-

rem 4.21. Suppose that p is unmmified in F/Q, and let R be a Wm-algebm
(Wm = W/pmw). Let Ta be the Igusa tower over M(C,r)/R' Let R' be a
Wm-subalgebm of R over which Ta is well-defined. Let f E HO(Ta/R,~7R) =
HO(T~/R'~7R) with k E Z[I]. We have the following assertions.
(1) f = 0 <===? aa,b(~,f) = 0 for all ~ E (ab)+ U {O};

(2) f E HO(Ta/R"~7R') <===? aa,b(~, f) E R' for all ~ E (ab)+ U {O}.
One may state a similar assertion for p-adic modular forms, which is left for
the reader to formulate.
Proof. The first assertion follows from the irreducibility of the scheme T a .
The same assertion holds for any R-module M and HO(Ta/R'~7R ®R M) for
the same reason. To see (2), we look at the exact sequence
Note that f E HO(Ta/R"~7R') if and only if the image of f vanishes m

HO(Ta / R', ~7 R' ® (R/ R')). Applying the module version of (1) for M = R/ R',
we get (2). D
Let R = ~mR/pmR be a p-adically complete W-algebra. We write Ig

for the Igusa tower TCXJ,CXJ over SCXJ = ~mSh(p) [il
®w R/pm R for a lift E E
HO (SCXJ , ~a(p-1)I) of the Hasse invariant H. For a subgroup K with Sl (pTIJ1) c
K c So(p TIJ1) for an integral ideallJ1 prime to p, the group K/S 1(pTIJ1) can be
identified with a subgroup ofT(Z/pTZ). We write K~ C T(Zp) for the pullback
of this subgroup K/S1(pTIJ1) C T(Z/pTZ). Then by abusing the notation, we
write Ig K for the quotient IgI(K~ x K(p)), where K~ acts on Ig through

the identification of T(Zp) with Gal(IgISh(p)). This is consistent because
IgK gives an open irreducible subscheme of the reduction modulo P of the
integral model over W of Sh K relative to the level K-structure of Drinfeld
type (see [AME] Chapter 3, [GME] 2.6.2 and 2.9, and [GCS] II.2). Thus we
have Igso(pr'J1) = Soo which is independent of r (but Igso(pr'J1) <;: Sh~;('J1)).
4.2.5 Hecke Operators as Algebraic Correspondences
We write ShK for ShK(G, X). For any 9 E G(A(oo)) and open compact sub-
groups K, K' c G(A(oo)), we have two projections PI : ShK9nK' -+ Sh K ,
given by ([x, h] mod Kg n K') H ([x, h] mod K') and Pg : ShKBnK' -+ ShK
given by ([x, h] mod Kg n K') H ([x, hg-l] mod K), where Kg = g-1 Kg
and gK' = gK' g-l. The image of Pg x PI in Sh K xQ Sh K , is an algebraic
correspondence (KgK') C Sh K xQ Sh K , which only depends on the double
coset K gK' as easily verified, using the fact Sh K = Shl K. We now assume
(sl) We have a vector bundle £/Sh (either an algebrogeometric locally free
sheaf of finite rank on the algebraic variety Sh/ Q or a locally constant
sheaf on the topological space Sh(C)) which descends to a sheaf £K/ShK
of the same type for sufficiently small K so that 7r* £ K = £ for the
projection 7r : Sh -+ Sh K ;
(s2) There exists an open semi-group ,1 C G(A(oo)) such that ,1 acts on
£/Sh via g* making the following diagram commutative,
1 1
Sh - - - t Sh,
g-1
where the bottom arrow is given by the action of g-1 E ,1-1 on Sh and
the top arrow is the pullback left action (because the action of G(A(oo))
on Sh is a right action). For an open compact subgroup K c ,1, we
require that the quotient £1 K under the above action k* of k E K give
rise to £K/ShK; so, we have HO(Sh K , 12 K ) = HO(K,HO(Sh,£));
(s3) We have the trace map on the cohomology groups
for K c K' c ,1 satisfying that TrKII/K' 0 TrK'/K = TrK"/K for

K c K' c K", and Tr K' / K 0 Res K' / K for the restriction map Res K' / K
is the multiplication by the covering degree of ShK -+ Sh K ,.
Since Shl Shs is etale and Shs is smooth if S is sufficiently small, choosing
such an open compact subgroup S and a sheaf £s on Shs as in (sl), the
two conditions (sl) and (s3) are satisfied by the quotient C/ K for C = rr* Cs
as long as K c S. As we show in the following discussion, the invertible
sheaf ~i:- = ~'ShK and the locally constant sheaf L( K,; R) K = L( K,; R) / ShK
presented in the introduction satisfy these properties.
Hereafter in this section, we limit ourselves to coherent sheaves CK (topo-
logical sheaves are dealt with in Section 4.3.3). Under (sl-2), for subgroups
K and K' of Ll, we define the Heeke operator [KgK'] : H·(ShK,C K ) --+
H·(ShK"CK') associated with the correspondence (KgK') by
[KgK'] = I det(g)IA' Tr K, /(K9nK') 0 [g] 0 ResK/(Kn9K'), (4.37)
where [g] : H·(SKn9K"CKn9K') --+ H·(ShK9nK"CK9nK') is induced by the

pullback sheaf morphism g* of the left action g-l : ShK9nK' ~ ShKn9 K', and
lalA is the adele norm of a E F;.
In this way, for any open compact subgroup K of G(A,.(oo)) inside Ll, the
double coset ring R(K, Ll) with respect to K c Ll acts on the cohomology
group H·(ShK,CK). Here R(K,Ll) is the free Z-module formally generated
by double cosets KgK for 9 E Ll with the multiplication law as given in [IAT]
3.1 (or [MFG] 3.1.6; see also Section 5.1.2 in the text). We call R(K, Ll) the
double coset ring for K c Ll. An important point is that we need to specify
the action g* of 9 E Ll, and a given sheaf C may have different actions (which
could be canonical from different points of view). The module structure of the
cohomology group over R( K, Ll) depends on the action of Ll on C K.
For subgroups K with Sf(SJt) eKe So(SJt), we define a standard semi-
group Llo(SJt) as follows. Let Ll(I)1) = M 2 (O) n GL 2 (FA ('JI=l). At SJt, we define
D = {(g~) E M 2 (01)1)la EO;', dE 01)1},
and put LlI)1 = (So(SJt)D . So(SJt)) n GL2(FI)1). These semi-groups are char-
acterized by the fact that 8 . U ( 01)1) 8 -1 ::) U ( 01)1) for the upper triangular
unipotent subgroup U c G (so D and LlI)1 are called expanding semi-groups
at SJt). The involution 9 H det(g)g-l = g' brings an expanding semi-group to
the opposite shrinking one. We define Llo(SJt) = LlI)1 x Ll(I)1).
4.2.6 Modular Line Bundles
The diagonal torus Tc of G is isomorphic to T2; so, its weight K, is a pair

(K,1, K,2) of weights of T. We associate with each dominant weight K, and a
"Neb en" character E modulo SJt, a coherent line bundle ~ and its cuspidal sub-
bundle ~I<,E equipped with the action of Llo(SJt) satisfying the three conditions
(sl-3) and giving rise to the space of modular forms or cusp forms of weight
K, and character E. Since we have well-defined ~k and ~k for each dominant
weight k of T on each geometrically connected component of Sh or Sh(p), we

have well-defined ~k and ~k on Sh. Thus our discussion is focused on how
to define a good action of Llo(SJt) on these line bundles. The quotient ~i:
over ShK equipped with thus-defined action g* of 9 E Ll really depends on
the weight K, of Tc (not just the weight k of T) and the character c. This
involves a fair amount of technicality. However, once we succeed, we will have
a coherent line bundle with eigensections (of all Hecke operators) associated
with motives of the Hodge type directly related to the weight K" whose Galois
representations have determinant character dependent only on c and K,. Thus
the reward for our endeavor is worth going through this technicality. We fix
a base ring B which is often one of the following rings: the valuation ring W
of residual characteristic p > 0, its residue field IF, lFp, Z(p), Q, or C (in any
case, B is an algebra over the ones specified above). Thus the line bundle and
its section are supposed to be defined over a B-algebra R.
We now start our work. We have a natural map 9J1'(c, r(N)) -+ ShS(N)
bringing (A, A, ¢N) to (A, >:, 7]) so that 'f) mod S(N) gives rise to ¢N. On
9J1'(c,r(N)), we have a well-defined invertible sheaf ':':l..k. Since ShS(N) is the
union of its geometrically irreducible components 9J1'(c, r(N)), we have well-
defined ':':l..~(N) for N ~ 3 (and its cuspidal counterpart ':':l..k,S(N)) on ShS(N),
which after taking pullback to Sh, gives rise to a line bundle ':':l..k defined over
Sh. This sheaf satisfies (sl) and (s3), but the resulting sheaf ':':l..k and the Hecke
operators acting on it would depend on our choice of the action of d; so, we
need to clarify this to have a clear picture of the modular forms on Sh and
the Hecke operator action.
Let us explain this process of defining ':':l..k more carefully. We first choose
a complete representative set {c} of strict ideal classes of F. We choose c
prime to p. Once c is chosen, we can always specify a c-polarization A in
the polarization class >:. The endomorphism t(a) E End(A) (a E F) acts on
A by t(a)* A = a 2 A for a test object (A, A, 'f)(p)). Thus even if we take a c-
polarization A, for a unit f E Ox, (A, t(f)* A, 'f)(p) a f) is another test object
with c-polarization representing the same point of Sh(p).
The problem in defining the action g* on ':':l..7Sh is that each point of Sh~)
corresponds to an isogeny class (not a single abelian variety); so, we have
inherent ambiguity coming from suppleness of isogeny. In other words, the
point here is to associate without ambiguity a section w k of ':':l..k with each
section w of HO(A, [lA/S) for a given (A, A, 'f)(p)) E EX') (S) only depending on
the point of Sh representing the test object.
We can do this job of specifying w k functorially by (connected) component
by component; so, we may assume that>: determines a strict ideal class of
c. Then we choose a representative A with the given polarization ideal c.
Let L = ()-l ED c 1 C V and A : L 1\0 L ~ c* be induced by the pairing
((a, a'), (b, b')) = a'b - ab'. We have a canonical alternating pairing (-, -).>- :
V(p) (A) x V(p) (t A) -+ FA.'poc) , which via A and the level structure 'f) induces
a linear map A(p) : ,£(p) I\(j(p) £Cp) ~ (?,)(p). We consider a rigidity condition:
(4.38)
If we change TJ by TJ 0 u for u E K(p), this condition fails if det( u) i- 1. Thus

under (4.38), we need to assume that K(p) c GdA(poo)) in order to make the
class fj = TJK(p) compatible with this condition.
For a finite etale morphism A/ s ~ Y/ s , we have a canonical exact se-
quence (e.g., [GME] Proposition 1.5.4) Dy / s ~ DA / S -+ DA / y -+ O. If a is
etale, a* is an isomorphism; so, we have an identification a* : Dy / s ~ DA / S .
If a : A -+ Y is an etale isogeny, we can factor a = (30')'-1 for finite etale
morphisms A' -4 A and A' ~ Y. In this case, we define a* : HO (Y, Dy/ s) ~
HO(A, DA / S ) by a* = h*)-l 0 (3*.
For (A,A,fj(P))/s E t'X')(S) with c-polarization A, as was done in Sec-
tion 4.2.1, bringing the functor t'X') to t'~ isomorphically, we find an etale
isogeny a : (A, A, TJ(p)) -+ (Y,,XI, TJ'(p)) over S with TJ'(p) (L) = T(Y) ®z Z(p)
and a c-polarization A'. For the point x E Sh~)N)(S) representing (A,A,TJ(P))
and w E HO(A, DA / S ), our idea is to take (a-*w),81k as a section (represent-
ing wOk) of ~~(N) at x. Here we have written a-* for (a*)-l for simplic-
ity. This idea works well under (4.38). Indeed, if we have another isogeny
(3 : (A, A, TJ(p)) -+ (Y, A', TJ'(p)) over S with TJ'(p) (L) = T(Y) ®z Z(p), the com-
posite a- 1 0 (3 is an automorphism of (Y,,XI, TJ'(p))/s E t'~(S), which has to
be induced by a unit 10 E Ox with 10 E K and f*,XI = ,XI by (4.38). It is easy to
see that if K(p) c S(N)(p) with p f N ~ 3, the unit 10 has to be trivial. Thus
we have a without ambiguity, and hence (a-*w)0 k without ambiguity.
Without assuming (4.38), just to prove the representability of t'X'), bring-
ing it isomorphically to t'~ by an isogeny normalized with respect to the lattice
L works perfectly well, but this process lacks rigidity to determine the section
wOk of ~k uniquely. Since the central element ~ E Z(Q) acts through the en-
domorphism L(~) E End~(A) on wE HO(A/ c , DA/d by L(~)*W = ~oow, we en-
large our group G(Zp xA(poo)) to G(Zp xA(p))+ := G(Zp) x G(A(poo)) x G(~)+
and let G(Zp xA(p))+ act on Sh(p) through its quotient G(Zp xA(p))+ /G(~)+.
We write Sh~) for Sh(p) / K(oo) for a closed subgroup K in G(A)+ with
Koo = G(~)+ and Kp = G(Zp). Thus we have eased the condition K c
G(Zp) x G 1(A(poo)) to a milder one: K c G(Zp X A(p))+. Each element in
K n Z(Q) is a root of unity up to units in 0; so, if k = mI for an inte-
ger m (and K is sufficiently small), (a-*w)0 mI is uniquely determined, since
102m I = 1 for all 10 E 0 x. Our argument now gives a well-defined ~KI for a
sufficiently small open compact subgroup K c G(Zp X A(p))+ (maximal at
p) and a well-defined ~~ for general k with a sufficiently small open compact
subgroup K c G(Zp) x G 1(A(poo)) x G(~)+.
We write K1 = (KnG 1 (A(poo))) x G(~)+ for a given K in G(Zp x A(p))+.
We have a well-defined ~k over Sh~; = Sh(p) /Ki oo ) for a sufficiently small
K; so, we have ~k over sMp) just by pullback. If we can define a good action
of G(Zp x A(p))+ on ~k over Sh(p), we can descend ~~1 to ~~ so that the
global sections over Sh~) are made up of global sections over Sh~; fixed
by K. Similarly, if we can define a good action of Ta(Zp) x G(A,.(p»)+ on
I£.k over the Igusa tower 19I Sh(p) , we can descend I£.~ I 9 to I£.} for K with
Sl(pn)p C Kp c G(Zp).
For the moment, we assume that the base ring B is W. To define the desired
W-integral action of ..1, starting with a K 1-test object (A, A, 'i'j(p) 0 g-l), we
define (Yg , Ag , 'i'jg (p») E £~Kl for 9 K1 = gK1g- 1 equivalent to (A, A, 1](p) 0 g-l)
uniquely in the same manner as above, replacing 1](p) by 1](p) 0 g-l, since we
have the left (inverted) action g-l of 9 in (s2). We need to assume that g-l
for 9 E ..1 has an associated etale isogeny a g : (A, A, 1](p) 0 g-l) -t (Yg , Ag , 1]~p»)
for all p-ordinary test objects A = (A, A, 1](p»). Here we say A p-ordinary if
A is ordinary at p. Also we mean by an "etale isogeny" a g : A -t Yg an
isogeny given by /3 0 a- 1 for etale finite morphisms A' ~ A and A' ~ Yg .
Let ..1 o(pool)'l) = nr
..10 (prl)'l). If 9 E ..10 (pool)'l) , then a g is an etale isogeny.
Thus we assume that ..1- 1 is contained in
(4.39)
where ip : /Lp 0 D- 1 y AlP] is a p-ordinary level structure, and S runs over all
W-schemes. Strictly speaking, the isogeny a g can be defined over a faithfully
flat covering S' / S, which could be inseparable (see Section 4.4.2 for separa-
bility of a morphism of schemes).
If B = W, we require ..1p = ..1o(pOO)p. (4.40)
If B = W, we may extend ip to a level pOO-structure 1]~rd : /Lpoo 0 D- 1 Y Alsf

for the base change AI S' of AI S to an etale covering S' / S. Normalize the first
coordinate a of (b) E a* EB b = L (a E a* and b E b) so that it is associated
with 1]~rd. Then the above choice (4.40) of ..1 satisfies the requirement (4.39)
for A, because gp E ..1p by definition preserves the a coordinate and hence the
connected subgroup i p (/Lp0D- 1 ) inside AlP]. The normalization of coordinates
is equivalent to choosing a point carrying (A, 1]~rd, 1](p») on the component of
the Hilbert modular Igusa tower containing the cusp (a, b) (see the proof of
Theorem 4.21). Hereafter we write 1] for 1]~rd X 1](p) in order to treat the case
of Sh(p) and Ig uniformly. When we are dealing with Sh(p) , we just ignore
1]~rd and the ordinarity of A.
We may further impose (4.38). If we do impose (4.38), the isogeny
a g : (A, X, 'i'j 0 g-l) -t (Yg , Xg , 'i'jg) with a g 0 'i'j 0 g-l = 'i'jg
uniquely determines g-lw = a;*w for wE HO(A, DAIs) (which is a section of

I£.YK over ShYK). Since diJk is a section of I£.} over Sh K at (A, X, 'i'j), we have
a natural left action w 0k M g-l . w 0k := (g-l w )0k = (a;*w)0k of 9 E ..1
making the following diagram commutative:
1 1
where the bottom arrow is given by [x,h] H [X,hg-l].
If B is a Q-algebra, we can ignore the condition related to v = p (in
particular ip) in (4.39), because ordinarity and etaleness in (4.39) are always
satisfied in this case. Taking now B = Q and .::1 = G(A)+, by the same process
as above, we can define the action of G(A)+ on i,<d.k over the Shimura variety
Sh/ Q . If B is a W-algebra, similarly, we have a well-defined left action of
G (A (poo)) on differentials over S h;% (and I 9/w ).
The problem we now encounter is that the action of .::1 we have so far
defined may not factor through G(A)+ /Z(Q). In other words, the descent to
Sh K we desired to have is for an open compact subgroup K of G(7Lp x A(p))+,
but so far, we described it for K inside G(7L p) x G 1 (A (poo)) x G(JR)+. To make
this wider descent allowing all K(p) in G(A(p))+, we want to have the action
factoring through K/(K n Z(Q)) in order to make the quotient of the sheaf
over Sh by K. We have an isogeny E : (A,C 2A,1],EW) -7 (A,A,1] 0 E,W) for
E E Ox; so, E . W = EW. Thus the problem is to somehow modify the unit
action on differentials w@k H EkW@k for E E K n Z(Q) C Ox. This problem
can be solved if the base ring is complete; so, we assume that B = e or
W = ~nW/pnw. We write v for the place p or ()() (of Q), and use the
symbol B to indicate the base ring depending on our choice of v; so, B = e if
v = ()() and B = W if v = p.
To state the solution (found first in [H91] Section 3 and later developed and
brought into the present form in Fujiwara [Fl]), recall the diagonal torus T c
of G; so, Tc ~ T2. We choose a character wE 7L[I] = X(T) and a continuous
character C : Tc(i) -7 W X such that c+(E)-lE k - 2w = 1 for all units E, where
C+ : Z(.~) -7 W X (for the center Z of G) is given by c+(z) = cl(Z)c2(Z) writing
c(diag[a,d]) = cl(a)c2(d). Define C : T(Z) -7 W X by c(z) = C2 1 (Z)cl(Z).
We assume that c+ extends to a Hecke character c+ : Z(A)/Z(Q) -7 ex with
c+(x oo ) = x;:-k. Here W is equipped with the discrete topology; so, cJ is
of finite order, although c+ extended to Z(A) could be of infinite order. In
particular, C factors through (0 /91)X for an ideal 91.
We fix a prime element Wq for each prime ideal q. We suppose that wp = p
for pip (because pis unramified in F/Q). Sending x E F qX to (u, w~) E O~ xw~
if x = uw~, we have the identification Fqx = O~ x w~. Hence T(A(oo)) =
T(Z) x (EBq w~). Recall that v = p or ()() is the fixed place of Q. We now
extend our Neben character c = (cl,c2,c+) to T(A(oo)) as follows.
(exO) For each wE 7L[I], we extend the character xp H x"; oJT(7Lp) to T(Qp)
trivially on EBp Ip w~. Thus p-2! = 1.
(exl) We extend C2 to T(A/oo)) trivially on EBq w~ and then extend Cl to

T(A(oo)) by clc2(X) = c+(x(oo)). We put c;;(a) = c21(a)cl(a) for
a E T(A(oo)). Thus identifying T2 ~ Tc by (a, d) f-+ (0 ~), we get
an extension of c to Tc(A).
(ex2) We extend the character c of Tc(Z) n So((v) n 91) to a character of the
semi-group LlO (( v) n 91) by
c~(6) = det(6);;I c2 (det(6))c;;(a(v)n'l1)
for 6 = (~ ~) E LlO ( (v) n 91) U LlO ( (v) n 91) ~. Here (v) n 91 = (p) n 91 if
v = p and (v) n 91 = 91 if v = 00.
(ex3) Since the character Z(A(oo))/Z(Q) ::1 z f-+ Iz(oo) 1.z2z;;-21 c+(z) and c~
on So((v) n 91) coincide on So((v) n 91) n Z(A(oo)), we may extend
c~lso((v)n'l1) to cs = cs: Z(A)So((v)n91) = Z(A)Sl((V)n91) -+ B X by
cs(zs) = Iz(oo)I.z2z;;-2Ic+(z)c~(s) for z E Z(A) and s E Sl((V) n 91).
If we need to emphasize the dependence on v, we add the superscript v, as in
cs for Cs. Note here that cs depends on v but the central character c+ and
cJ (j = 1,2) do not. Also cl(wq) could be nontrivial, but c2(wq) = l.
By computation, we have
cs(z) = c+(z)lz(oo)I.z 2z;;-21 for z E Z(A),
and c~ (( 0 ~)(v)n'l1) = a;;;I d;;;I cl(a(v)n'l1)c2(d(v)n'l1)' (4.41)
We define new actions w®k f-+ gw,c,vw®k of 9 E Z(A)So(91) and w®k f-+ g*w®k
in (s2) of 9 E Llo(91) (or 9 E Llo(91)~) by
gw,c,vw®k = cS(g)~l det(g);;-W(g. w®k)

and g*(w®k) = c~(g) det(g):;'(g~l . w®k). (4.42)
The first action is a left action, and the second is considered to be a right
pullback action associated with Sh ::1 x f-+ xg~l E Sh for 9 E Llo(91). These
two actions coincide on So(91) n Z(Q) = Z(Z) and commute each other, and
hence the action of [KgK'] with respect to the second action is well-defined
on the cohomology group of the coherent sheaf on Sh K obtained from f!d.7Sh
by the descent via the first action. For the second action to be valid, one
has to assume that a g is an etale isogeny (which always holds if v = (0).
e
Then, writing ~(voo) for ~(oo) or poo ) accordingly as v = 00 or p, we have
~w,c,vw®k = 1~(voo)I.z2~;;-2Ic+(~(CXJ))~1~~2w~kw®k = w®k (for ~ E O(v)+) by
the product formula 1~(vCXJ)ll~;I = 1, and we are done. Here O(v)+ = 00,)+ if
v = p and O(v)+ = F"; = 00. The action of K(p) is well-defined over W,
if v
because det(g)~ = 1 if gv = l. We have K(p) / Kl ~. K/(Kl X Kp), and the
action of K(p) / Kl is independent of v. Using the action w®k f-+ gw,c,vw®k, as
4.2 Hilbert Modular Shimura Varietie:; 171
long as "l..txKp/B is well-defined for K c Z(A)So(IJl), we can make a descent

from "l..t xKp/W to "l..~,o,K/W' and we have for any W-algebra B,
where the HO(K(p),?) is with respect to the action wOk r l gw,0,vw0k. We

write this new sheaf as "l..~ 0 K on Sh K ·
There are some other c~s~s where the above definition yields "l..~,o,KiB over
Sh K . If v = 00 and B = C, the only requirement is to have smooth Sh K . If
v = p and B is of characteristic p, "l..t xKp/B over Ig K1 xKp is well-defined, if
Ig/IgKIXKp is etale and Kp ::J U(Zp). Here we recall the unipotent radical U
of the upper triangular Borel subgroup of G 1 . Thus if K ::J U(Zp) and I g/ I gK
is etale ({o} Ig K is smooth), "l..~,o,K/B is defined over IgK .
Abusing the language slightly, we call this Hecke character 10+ the centml
chamcter of "l..~ 0 K' because, as we show later in (4.55), modular forms f
on ShK(C) givi~~ rise to global sections of "l..~ 0 K satisfy fl(z) = E+(z)f
for the normalized diamond opemtor (z) = IzlA"2[KzKJ with z E Z(A(P<Xl»)
(the actual central character of f is I . 1A"2 E +). We define the v-type of the
central character of the sheaf "l..~,o,v,K to be the character of Z(Q) given by
~ rl ~:'(k)~; so, we find the v-type to be 2w - k. We now introduce the
"new" double-digit weight K, = (K,I' K,2) E '1'.[1]2 = Hom(Tc, Gm) of "l..~ 0 v K
by K,2 - K,1 + I = k and K,1 + K,2 = k - 2w + I. Hereafter we denot~ this
sheaf "l..~,v,K by "l..~,K and its cuspidal subsheaf by "l..1<,o,K' As we have seen,
these sheaves ~ and "l..1<,0 satisfy the three axioms (sl-3). This change of our
formulation of weights comes from a change of identification T2 ~ Tc: (k, w)
(resp. (K,I' K,2)) corresponds to T2 :3 (a, d) r l d- 1 (~ a~l) (resp. T2 :3 (a, d) r l
(g ~)), identifying T with Reso/zGm (see (4.70) for the description of the
automorphic factor of G giving rise to the sheaf "l..~).
Summarizing all these, we get
Theorem 4.24 Let the notation be as above. We have
(1) If the base ring B is C and Sh K is smooth, then "l..~ K and its cuspidal
subsheaf "l..1<,o,K are well-defined over C; ,
(2) If K is maximal at p and Sh~) is smooth, then "l..~,K and its cuspidal
subsheaf "l..1< 0 K are well-defined over W;
(3) If K contai'n~ U(Zp) and Ig K/ B is smooth for the base ring B of chamc-
teristic p or B = W, then "l..~,K and its cuspidal subsheaf "l..1<,o,K are well-
defined over I gK/B, where Ig K/ w is the formal scheme \!!!! m I gK/W='
The results of (1) and (2) in Theorem 4.24 give us a natural integral
structure over W of the p-adic sheaf "l..:/w and the Archimedean sheaf "l..:/c
over Sh~) (as long as they are well-defined). In other words, we have
(4.43)
at the same time. When E restricted to T(Z) is the identity character 1 and
E+ is a power of the modulus character I . lA, we drop E from the notation,
because then K determines E. As we observed already, when k = mI (m E Z),
we had a well-defined ~:;"{,, and we can easily verify that ~/0 = ~:/w for
K = (0, (m - 1)1) and E2 = 1 and E+(X) = Ixli- m . In some of the earlier
works of the author (e.g., [SGL] and [MFG]) , Ll(sn)" and E(X L ) are used in
place of Llo (sn) , and also the roles of K1 and K2 are interchanged (particularly
in the Hodge type of the motive associated to cusp forms of weight K below).
In other words, our formulation is cohomological (and hence Hecke operators
act from the right, because cohomology functors are contravariant), while the
formulation in [SGL] is homological; so, the Hecke operators act from the left.
The two formulations are related by [KgK'] B [(KgK')"l, and the role of K1
and K2 is interchanged in the homological formulation (thUS (m, n) = (K2' Kd
under the notation in [SGL]).
This double-digit weight K has a geometric meaning that the motive M f
associated with a Hilbert modular Hecke eigenform f E HO(Sh K , ~K'E:,K/d
constructed by Blasius and Rogawski ([BlR] for k > 2, [H81] and [Bl] for
k = 2) has Hodge weight (K2,0-,K1,0-) and (K1,0-,K2,0-) at the Archimedean
place (J of F, and by writing ~al for the €-adic Galois character associated
with the Hecke character E+, ~alN£ with the €-adic cyclotomic character N£
gives the determinant of the Galois representation of f (see [BlRl, [Tal, and
[MFG] 5.6.1 for a description of Galois representations and Section 4.2.7 in
the text for ~al).
Our definition of ES in (ex3) looks technical; in particular, the extra factor
Iz(voo)liz;I in (4.41) appears to be strange, but it is placed in order to kill
the factor I det(z)IA in (4.37) (and (4.44)) for the central element z E Z(A).
This is why the central character is given by E+.
When v = p, by our definition of w : T(Qp) -+ W X , g* is well-defined
for all g in the semi-group in (4.39). Still the multiplication by I det(g)IA in
the definition of [KgK'] may not be well-defined in characteristic p (because
Ip(oo) IA is not p-integral). We justify this later by means of the q-expansion
principle.
We now relate the new automorphic definition of Hecke operators to the
geometric one already given in Section 4.1.10. Let K1 = K n (G 1(Qv) x
G(A(v))). Decompose K1g. K,l = Uh K 1h for g E Llo((v) n sn), which yields
KgK' = Uh K . h. Then in the definition (4.37) of the operator [KgK'l,
the factor 'IrK' /(KgnK') ° [g] in down-to-earth terms is to average the image
h*(fw 0k ) over the coset representative h. Strictly speaking, this statement is
only valid when the degree of the isogeny cx g is invertible over the base scheme
S of the test object; otherwise, we need to stick to 'IrK' /(KgnK') because
ShKgnK-/ShK' (or IgKg nK , /Ig K') could be inseparable. If K· h = K· h', we
have (T/oh- 1 )oK = (T/Oh,-l)oK as K-orbits. Thus we have a well-defined value
f(A,"X, 'r/ 0 h- 1, w) for modular forms f of level K, because (A, "X, 'r/ 0 h- 1, w)
is a level K test object if so is (A, "X, fj, w). We suppose that 9 E <10(1).
We recall our construction briefly: for a test object (A,A,'r/)/s, the subgroup
C = 'r/(h- 1LIL) for L = 'r/-1(T(A)) is a subgroup of A defined over S. Let
Yh = AIC, and write ah : A --+ Yh for the projection. We thus have (Yh, Ah, 'r/h)
with 'r/h = ah 0 'r/ 0 h- 1 and Ah induced by ah from A. Write x E Sh(S) for
the point represented by (A, "X, 'r/). Then xh- 1 is represented by (Yh , Ah, 'r/h).
Pick Wx E HO(A, [lA/S). Then Wxh-1 = h- 1 . Wx for the corresponding section
over xh- 1 is given by ah*w X' For a global section s E HO(ShK,!,l) given by
s(x) = f(A,"X, fj, wx)w~k of ~l with a modular form f, we have
fl [KgK']v(A, "X, fj, wx)w~k = Idet(g)IA L h*(f(A, "X, fj 0 h- 1, WXh-l)W~k)

h
= det(gv)W I det(g)IA L c~ (h)f(Yh , "Xh , fjh, ah*wx)w~k. (4.44)

h
In the above formula, the summation symbol Lhc~(h)f(Yh) actually rep-

resents TrKI/(KgnKI)(c~(g)f(Yg)) if inseparability is involved (when a g has
noninvertible degree over the base of the test object) as we already remarked.
The operator [KgK']v really depends on det(gv) (because of the involvement
of (det(gv))Wc~(h)).
For an element Z in the center with Zoo = 1, we define the diamond operator
(z)v = IzlA:2[KzK'] by
fl(z)v(A,"X, 'ii, wx)w~k = I det(z) IAcS(Z) (f(A, "X, fj 0 z-l, WXZ-l )w~k)
= c~(z)z;2I f(Yz , "x z , fjz, (a;)-lwx)w~k. (4.45)
We have (z)p = (z)oo if zp = 1. It is customary to write (3) for (z)v if

zO n F = 3 and ZplJt = 1 (in this case, the operator is independent of v).
If we take the shrinking semi-group <1o(SJt)' in place of our choice <1 o(SJt)
here, the Hecke operator is induced by the right action of g' : ShK --+ Sh Kg"
and we need to replace the covering (A, "X, 'r/, WXh-1) ~ (Yh , "X h , 'r/h, (a h)-lwx )
by another one (Y~,,"X~','r/~,,(a~,)*wx) Q~') (A,"X,'r/,w x ) for the operator
[(KgK')'] in (4.44) (or simply to make the following variable change, 9 H g-"
as was done in [H88], page 307). We have made our choice of the cohomolog-
ical action, because it is directly related to the classical Hecke operator T( q)
already defined in Section 4.1.10.
We assume that Si(SJt) eKe So (SJt). If we consider a test object
(A, "X, fj) sitting over a point of S'K,oo = ShK[~], we have the connected com-
ponent Co(p) of AlP], and all finite subgroups C of A with C n Co(p) =
C XA Co(p) = {O} (scheme-theoretically) are etale. Thus if the test object
is p-ordinary and sitting over TK,m,a. for a > 0, we may normalize L to
be given by 0* EB ,-1 C V with 'r/(L) = T(A), and we may assume (by
the irreducibility of TK,m,Q proven in Theorem 4.21) that 7](1'* jO*) = Co(p)
at the characteristic 0 fiber of A. For any prime factor 1', taking the un i-
formizer wp = p of Op, we find KO(p)l (6 ~p) KO(p)l for Ko(pr) = Knih(pr)
is decomposed into Uu mod p KO(p)lhu for hu = (6';:p). Thus all etale
cyclic subgroups of order I' in A are given by 7] (h;/ LjL), and the def-
inition of the Hecke operator U(p) in (4.16) is identical to the operator
[Ko(p) (6 ~p) Ko(p)]oo = [Ko(p) (6 ~p) Ko(p)]p given by (4.44) (because
n
W;"" = 1 by (exO)). We write Tv(l, y) for [K (6 K]v, which really depends
on y E (A(oo))X (as we already remarked). We have Tp(l, y) = y;""Too(1, y).
When Sf((p) n SJ1) eKe So((p) n SJ1) and y(p'Jl) = 1, we write Uv(y) for
Tv (1, y). If YP = 1, these operators do not depend on v; so, we drop v from
our symbols if the choice v is clear from the context. Then we have
(4.46)
Note here that W';"" E W Xby the definition of W';"" in (exO) for any other
generator w~ of pOp, and hence
e = IIep with ep = n---+oo

lim Up(w~t! = lim Up(wp)n! = lim U(pt!·
n--+oo n-+oo
pip
When a prime qlSJ1, we have a specific subgroup Co(q) = 7]((cq)* jc*), and
recalling the uniformizer Wq of Oq, we find
Uv(w~) = [K (6 ~q) K]~ = [K (6~;) K]v for v = p and 00. (4.47)
If q is outside pSJ1, we have
T(l,w~) = [K (6~;) K]v for v =p and 00. (4.48)
The operator Tv(l, y) with YP = 1 does not involve y;"" and lyplA =
I det(diag[l, y])IA; so, it is integral over W. We record what we have proven.
Corollary 4.25 Let the notation and the assumption be as in Theorem 4.24
(2). Suppose that c = C21cl factors through (OjSJ1) X • Then we have W-
integral line bundles !;d.""E: C !;d.~ satisfying the three axioms (sl-3) with an
action of Llo((p) n SJ1) in (s2) specified in (4.42) for the coefficient rings W
and <C. These line bundles descend to Sh~jw for K C Ll o(SJ1) maximal at p
as long as Sh(p) jSh~) is etale. The Hecke operators Tp(l, y) = Too(l, y) with
yp = 1 are W -integral.
In Theorem 4.28, we describe when W-integrality for Tp(l, y) and Up(y) with
nontrivial YP holds.
4.2.7 Sheaves over the Shimura Variety of PGL(2)
We have constructed ~:,K for a closed subgroup K of ..:1 = ..:1 o((p) n 5)1). If
K(p) contains the center Z(A(pOO)) , the center acts on the sheaf ~ K by the
character c~, and we may regard ~ as a line bundle defined over th~ Shimura
variety Sh(PG, X) of the group PG = Reso/zPGL(2), because the central
action is specified. In this subsection, we construct a slightly larger family of
sheaves on Sh(PG, X) containing all ~: so far constructed.
We interpret the sheaf~: as a sheaf over Sh(PG,X). For a E Z(A(oo)),
writing 0 for F n aO, we find 7(A 129 0) = 7(A) @z 0 and 1]-l(7(A 129 0)) =
a1]-l (7(A)) for a test object (A,.x, 1]). Tensoring A with 0 c F :J 0, two
maps A 129 0 -+ A 129 F +-- A induce an isogeny a E Hom~(A 129 0, A). We
get (A, X, 1] 0 a) ~ (A 129 0, a*X, a- l 0 1]). This central action brings the
c-polarization to co 2 -polarization a*X as seen in Section 4.1.9.
Since the Shimura variety Sh(PG, X) is given by Sh(G, X)/Z(A(oo)), we
identify (A,X,1]oa,w) = (A@o,a*X,a-lo1],a*w) with (A,X,1]). If~ de-
scends to Sh(PG, X), the center Z(A(oo)) has to act trivially on ~~; so, we
find that k - 2w = 0 (hence k E 2Z[I]) and that 10+ is trivial. This means
K, = (I - w, w). We write the sheaf ~F-w,w) as ~ for w E Z[I]. We regard
~~ as defined over Sh(PG, X) = ShZ(A(oo»)(G, X).

Fix the Neben character 10 with trivial 10+, and consider ~~ as specified
above. We can generalize the above construction slightly allowing nontrivial
central characters X. As before, we write B for the base ring (either W or C).
The classification functor of A = (A, X, fj(p) , w),
is represented by a T-torsor M~)} over Sh(G, X)(p). Composing
we may regard M~)} as a g-torsor over the variety Sh~(A(OO») (G, X) for
g = (Z(Zp x A(poo))/Z(Z(p))) x T. Here Z(Zp x A(poo))/Z(Z(p)) is a
constant group. Take the vector bundle [2 associated with this torsor over
Sh~(A(OO»)(G,X). For each character (X,2w) : g -+ B X sending (z,x) to
X(z)x 2w (x E T and z E (Z(Zp x A(poo))/Z(Z(p)))), we can cut out a
line bundle ~~ [X]/B given by the sheaf of eigensections under character
g 3 (z,x) f--t X- l (z)x- 2w . In other words, we let the center Z(A(poo)) act
on ~:/Sh(P)(G,X) by X and descend the sheaf ~:/Sh(P)(G,X) to ~[X]/B over
Sh(p) (PG, X) by this new action.
Since Z(A(oo)) :J Z(Q) n Z(Zp x A(poo)) = Z(Z(p)) c Z(Zp x A(poo)), we
have
Z(Zp x A(poo)) Co-' Z(Zp x A(poo))Z(Q) Z(A)

Z(Z(p)) Z(rQ) Z(rQ)Z(lR)
Thus the characters of Z(ZpxA(P=)) are in bijection with continuous even char-
Z(z(p))
acters of the Galois group Gal(Fab / F) ~ Z(~~~~)+ by the Artin reciprocity

law of class field theory. A continuous Hecke character X : Z(A)/Z(rQ) =
F; / F X -+ ex is called arithmetic (in [MFG] 1.2.2) if we have Wi E X(Z) =
Z[I] such that X(x oo ) = x~w' whenever Xoo is totally positive. Suppose that
X is arithmetic and that the base algebra W is so large that Wi E X(Z) is
defined over W (i.e., x w' E W X if x E Z(W)). We have X(x(oo)) E rQx c rQ;
for x(oo) E Z(A(oo)). Then we define a p-adically continuous character
X : Z(A(oo))/Z(rQ)+ = FA(oo) x / F; -+ ij; by X(x) = ip(X(x(oo)))x;W' (see
[MFG] Theorem 1.1). By the Artin reciprocity map, we can regard X as a
Galois character Xgal . Thus if X is even, we have well-defined ~~ [X] := ~~ [X]
as an invertible sheaf over Sh(PG, X). The central character of ~~[X] is given
by Z 1--+ X(Z(P))-l z;:'.
Since the weight of ~~ is determined by its character e+
and K, we may define the weight of ~~[X] to be K = (~' + I - w, ~' + w).
Choosing ex : TcCZ) -+ W X with ex+ = X, we descend ~[X] to ~~cx,K over
SFJ;) if Wi E 2Z[I] for K with S}(SJl) eKe So(SJl) , where K = Z(A(oo))K
(so, K/Z(A(=)) is an open compact subgroup of PG(A(oo)) maximal at p).
When X = 'ljJ2, we can interpret the central twist ~ [X] of ~~ as an external
twist ~~ @ 'IjJ. We now introduce a twist of the sheaf ~~ over Sh(PG, X) by the
square root 'IjJ of X. The pullback of ~~ to Sh( G, X) has the natural action
gw,c of 9 E L1 in (s2) as we already specified. Using determinant character
det : G -+ T, we can twist the action gw,c on ~~ by w®k -+ gw,c,xw®k :=
'IjJ-l(det(g)(=))gw,cw®k. We denote this sheaf as ~~ @ 'IjJ. This sheaf ~~ @
'IjJ has the central character Xe+, and locally, u = (~~) E So(SJl) for the
conductor SJl of C acts on the sheaf by C(a<Jl)c2'IjJ(ad - bc). Defining e @ 'IjJ
by (el'IjJ(oo),c2'IjJ(oo)), we find (c@'IjJ)+ = e+X and
'" (0,
~c '61
0/'
'f/ = fid cI<+(w'/2,w'/2)
rv
®'1fJ ;
. 1ar 1y,
partlcu I< 0/'
~c @ 'f/ = ~cW[]
X 'f e+ IS
1
., 1.
. tnvla
(4.49)
Here is a criterion for the existence of 'IjJ = JX.

Lemma 4.26 If a finite-order continuous character X : F; / F X -+ ijx fac-
tors through F; / F X F;;' = F;(oo) / F X , then there exists a continuous character
JX: F; /F x -+ ex with JX2 = X.
Proof. Suppose that e
= T] E F X for ~ E FA(oo). Then, all primes of F
split in F[y07]; so, F = F[y07] by class field theory. Thus F;(oo)/F x F;;' is
2-torsion-free; so, we can find a continuous character JX as above. 0
A character of Gal(pab j P) is called totally even if it is trivial on any complex

conjugation (i.e., i-I 0 c 0 i for any embedding i : pab y C). If a Galois char-
acter X: Gal(pab j P) -+ B X is totally even, the corresponding idele character
factors through P:(=) j px, and we find a character 'IjJ : Gal(pab j P) -+ B X
with X = 'ljJ2.
4.2.8 Heeke Algebra of Finite Level

Let R be a W-algebra. Fix a Neben character E = (E1' E2, E+) with values in
R X as in Theorem 4.24 and a dominant regular weight K = (K1' K2) in X(Tc).
The weight is dominant regular if K2 > K1 and is just dominant if K2 ~ K1.
We define, for SJ1 prime to p,
G,JSJ1, E; R) = HO(S/J;;C'Jl.)/ R' iJd.~/ R) if K = (2: . 1)2,
and SI«SJ1,E;R) = HO(S/J;;C'Jl.)/R,iJd.I<,c/R), (4.50)
where K = Z(ACoo»)K (so, 8 0 (SJ1) = Z(A Coo»)So(SJ1)). The group KjZ(ACoo»)

is an open compact subgroup of PG(ACoo»). We then define the Hecke alge-
bra hl«SJ1, E; R) with central character E+ and Neben character E to be the
R-subalgebra of End R (SI«SJ1,E;R)) generated by T oo (1,w q) for all primes q
outside SJ1 and Uoo(y) (with all y E p~ n O'Jl.) for all primes qlSJ1.
For more general K maximal at p, we define
GI«K,E;R) = HO(ShK/R,iJd.~/R) if K E (2:. I?,

and SI«K, E; R) = HO(ShK/R,iJd.I<,c/R). (4.51)
Since K contains all the information of the center, the operator (z) (z E
Z(ACpoo»)) acts on GI«K, E; R) through the scalar multiplication by E+(Z)
(see (4.55)).
For K with Sf(SJ1) eKe SO(s:n:) , we define the Hecke algebra hl«K,E;R)
with coefficients in R by the R-subalgebra of End R (SI< (K, E; R)) in the same
manner as above.
In the following subsection, we show that the operators Too(l, w q ) and
Uoo(wq) are W-integral for almost all cases, at least if K1 + K2 ~ 1 and K is
dominant; see (hl-6) for more details.
More generally, for a subgroup K with (Sl(pT) n Sf(s:n:)) eKe SO(pTSJ1)
with r > 0, we define for a profinite W-algebra R,
GI«K, E; R) = HO(I9K/R' iJd.:/R) if K = (2:.1)2

and SI«K,E;R) = H°(I9K/R,iJd.I<,c/R). (4.52)
Here 19K/ R is regarded as the quotient of the Igusa tower 1 9 K / R by the action
of Z(2:p x ACp»). Since Ig K is an open formal scheme over W, these spaces
may not be of finite type over R. However, its ordinary part is an R-module
of finite type (Theorem 4.11).
Proposition 4.27 Suppose that K, is a dominant weight of Tc. If K is max-

imal at p, S,,(K, 10; R) and G,,(K, 10; R) are R-modules of finite type.
Proof. Since K, is dominant, £ = ~:j R for K, E (Z . 1)2 or £ = ~""E/ R for

general K, extends canonically to a line bundle over the smooth projective
toroidal compactification Sh'Km of Sh~). By the Koecher principle (see (4.10)),
we have HO(Sh~~ R' £) = HO(Sh R ( R' L) for £ as above if K, E (Z· Ij2. Taking
K,' = (0, (m -1)1) for a sufficiently large m, ~::;; for 10' = (I' lAm, 1, 1.1:- 2)
has a global section s over Sh'Km such that the multiplication by s gives an
injection of S,,(K, 10; R) into S", (K, 1010'; R), because ~';(1 is very ample for m
large. This shows that these spaces are R-modules of finite type (e.g., [ALG]
IIL8.8). 0
We define hK.(K, 10; R) and h,.;JpT>n, 10; R) by the R-profinite subalgebra of

EndR(S",(K,c;R)) topologically generated by Toc,(I,wq) for all primes q out-
side >n and UCXJ(wq) for all primes ql>n.
4.2.9 Effect on q-Expansion
If p is unramified in F IrQ, by Theorem 4.21, 9J1( c, r) /w and the Igusa tower

TCXJjw= over it are geometrically irreducible; so, the q-expansion f(q) =
f(Tatea,b(q)) determines the section f of ~:/R for any W-algebra R (see
Corollary 4.23). This fact we refer to the q-expansion principle over W. Here,
for a given open compact subgroup K ~ Sf (>n) (giving rise to the level struc-
ture of type r), W is assumed to be a sufficiently large discrete valuation
ring inside ij giving the p-adic place induced by ip over which the generically
irreducible component of Sh~jw is fiber-by-fiber geometrically irreducible.
We recall W = ~nWlpnw.
We compute the q-expansion of fit for Hecke operators t in order to show
that the Hecke operators preserve W-integrality. When we do this computa-
tion in the classical setting of ~k over 9J1(c, r), we assume f E Sdc, r; R).
When we deal with adelic Hecke operators Tv(I, y) and Uv(y), we assume that
f E SK.(K,c; R) for K ~ St(>n); so, in particular, f has the central character
10+. We prove
Theorem 4.28 Let the notation be as above, and assume that p is unramified
in FlrQ. Let>n be an O-ideal prime to p and R be a W-algebra.
First let K be a subgroup with Sf(pT>n) eKe So(pT>n) for r > O. Then
the operators [KgK]p for g E Llo(>n) with gp = I, (z)p for z E Z(A.(PCXJ)) ,
Tp(I, y) and Up(Y) (y E F; n 8) are well-defined on G",(K, 10; R) and
SK.(K, 10; R) if one of the following conditions is satisfied.
(hI) R = BlpmB for a W-algebra B, and the Hecke operators are well-defined
for GK.(K, 10; B) (if K, E (Z· Ij2), and S,,(K, 10; B) (for all K,);
(h2) R = W or Wm = Wjpmw and (Sl(pT) n Si(>n)) eKe SO(pT>n);
(h3) R is p-torsion-free.
Suppose that SUI)1) e K e So(I)1). Then the operators (z)CXJ for z E
Z(A(CXJ)), [KgK]= for g E Ll o(l)1) with gp = I, and TCXJ (I,p) are R-integral
under one of the following conditions.
(h4) We have R = H/pmH for a W-algebra H, the Hecke operators are well-
defined for G",(K,c;H) and S",(K,c;H), and Lc!~ is very ample on the
minimal compactification Sh*x:/13 of Sh K / 13 ;
(h5) R = W or Wm = W /pmw, ""1 + ""2 2' 0, and st(l)1) eKe So(I)1);
(h6) R is a Q-algebra.
Under (h5), if we further assume ""1 + ""2 2' I, we have TCXJ(I, w-p) == Up(w- p )
mod p in h", (K n So ((p)), c; W), where Up (w-p) is defined on the bigger space
G",(K n So(p), c; W) than G",(K, c; W) on which Too(l, w-p) is defined.
Proof. We first treat the case where (h4) is valid. Since Lc!~ is very ample, it
is generated by global sections, and G",(K, c; H) 013 R = G",(K, c; R). Since
G",(K, c; H) is stable under the Hecke operators, G",(K, c; R) is also kept by the
Hecke operators. Once we know that G", is stable under the Hecke operators,
S'" is also stable, because the cusps are permuted by the action of JL E Ll.
To see the assertion under (hI), we consider the normalization I g K of the
minimal compactification M'K of sh~lp)xG(7Lp) in Ig K · Let S'K be the formal
completion of M'K [i 1 along the modulo p special fiber for a lift E of the
Hasse invariant. Then we put Ig'K = Ig K XM K S'K. Then S'K and Ig'K are
affine schemes, and I g'K is etale over S'K. This is because the Igusa tower
Igkm over the smooth toroidal compactification Sh~,?p)XG(7Lp) is etale and the
infinitesimal structure of the projection I gkm ---+ I g'K around each cusp of I g'K
is isomorphic (infinitesimally) to that of the projection Sh~,?p)XG(7Lp) ---+ M'K.
Note that Lc!~ for "" E (Z· 1)2 and Lc!c: '" extend to Ig'K and H°(Ig'K, £) =
H°(Igkm , £) by the Koecher principle' for £ = Lc!~ and Lc!c:,,,,. Since Ig'K is
affine, we find FK(K, c; H) 013 R = F",(K, c; R) for F = G and S. Then we
conclude the assertion in the same way as in the case of (h4).
The operators may not be well-defined, because of the factor I det (g) IA in
(4.37), (4.44), and (4.45) (see also (4.16) and (4.17)). If R is a Q-algebra, this
does not cause any problem; so, the assertion under (h6) follows.
To show the assertion under (h2), (h3), or (h5), we apply the q-expansion
principle as described at the beginning of this subsection and Corollary 4.23
to verify that fl [K gK] and fl (z) are R-integral as long as f is R-integral. We
first compute the q-expansion of fl (3) for f E Sk( c, rl (1)1); R) and a fractional
O-ideal3 prime to p. Since we see from (ex3) applied to z E Z(A(PCXJ)), the
factor I det(z)IA = Izli in front of the summation symbol Lh is canceled out
by the factor Izl.,;:2 in (4.41), we only need to write explicitly Tatea,b(q) 03- 1
for 3 = zOnF in order to compute the q-expansion at the cusp (0, b) of fl(3).
Since Yz in (4.45) is given by A03- 1 = A/A[3], for an integral idele z with
3 = F n zO, we find
Tensoring G m with the exact sequence: 0 -t 0* -t (JO)* -t °/J -t 0, we get
o -t (/-LJ ~) TorI (0 h G m) -t G m 00* -t G m 0 (Jo)* -t O.

This shows
Tt ,o.._1_Gm00*/qJ-lbc,;Gm0(JO)* T ()
a en ,b()
q '<YJ - /-LJ - qJ Ib = ate'n,,-lb q,
0 0
which implies
(4.53)
For an ideal J prime to p generated by a totally positive element a with

a == 1 mod N, we can identify Tate.-ln,Jb(q) with Taten,b(q) by t H to. for the
variable t E G m 0 0* composed with variable change q H qo.-l. The variable
change q H qo.-l is introduced to neutralize the effect of the endomorphism
t H to. on the period q of Taten,b(q). When (0, b) = (0, c- l ), this induces an
equivalence up to isogenies,
- _ dt - _ -1 -1 dt
(Tate.-l,JC- 1 (q), A can , 'T]can' T) rv (Tateo,c-l (q), A can , 'T]can 0 a ,a T)'
where Acan is the canonical polarization on Tateo,c-l (q) induced by

(0- 1 EEl c- l ) /\ (0- 1 EEl c- l ) ~ c*
sending (x, y) /\ (x', y') to x'y - xy', and 'T]can is the projective limit of the
level T(N)-structure
(0- 1 0 Z/NZ) EEl ~~~l 3 (( 0 a, b) H (exp ( -;) 0 (, (liN mod qC-

1
))
and fican = 'T]can 0 8 0 (sn). Here we choose N so that sn n Z = (N). We write

(z)v (z E Z(A.)) for the center action as in (4.45) which is independent of v
if zp = 1. By the above computation applied to J = (a), writing a(voo) for
a(poo) or a(oo) accordingly as v = p or 00, we have, if idl< = idmI for an integer
m, ac-l((,fl(a(voo»)v) = a k N(a)-2a c l((,j) for a E F; outside v. Here the
factor a k in the right-hand side is the result of
dt
= f(Tateo,c-l (q), a- 2 A can , 'T]can 0 a-I, a-IT)
k -2 dt
=a f(Tateo,c-l (q), a A can , 'T]can, T)'
and the factor N(O'.)-2 comes from the shift by I det(O'.(voo))IA in (4.45) of the
action of (O'.(voo))v' For general r\" the action of O'.w,c,v is trivial,
and hence O'.};;,r::2w®k = c+(O'.(voo))-l w®k. We have from this (replacing a. in

the above formula by 0'.-1)
ael (~, fl(O'.(VOO))v) = c+(O'.(VOO))ac_1 (~, f) = O'.",,+K-2- I ac _1(~, f). (4.54)
For the sheaf ~ with general r\, and c, since we have the central character c+,
for more general Z E Z(A(oo)) with Zv = 1, we find from (4.45),
(4.55)
This is the reason why we call c+ the central character. Also the two identities
(4.54) and (4.55) for Z = O'.(v) coincide.
Although we do not need this, we now insert the computation of the
effect of U(p) for I,!dmI on the q-expansion, using (4.17), because I,!dmI ex-
tends naively to Sh K giving rise to ~ for r\, = (0, (m - 1)1), C2 = 1, and
c+(x) = Ixll-m. In this special case, (4.16) yields the correct result. By def-
inition, Tateo,c-,(q)[p] is isomorphic to /-lp x q(pC)-I/C- ' . The nonconnected
subgroups over Ware given by Cc, = {(mQmlm E 0/p} for ( E ILp and
a generator Q of qpc- ' /c- l . Thus we see Tateo,c-' (q) /W[/LP][[q~ll~E,-1 ICc, =
Tate(pc)-I,o((q)/W[/Lp][[q<lleEP_I,_I' Writing ap-Ic-l (~, f) for the q-expansion
coefficient of f E GmI(rl (cp, l)1); R) at the cusp (p-1 c-1, 0), we find the
q-expansion coefficient of fIU(p) at the cusp (c-1, 0) is given by
ael(~,JIU(p)) = N(p)-l (2: (~) ap-lc-l(~,f)

C,E/Lp
for ~ E p- 1 C- 1.
By the orthogonality relation of characters ( I--t (~ of /-lp, we find that only the
coefficient of q~ for ~ E c- 1 (not for ~ E p-1 C-1 - C 1 ) survives and is equal
to ap-Iel (~, f). Thus, the division by N(p) is canceled if R is p-torsion-
free, and the operator U(p) is well-defined on GmI(rl (cp, l)1); R) even for
R with nontrivial p-torsion if GmI (rl(cp,l)1);R) = GmI (rl(cp,l)1);W) ®w
R. In particular, if m » 0 so that I,!dmI is very ample, U (p) is well-defined
over R. Since the ordinary locus S;,. of the minimal compactification M;" =
M* (c, rl (l)1)) is affine, by Koecher's principle, we have
HO(S'/x,/R,l,!dmI) = HO(S~/R,l,!dmI)
= HO(S~/w,l,!dmI) ®w R = HO(S'/x,/w,l,!dmI) ®w R.
Thus the Heeke operator U (p) is well-defined on the space of p-adic modular
forms with coefficients in any p-adically complete W-algebras R.
Suppose that q is a prime factor of (p) n 1)1. Let w be a generator of q

with w E O~)+, We compute U(q) for I.Jd. mI using (4.16). Since we identify
each e-polarization with a we-polarization by ), >-+ w- 1 )" we are identifying
the q-expansion at the cusp (0,e- 1 ) with that at (0,w- 1,-I) by inclusion
R[[qEllEEc-1 c R[[qw-1<:llEEC-1. Since the identification), >-+ w- 1 ), is only
canonical up to O-units, the operator U(q) thus defined as an endomorphism
of GmI(e,r;R) (for r = r O(I)1),rl(I)1), and rl(I)1)) depends on the choice
of w; so, we wrote it as U(w). Then by the same computations as above,
we have, if w is a global generator of q, ac-1(~,fIU(w)) = ac-1(W~,f). A
slightly more generally, if a totally positive, w E F generates ph for an integer
h> 0, the operator U(w) = U(p)h is intrinsically defined on I.Jd.k for general k
(not necessarily equal to mI). The above computation can be performed for
general k if ph = (w) and yields the same result
(4.56)
In general, we have
(4.57)
Now we generalize the above formula to 1.Jd.~ using (4.44). Recall the prime
element Wq we have chosen in (exO) so that E2(wq) = w~' = 1. Then by
(4.41), we find that E~ (6 -d:q ) = 1. If w = Wq, then by (4.44), the definition
of Uv (wq) is identical to (4.16). Then the same computation for the sheaf 1.Jd.~
yields
For a general local generator w~ of qO q' we have
where w';,~'E2(w~) is always a p-adic unit by our way of extending E2 to

T(A(oo)) (see (exl) in 4.2.6) and the definition of d;1I:1 in (exO).
The formulas (4.56) and (4.57) are the special case of (4.58) for E = 1 and
"'1 = O. This finishes the proof for the assertion under (h2), because we have
now established the W-integrality of fIUp(wp) for all prime factors pip by
Corollary 4.23 if f is W - integral. Hence the assertion under (h2) for R = W m
follows from (hI). Since Dc is dense in the space of p-adic modular forms
(Theorems 4.9 and 4.10), a global Wm-integral section f of ~ is a reduction
modulo pm of a W-integral section in D cjw for a sufficiently large g. By the
stability of GII:(K, E; W Q9z Ql) under the Hecke operators, Dc is stable under
[KgKjp for g E Llo (prl)1) , and we have proven that the space of p-adic modular
forms V(r; W) in (4.20) is also stable under the Hecke operators.
Under (h3), R is W-fiat, and GII:(K,E;R) = G",(K,E; W) Q9w R (fiat base
change). Thus the assertion under (h3) follows from the assertion under (h2)
already proved.
Although we do not need this, we again insert the computation of the

effect of T('CD) := T(l,'CD) for f=d.mI on the q-expansion, using (4.17). For a
prime ideal q outside 91, we have one more (order q) etale cyclic subgroup /-lq
of Tateo,c-,(q) besides Cc, = {(mQm E Tateo,cl(q)lm E O} for a generator
Q of qq-lC-l/C- l . Consider the q-power map Tateo,c-,(q) ~ Tateo,qC-l(q)
with kernel /-lq. If q = 'CDO ('CD E F.n, this map is realized by t H tW.
Then wcan(q) = '¥
on Tateo,c-' (q) is sent to (7f~)-IWcan(q) = 'CD- 1Wcan (qW).
Since we identify e-polarization with 'CD-Ie-polarization by A H 'CD A, we are
identifying the q-expansion at the cusp (0, C 1) with that of (0, 'CDe- 1 ) by the
inclusion R[[qW~ll~Ec-l C R[[q~ll~Ecl; so, we find, if q is a prime ideal outside
91 generated by a totally positive 'CD E F,
(4.59)
where, as a convention, we have acl (~, f) = 0 if ~ rf. e- 1 . Here the 'CD mI in

front of the second term comes from
For a prime ideal q outside (v) n 91, we now perform a similar computation
for the sheaf ~ on which .do(91) acts. In this setting, Too (1, 'CDq) = Tp(l, 'CDq);
so, we may assume that v = 00. We implicitly suppose all order q subgroups
of test objects are etale (though after computing q-expansion, we will know in
many cases that the formula is valid even if it is not the case). The central el-
ement z E Z(A) acts by the character Z H Iz(00)1A: 2z;2Ic:;'(z(00)) (see (4.41)).
By the decomposition Kq (6 ~q) Kq = Kq (~q U (Uu n mod q Kq (6;q)),
the latter half has the same effect as c~ (6 ;q ) U ('CD q)
for c~ (6 ;q) =
C2('CDq) = 1; so, we need to compute the effect of the first coset Kq (~q n.
Write h for (~q n. Then h = 'CDqh' for h' = (6 w~l). By (4.41), c~(h') = 1.
The central action of 'CDq gives rise to the multiplication by N(q)2c+('CDq),
because f=d.;,x is the quotient of f=d.k by the action w®k H gw,E,vw®k. Thus
I det(h)IAN(q)2c+('CDq)c~(h') is integral if /';;1 + /';;2 2: O. The effect of h' can
be computed by the etale isogeny,
sending t mod qqc- ' to t mod qC- l . By the equivalence up to isogenies:
- dt - ,-I dt
(Tateo,qc-l (q), Acan , 'T/can, t) '" (Tateo,c' (q), Acan , 'T/can 0 h 't)'
we get, if q t 91 is a prime,
(4.60)
because Wq,v = 1 if q is prime to (v) n 91. The sheaves id k and id~ have the
same pullback map g* for 9 E Llo(91) only when k = mI and (El' E2' E=;O) =
(l'I Am , 1, 1·1~-2). In this special case, it is easy to see that the formulas (4.60)
and (4.59) match.
We assume (h5). Since we have [KgK]oo = [KgK]p if gp = 1, the assertion
for [KgK] with gp = 1 follows from the argument for (h2). Thus we only need
to take care of Too(l, wp) for pip. We see that N(p)E+(wp) is p-integral as
long as 1>:1 + 1>:2 2' 0 and that N (p )E+ (wp) == 0 mod p as long as 1>:1 + 1>:2 2'
I, because E+ is the finite part of an arithmetic Hecke character of infinity
type -1>:1 - 1>:2 + I. The operator Too (1, wp) is originally defined only over 1Ql.
However by the above formula, if 1>:1 +1>:2 2' 0, it sends the W-integral subspace
GK(K, E; W) into itselffor K with S£(91) eKe So(91). Since N(p)E+(wp) ==
o mod p if 1>:1 + 1>:2 2' I, Too(l,wp) == Up(wp) mod p by the q-expansion
principle (Corollary 4.23 (1)). Thus for K with S£(91) e K e So (91), if
GK(K, E; W) ®w R = GI«K, E; R) and 1>:1 + 1>:2 2' 0 (under unramifiedness of p
in F/IQl), the operator Too(l,wp) is actually well-defined on GI«K,E;R), and
Too(l, wp) is equal to Up(wp) if 1>:1 + 1>:2 2' I and pR = O. D
We record what we have seen for p-adic modular forms in the above proof.
Corollary 4.29 Let T be a group with rl (91) eTc Fa(91). For each p-adic
W -algebra R, the space V(T; R) of p-adic modular forms defined in (4.20)
is stable under the Heeke operators Tp(l, y) with YP = 1, (Y)p for yEO n
(Fipoo)y and Up(Y) for y E Fpx n G p.
By the universality of V (c, T; W) (which classifies (A, X, ¢;, ip) / R for p-adic
W-algebras R), we find V(c, T; R) = V(c, T; W)0wR; so, the assertion for R
follows from the same assertion for W.
4.2.10 Adelic q-Expansion
In this subsection, we introduce adelic q-expansion coefficients ap (y,1) for

a modular form f on ShK, whose Archimedean counterpart is the Fourier
coefficients of a modular form on G(A) with respect to the canonical Whittaker
function (whose Mellin transform is the T-factor of the standard L-function
of GL(2)). Let B be the base p-adic W-algebra. The coefficients ap(Y, 1) for
ideles yare useful in the following section to describe the duality between the
Hecke algebra hI< (91, E; B) and the space of modular forms SI< (91, E; B).
Choose a complete representative set {c} C (Fi poo )Y for the class group
Cl~ = FJ; /Ox F X F~+ as in Section 4.1.3, and write c = cO n F. Recall that
Shso('Jt) = UCEC1 + 9J1(c, To (91)) with
F
where c= (8?). A modular form f E SI«SO(91),E;B) can be considered as

a tuple (fC)CEC1"t, with fc(A,X,Tj,w)w@k E H O(9J1(c,To(91)),idl<,E/S) for k =
"'2 - "'1 + I. A p-adic modular form f is a tuple Ue)e indexed by C E elt of

sections fe E Vcusp( c, ro(SJ't); B) of the structure sheaf of the formal completion
of 8 = 9Jt( c, ro (SJ't)) [1; 1 (for a lift E of the Hasse invariant) along its modulo p
fiber 8 1 . For a given modular form f = Ue), it has a q-expansion at the cusp
(O,c- l ): fe = EO«:~Ee-l acl(~;fe)q€. We want to compute the action of the
matrix 9 = ( '(;;0) ~) for a totally positive unit E E a on the q-expansion of f.
More generally, we define fig for f E G,,(8(N), c; B) and 9 E ,1 by
(4.61)
where w 0k f-7 g*w 0k is as in (4.42). This is a right action. Taking v = p and

,1 = ,1(pOO)p x G(A(p)), f f-7 I det(g)IA I Ulg) gives rise to a right automorphic
representation of the group ,1(pOO); x G(A(p)) on liI!;NHO(8FJ;;N),w..~/B).
Even if rJ(p) misses its p-component, gp acts nontrivially on the differential w
(so nontrivial gp does make difference). Here N runs over all positive integers
prime to p. If B is a Q-algebra, taking v = 00 and ,1 = G(A(oo)), we can extend
the action f f-7 I det(g)IAUlg) to G(A(oo)) in exactly the same way, replacing
(8h(p), rJ(p)) by (8h, rJ) and taking the limit over all positive integers N. Note
that the choice of v = p and v = 00 makes the resulting representations
different at their p-component. Over C, for the choice v = 00, this is the
right regular automorphic representation of G(A(oo)) (with central character
c+) on the space of holomorphic Hilbert modular forms and is known to be
a discrete direct sum of irreducible admissible (automorphic) representations
with multiplicity at most 1 ([AAG]). The twist ®I det(g)IA has to be there to
undo the reverse twist IzlA2 appearing in the definition (ex3) of c:s(z). We
have done such a twist in order to formulate the effect of Hecke operators in
(4.58) and (4.60) in a traditional form. Without the twist, the right-hand side
of (4.58) and (4.60) for v = 00 will be divided by N(q). In any case, we also
obtained a rational structure over Q[c+l on the (right regular) automorphic
representation acting on holomorphic modular forms.
We compute the q-expansion of fig for 9 = ('(;;0) ~) for E E Ox. Take
f E G,,(8(SJ't), c; B). We consider the effect of t f-7 t' on the Tate AVRM
Tateo,e-l (q). This map is induced by multiplication by E on G m ® ()-1, which
induces multiplication by E on f1.<Jl (corresponding to the action i f-7 i 0 E on
level n (SJ't)-structure i). Since the level r l (SJ't)-structure i gives the identifi-
cation of (SJ't* EB c l ) I L with f1.<Jl C G m ® ()-1 (for L = ()-1 EB c- l ), the action
t f-7 t' on level r l (SJ't)-structures corresponds to the action of g-1 for the ma-
trix 9 = ('(;;0) ~). Indeed, the isogeny cx g : A -+ Yg for 9 E ,10 (N) is just the
quotient map A -+ Yg = AI Ker( cx g) (Ker( cx g) = rJ(g-1 L IL)); so, it does not
have any effect on L. The action t f-7 t' has an effect on T(Tateo,c-l (q)) which
is canceled out by the action of g-1 on T(Tateo,c-l (q)). Since cx g and 9 act
trivially on q<Jl-1c- 1Iqc- 1 , to describe the action of 9 on Tateo,c-l (q), we need
to make variable change q f-7 qE- 1, to compensate the effect on q<Jl-1c- 1Iqe- 1 •
The action of 9 is reflected by the q-expansion of f E 8,,(SJ't, c; B) as follows,
(felg(A,:\, T/, w))w@k = g~,E(fe(A,:\, T/ 0 g-l, w)w@k)

= fe(A, :\, T/ 0 g-l, W )(g~,Ew@k) = E~ (g(OO)) (det(g)v)W fe (A,:\, T/ 0 g-l, w)w@k
= E1 (E p 'l1)E- fe (A, :\, T/ 0 g-l, W )w@k,
K01
where the last identity follows from K:1 = I - w. Note that E 1 ( Ep 'l1) = E 1 ( E) by
(ex1) and that E+(E) = 1 for a unit E E O~. This shows
(4.62)
In Section 4.1.9, we only considered p-adic modular forms on PG with

a trivial central character. Since we have created the line bundle id; over
Sh(PG, X), moving around K: fixing the central character E+ : Z(A(oo)) -+
W X and E J : Tei) -+ W X , we have a similar theory for p-adic modular forms
with a fixed central character. We now describe the theory. Write r
(resp. r)
for r J (;;t) (resp. r J (;;t)) for j = 0, 1. Define Vcusp (r, E; B) to be the space of
p-adic modular forms (with coefficients in a p-adic W-algebra B) with central
character E+ and Neben character EJ . Here we assume that the conductor of
c is a factor of;;t so that ELl is well-defined over Llo(;;t). Then Vcusp(T, E; W)
is the space of global formal sections of the sheaf IdKO,E = ~ mid""E / s= over the
Igusa tower Too,oo/Soo, where Sm = Sh(p)(PG,X)/So(N) [iJ/w= for a lift
E of the Hasse invariant. Here Sh(p)(PG,X)/So(N) = UcEC [l2]9J1(c,r)/w=
for a complete representative set {c} for the two-torsion group Cl[2j =
Cli/(Cli)2, and physically as the T(Zp)-module (forgetting the Hecke oper-
ators) , we have
VCUSp(T,E;B) = EB Vcusp(c,r;B).
cEC112]
In other words, if we identify Vcusp (C,,2,r) with Vcusp(c,r) by f +--+ fl(,,) for
the operator (J) introduced in Section 4.1.9, we get Vcusp(r, 1; B) = Vcusp(T)
of Section 4.1.9. Instead we identify Vcusp (C3 2, r; B) with Vcusp (c, r; B) by
E+(J)-l (J). We still have Vcusp(T, E; B) = EB cECI 12] Vcusp(c, r; B) (under the
different identification). Thus, forgetting about the action of T(i(p)), as a
module over T(Zp), we have Vcusp(T, E; B) = Vcusp(T; B) physically. So, The-
orems 4.11 and 4.12 remain valid for Vcusp(r, E; W) with T = ro(;;t) and
T1(;;t). In particular, the W-dual of vc':s~(r, E; W) is W[[T(Zp)]]-projective
(of finite rank) at least if p ;::: 5.
Replacing T/(p) by the equivalent cyclic subgroup C of order ;;t to indicate
that this is a level ro(;;t)-structure and denoting by ip : Mp= (>9 Zl-l '-1 A[pOO]
for level r 1(pOO)-structures, we may think of f E Vcusp(ro (;;t), E; B) as a func-
torial rule (A,:\, C, ip) / R H f(A,:\, c, ip) E R satisfying the following three
properties for (variable) p-adic B-algebras Rand R'.
(G p1) (A,A,C,ip)/R ~ (A',X,C',i~)/R '*

f(A,A,C,i p) = f(A',X,C',i~);
(G p 2) If p : R -+ R' is a continuous W-algebra homomorphism, then
f((A,A,C,i p) XR,p R') = p(f(A,A,C,ip));

(Z) fl(3) = c+(3)f for all fractional ideals 3 prime to p91 ({::} flz = c+(z)f
for all z E Z(A(P')1°o»)).
By Koecher's principle, we do not need to have the condition correspond-
ing to (G p 3) in Section 3.2.9 if F i= Q; so, we have omitted it. We can let
a E T(Zp) act on f E Vcusp(Fo(91) , C; B) by f(A, A, C, ip) r-+ f(A, A, C, ip 0 a).
In the above definition, replacing the level ro(91)-structure C by a level
r l (91)-structure i')1 : j.J,')1 '-+ A [91] , we get the definition of the space of
p-adic modular forms Vcusp (h(91),c;B) on rl(91). Since 8h s1C ')1) is a dis-
joint union UeECI + m1( c, r l (91)), we have an embedding VCUSP(Fl (91), C; B) into
F
EB eECI + Vcusp(c, rl(91); B).
F
We return to the level ro(91). We write Vcusp (Fo

(91) , C; B)[Kl' cl] for the
subspace on which T(L) acts by the character WCl : z r-+ z;t;;lcl(Z) (note that
Kl = 1- w). We have a natural map 8t;;(91,c;B) '-+ Vcusp(Fo(91),c;B)[Kl,cl]
by defining f(A,A,C,i p) := f(A,>',r;Cp),i p,*('¥ Q9 1)) for the level 8 0 (91)-
structure r;Cp) induced by C and for the differential '¥ on G m Q9 ()-l. By
the q-expansion principle Corollary 4.23, this map is injective. We verify that
(4.62) still holds for f E Vcusp (Fo(91),c;B)[Kl,cl] by a similar computation.
Choose a finite idele c so that cO = c and cp = 1. For finite integral
ideles y and f E Vcusp (Fl (91), C; B) [Kl' cl], we introduce a continuous function
y r-+ ap(y; f) E B as follows: Choose ideal representatives c (prime to Np) so
that yO = ~c for ~ E (;:1, and write y = u~c for ideles u, c with cp = c[ = 1
for all prime factors [ dividing the conductor of C and cO n F = c. Define
(4.63)
The definition (4.63) does not depend on the choice of c. Indeed, taking an-
other c' and c' = F n c'O with y = u'(c', by Tateo,e-l(q) ~ Tateo,e,-l(q)
given via t r-+ t ro for w = ~-l( (which pulls back '¥ on Tateo,e'-l (q) to
('b
w,¥ on Tateo,c-l (q)), we have ac- 1 (~; fe I (g ~)) = ae,-l ((; fe' I ~)), be-
cause uu,-l = w- l which compensates the action of t r-+ t W on the level
structure. By definition, if f E Vcusp(Fl(91),c;B)[Kl,cd, then ap(uy;f) =
U:;;t;;l Cl (u)ap(y; f) for u E TeL). Thus a p is well-defined independently of the
choice of c, and for an integral ideal n prime to p and the conductor Itl of Cl,
choosing a finite idele n so that nO = n, ap(nCpctlly; f) is well-defined inde-
pendently of the choice of n. We write ap(yn; f) = ap(nCpctlly; f). We extend
the function a p outside integral ideles by defining the value to be O.
The above definition of ap(Y, f) extends without modification to p-adic
modular forms f in Vcusp (Fl (91), C; B) regarding Vcusp (Fl (91); B) as a subspace
of the direct sum EBcEClt Vcusp ( C, r 1(1)1); B), because we have a naturally de-
fined action of (t ?) for u E T(i) on such p-adic modular forms f given by
fl (t ?) (A, A, il)!, ip) = 101 (u)f(A, A, il)! 0 u;r/, ip 0 U;1). By Corollary 4.23, the
p-adic modular form is determined by the function a p on integral ideles. We
have now proved
Vcusp (ro (1)1) , 10; B)[K:1' cd =
{f E Vcusp (Fa (1)1) , 10; B) lap ( uy, 1) = 101 (U)U;'" ap(Y, 1) \fu E T(Z) } .
We compute the effect of Tp('Wq) := Tp(l, 'Wq) for a prime q t 1)1 on the
coefficients a p using (4.60). We have, if q pl)1, t
ap(Y, fITp('W q)) =acl (~; fcITp('Wq)1 (t?))
=aq-lc-l (~; fqcl (t?)) (4.64)
+ N(q)c+('Wq)aqC-l (~; fq-1cl (t?)) .
Note that y = u~c implies 'Wqy = u~c'Wq and 'W q 1y = u~C'Wq1. Then replacing
C (resp. c) by c'Wq (resp. C'Wq1) in the first (resp. the second) term of the
last equation of (4.64) and applying the definition (4.63) of ap('Wqy, 1) and
ap ('W q1 y, 1), we find, for all primes q outside pl)1,
The same formula is valid for Too ('Wq) := Too(l, 'Wq) including qlp for our cho-
sen prime element 'Wq (because c1('Wq) = 1); otherwise, the formula depends
on v = p, =. Similarly, we have
( 4.66)
We write Tv(y) (v = p, =) for the Hecke operator acting on G,,(I)1, 10; B)

corresponding to the double coset decomposition of
T(y) = {x E £1 0 (1)1)1 det(x)O = yO}
for yEO n Ft(oo)' Note that, if yO = yO n F is square-free (as an ideal), we

have Tv(y) = Tv(l, y), but otherwise, they are different.
Since we have a relation T('Wq)2 = T('W~) + N(q)('Wq) and U('Wq)n =
U('W~) (cf. [MFG] (3.27)), by iteration, we get from (4.66) and (4.65),
a p (1, fITp(y)) = ap(Y, 1) for all integral ideles y. (4.67)

4.2.11 Nearly Ordinary Hecke Algebra with Central Character
We now introduce the nearly ordinary Hecke algebra and prove the duality
between the algebra and the space of cusp forms.
We take limits e = lim n -+ oo up (p)n! and eo = lim n -+ oo Too (p)n! in the alge-
bra h",(K, 10; R) defined in Section 4.2.8 for a profinite W-algebra R, whenever
the Hecke operator Up(p) = Uoo(p) or Too(p) is in h",(K,c;R). We define for
K with Si(IJt) eKe So(IJt),
if pllJt,
if IJt prime to p,
(4.68)
if pllJt,
if IJt prime to p.
The algebra h~·ord(K, 10; R) is called the nearly ordinary Hecke algebra
on G of level K with Neben (or weight) character (,,-,c) and with coeffi-
cients in R. We split T(Zp) = r x ..:1(p) for the maximal p-profinite sub-
group r and the maximal prime-to-p torsion (finite) group ..:1(p). Since p
is unramified in F/Q, r is torsion-free. For each p-profinite W-algebra R,
any projective R[[T(Zp)]]-module of finite type is R[[r]]-free of finite rank.
We then define hn.ord(lJt, 10; W[[T(Zp)]]) for the W[[T(Zp)]]-subalgebra of
End(V~~~p(.fo(IJt), 10; R)) generated over W[[T(Zp)]] by the operators Tp(Y)
and (YIp for all integral ideles y.
Theorem 4.30 Let K be a subgroup ofG(A(oo)) with Sl(pr) nSi(lJt) eKe
So(prlJt) for IJt prime to p. By the pairing (t, f) = a p(l, fit) for modular forms
f and Hecke operators t, we have the following perfect duality.
(1) Suppose that K is maximal at p and that "- is a dominant weight of Tc
(<;=;. "-2 ::::: "-i). If one of the conditions (h4-6) in Theorem 4.28 is met
depending on the level, we have h",(K,c;R) ~ HomR(S",(K,c;R),R) and
S",(K, 10; R) ~ HomR(h",(K, 10; R), R) for an algebra R finite over W;
(2) If r > 0 and one of the conditions (hl-3) is met,
h~·ord(K, 10; R) ~ HomR(S~·ord(K, 10; R), R)
and s~·ord(K, 10; R) ~ HomR(h~·ord(K, c; R), R)
for a p-profinite algebra R finite over W;
(3) hn.ord(lJt, 10; W[[T(Zp)]]) ~ Homw(~':8~(.fO(IJt), c; W), W)
and ~':8~(.fO(IJt), 10; W) ~ Homw(hn.ord(lJt, 10; W[[T(Zp)]]), W);
(4) hn.ord(lJt, 10; W[[T(Zp)]]) ~ HomR(V~~~p*(.fo(IJt), 10; W), R)
and V~~~p*(.fo(IJt), 10; W) ~ HomR(hn.ord(lJt, 10; W[[T(Zp)]]), R)
for R = w[[r]].
Proof. We give a sketch of a proof. The Hecke algebras and the spaces of
modular forms involved in (1) and (2) are R-modules of finite rank. This
follows from Proposition 4.27 for (1) and Theorem 4.9 for (2).
Since the proof of the assertions (1) and (2) is the same and (4) follows
p=
from (3), we prove only (1) and (3). By the long exact sequence of ~""E:/w -'---+
~I<,E:/w --» ~""E:/w=, S",(K,E:;W) Q9 Wm (= HO(ShK/W'~""E:/w) Q9w W m)
injects into S",(K,E:;Wm) (= HO(ShK/W=,~""E:/w=)); so, the q-expansion
principle is valid for the finite module SI«K,E:;W) Q9 W m . For the pair-
ing ( , ) : (h",(K,E:; W) Q9 W m ) x (S",(K, E:; W) Q9w W m ) -+ W m , we have
(U(Y<n)T(l, y(<n)), f) = ap(Y; f) by (4.67), and hence, by the q-expansion prin-
ciple, f = 0 if (h",(K,E:; W) Q9w W m , f) = O. By the perfectness of the duality
of finite modules, we thus have a surjective morphism
of Hecke modules. Taking the projective limit relative to m, we get a surjective

homomorphism h",(K, E:; W) --» Homw(S",(K, E:; W), W). Since h",(K, E:; W)
acts faithfully on SI«K, E:; W), we conclude the injectivity. This implies
Since W is faithfully flat Over Wand W is a valuation ring, we get at the

same time S",(K, E:; W) ~ Homw(hl«K, E:; W), W). If R is flat Over W, we
find SI«K, E:; R) = SI«K, E:; W) Q9w R by flat base change (e.g., [ALG] 111.9.3).
This implies hI< (K, E:; R) = hI< (K, E:; W) Q9w R by the above two identities.
Replacing hl«K, E:; W) Q9w Wm and SI«K, E:; W) Q9w Wm by hl«K, E:; W m )
and SI«K, E:; W m ) in the above argument, we get the desired identity for R =
W m . Since any W-module of finite type is a finite direct sum of Wm-flat
modules for some m (including m = (0), the above argument using flat base
change works well to get the desired result for a W-module R of finite type.
This shows the assertion (1).
To prove the assertion (3), we simply write h for hn.ord(sn, E:; W[[T(Zp)]]).
We consider the pairing ( , ) : vg~~p (Fa (sn)) x h -+ K jW given by
(j, h) = ap(l, flh). Then again we have (U(Y<n)T(l, y(<n)), f) = ap(Y; f) by
(4.67) applied to p-adic modular forms and, hence, by the q-expansion prin-
ciple, (j, h) = 0 implies f = O. By the perfectness of the Pontryagin duality,
we thus have a surjective morphism: h -» vg~~p* (Fa (sn)) of Hecke modules.
Since h acts faithfully on vg~~p*(Fo(sn)), we conclude the injectivity. Since
we have vg~~p*(fo(sn)) ~ Homw(Vc':s~(Fo(sn), E:; W), W) (e.g., Section 3.3.1
particularly (3.4)), assertion (3) follows. Assertion (4) is a formal consequence
of (3) and the R-freeness of vg~~p*. 0
The above theorem combined with Theorem 4.12 implies

Corollary 4.31 Let sn
be an O-ideal with sn n Z = (N). Suppose that p ?: 5
is prime to Nd(F). Fix a K, E X(Tc) and a Neben characterE: = (E:l,E:2,E:+)
with C factoring through (Ojsn) x . Then we have for T = Reso/zGm,
(1) hn.ord(sn, E:; W[[T(Zp)]]) is a W[[T(Zp)]]-projective module of finite type;
(2) If K; - K~ ::::: 21 for K' E Z[Ij2 with K~ + K; = K1 + K2, we have a canonical

isomorphism of W -algebras
hn.ord(sn, s; W[[T(Zp)]]) ®W[[T(Zpl]]'''' W ~ h~/rd(sn, s; W)
sending Tp(Y) and Up(Y) in hnord(sn, s; W[[T(Zp)]]) to Tp(Y) and Up(Y)

in h", (sn, s; W). If we replace sn by psn in the right-hand side of the above
identity, it is valid under the milder condition K; - K~ : : : I.
This corollary gives us a finite fiat fibration Spec(h) --+ D = Spec(W[[T(Zp)]])
(h = hnord(so(sn), s; W[[T(Zp)]])) whose fiber at an algebraic regular weight
K is the (spectrum of the) nearly ordinary Hecke algebra of weight K with a
fixed central character s+. Here the p-adic rigid analytic space of D is just
a disjoint union of g-dimensional p-adic open unit ball (g = [F : CQl]); so, we
have found a nice family of algebras from modular forms.
4.2.12 p-Adic Universal Heeke Algebra
We generalize Corollary 4.31 allowing the deformation of the central character.

This means that we can expand the base D of the fibration Spec(h) --+ D
so that we can deform (or continuously vary) central characters of Hecke
algebras appearing as its fiber. Let Z = crj;(pOO) be the ray class group
modulo poooo, that is, ~rClj;(pr). We decompose Z = rz x L1z so that rz
is p-profinite and L1z has order prime to p. Note that any character X of rz
can be regarded as a character of Z, because Z / L1z is canonically isomorphic
to rz. If p > 2, any character X : rz --+ W X has a unique square root yiX.
Thus we have a canonical identity 1d",E ® yiX = 1d",E ®w[[rz]]'v'X W ~ 1d",E@X
for s ® X = (SlyiX, S2yiX, s+X) (see (4.49)). We define
hn.ord(sn, s; W[[T(Zp) x r z ]]) = hnord(sn, s; W[[T(Zp)]])0w W[[rz]],

( 4.69)
which is called the universal nearly p-ordinary Hecke algebra. In the class
of characters {s+xh we have the Teichmiiller lift So+ of s+ mod mw. By
construction, the above algebra only depends on the Teichmiiller lift So
of s mod mw of weight KO = (0,1); so, we write the above algebra as
hnord(sn, So; W[[T(Zp) x rz]]).
For an idele y, we have a well-defined ideal class y E Clj;(pOO)
FJ: /FX(O(pl)XF:;'+. We write [y] for the image of y under Z --+> rz. We
have the following consequence of Corollary 4.31 and the discussion in Sec-
tion 4.2.7.
Corollary 4.32 Suppose that p f 2Nd(F) and that p ::::: 5. Then we have
(1) The Hecke algebra hnord(sn, So; W[[T(Zp) x r z ]]) is a W[[T(Zp) x rzll-
projective module of finite type;
(2) When "'2 - "'1 :2: 2I for", E X(To) and cITG(Z) = COITG(Z) with c+ == co+
mod ffiW on Z(A(poo)), for every arithmetic Hecke character X of rz with
oo-type 2v, we have a canonical isomorphism
h n .ord (l)1, co; W[[T(Zp) x rz ]]) ®W[[T(Zp)xT'zll,I<<8Iv'x W

~ h~~(:,v) (1)1, c ® X; W),
which sends Tp(Y) ® [y] and Up(Y) ® [y] in h n . ord (l)1, C; W[[T(Zp) x r z ]]) to
Tp(Y) and Up(Y) in h~~(:,v)(I)1, c ® X; W). If we replace 1)1 by pl)1 in the
right-hand side of the above identity, it is valid under the milder condition
"'2 - "'1 :2: I.
4.3 Rank of p-Ordinary Cohomology Groups
We give a sketch of a proof of the constancy of the rank of nearly ordinary

cohomology groups when we vary the central character and the weight. The
idea is the same as in the proof of Theorem 4.9 after moving everything to
quaternionic automorphic forms by the Jacquet-Langlands and Shimizu cor-
respondence. It is a well-known result of Jacquet-Langlands and Shimizu that
if we choose the level appropriately, the space of holomorphic quaternionic au-
tomorphic forms can be embedded into the space of Hilbert modular forms
preserving the Hecke operator action. If the ramification of the quaternion
algebra is limited to Archimedean places, the embedding is surjective. Fur-
thermore, if the Archimedean ramification is maximal, the Shimura variety
has small dimension::; 1 so that we can make the argument more precise.
We first redefine the space of modular forms in 4.3.1 in order to state
the correspondence in Section 4.3.2. In Section 4.3.3, we interpret the corre-
spondence in terms of topological cohomology and make it canonical. Then
we shall give in Section 4.3.4 an exposition of the Eichler-Shimura isomor-
phism which connects the topological cohomology groups and the space of
cusp forms defined in Sections 4.3.1 and 4.3.2. At the end of this section, we
give a proof of the boundedness in Section 4.3.5. Contrary to (exO) , in this
section yW = TIO'EI a(y)wO' for all y E Fpx and wE Z[I].
4.3.1 Archimedean Automorphic Forms
Let us recall the definition of the adelic Hilbert modular forms and their Hecke
ring of level 1)1 for an integral ideal 1)1 of F (cf. [SGL] Sections 2.2-4). Let
To = G~/o be the diagonal torus of GL(2)/o; so, To = Reso/zTo. Recall the
characters c = (C1' C2, c+) as in Section 4.2.5. In particular, c+ : Z(A)/Z(Q) -t
ex is an arithmetic Hecke character such that c+(z) = c1(Z)c2(Z) and
c+(x oo ) = X-(1<1+1<2)+l. Suppose that the conductor of cJ (j = 1,2) is a
factor of an integral ideal 1)1' and that the conductor of c- = c;-lc1 is a factor
4.3 Rank of p-Ordinary Cohomology Groups 193
of SJtISJt / . Recall the "central" character cs : Z(A)So(SJt) --+ ex as in (ex3)

given by cs(zu) = c+(z)lzlA'2c:d'(u) for u E So(SJt) and z E Z(A). To simplify
the notation, we write c for the restriction of cs to So(SJt). We have extended
c to Llo(SJt) by cd'(u) = c2(det(u))C(a')1) for u = (~~) E Llo(SJt). This Neben
type has been used in [F 1].
For Archimedean automorphic forms f : G(Q)\G(A)j K --+ e of level K,
it is standard to define Hecke operators by the sum fl[KgK](x) = Lhf(gh)
if KgK = Uh hK without the determinant factor Idet(g)IA (in (4.44)) in
front of the sum Lh' To adjust our notation to this convention, we take the
twist (I . IA ® f)(g) = I det(g)IAJ(g) as the object corresponding to a section
HO(ShK(q,i!d.K,J if f restricted to ShK(q gives rise to the global section
of i!d.K,c· Indeed, we have (I' IA ® f)1[KgK](x) = I·IA ® (I det(g)IA Lh f(xh)).
This adjustment removes the factor I . 1A'2 from the central character I . 1A' 2c+
of the sections of i!d.K,c' Incorporating this adjustment, the automorphic factor
J K (g, z) of the weight K in the sense of Section 1.1 satisfies J K ( ( td t~ ) , z) =
t~l-I t~2 for ed t~) E To and is given by
JK(g,z) = det(g)KI-Ij(g,z)"'2- Kl+I for 9 E G(JR) and z E f/. (4.70)
Then we redefine SK(91, c; q to be the space of functions f : G(A) --+ e

satisfying the following conditions (e.g., [SGL] Section 2.2).
(SAl) We have the following automorphy
f(axuz) = c+(z)cd'(u)f(x)JK(u OO ' i)-l

for all a E G(Q), z E Z(A), and u E So(SJt)Cj for the stabilizer C j in
G(JR)+ of i = (R, ... , R) E 3 = SjI, where G(JR)+ is the identity-
connected component of G(JR);
(SA2) Choosing u E G(JR) with u(i) = z for each z E SjI, define a function
fg : 3 --+ e by fg(z) = f(gUDO)JK(U OO , i) for each 9 E G(A(oo)). Then
fg is a holomorphic function on 3 for all g;
(SA3) fg(z) is exponentially decreasing as Im(z) --+ 00.
As we now show, the right translations of the functions f satisfying (SAl-
3) generate the (left) automorphic representation f H Idet(g)IA(flg-l) for
9 E G(A(oo)) whose central character is c+ (we studied in (4.61) its in-
verted right automorphic representation). If we replace the words "exponen-
tially decreasing" in (S3) by "slowly increasing" at most polynomial growth
as Im(z) --+ 00, we rediscover the definition of the space of modular forms
GK(SJt, c; q.
The function fg in (SA2) satisfies the classical automorphy condition
fgh(z)) = c-1(g-1,g)fg(z)JKh, z) for all , E ro,g(91), (4.71)
where rO,g(SJt) = 9 . SO(SJt)g-lG(JR)+ n G(Q), and G(JR)+ is the subgroup of

G(JR) made up of matrices with a totally positive determinant. Indeed, we
have under the notation in (SA2),
fg('y(z)) = f(g('yu)ex;) J", (('yu)OCJl i) = f('y,,((ex;)-l g. Uex;)J", (('yU)ex;, i)

= 10- 1 (g-l"(g)f(guex;) J", ('y, uex;(i))J",(uex;, i) = 10- 1 (g-l"(g)fg(Z)J",('y, Z).
The same argument applied to a E G(Q)+ yields
(4.72)
By (SA3) combined with (4.72), fg is exponentially decreasing towards all

cusps of T o,g(I)1). It is easy to see that G", = 0 unless 11:1 + 11:2 = [11:1 + 11:2]1 for
[11:1 + 11:2] E Z. We simply write [II:] for [11:1 + 11:2] E Z if G", =f. o.
Let us now show that this new definition of 8",(1)1,10; C) is compatible
with the earlier one, recalling the theory expounded in Section 4.1.3. We
take 1'(A, A, C, w)w@k E H O(8h so ('Jl.)' C!d.""c:jd. The category of test objects
(A, A, C, w) with polarization ideal e is equivalent over C to the category of
triples (£, A : £ 1\ £ ~ e*, C) for a-lattices £ of Fc = C 1 and a subgroup
C of Fc! £ cyclic of order 1)1. The abelian variety A has a uniformization
A(C) ~ Fc! £ of complex manifolds. The differential w corresponds to du =
LO" duO" for the variable U = (uO" )O"E1. Since £ is projective of rank 2 over 0,
we find two fractional ideals (0,e- 1 ) of F and a base (Wl,W2) of £ so that
£ = O*Wl + e-1w2 with z = W2 Wl
E 3 and 0*1)1-1/0* = C. The alternating
form A is given by A(( -a· WI + b . W2), (-a' . WI + b' . W2)) = a'b - ab'; so,
the positivity Im(z) » 0 implies the positivity of the induced polarization
A : A --+ t A ([ABV] Chapter I). Multiplication by a = (21fi)w;-1 sends Fc! £
to Fc! £~ for £~ = £z in Section 4.1.3. We have (a*)du = aw. Thus putting
1'(£,A,C) = 1'(A,A,C,W), we find f'(£,A,C) = a-k1'(£z,Az,Cz ) for C z =
aC = (0*1)1-1 + e- 1z) 1£z. In other words, the restriction of the untwisted
1 . IAI @ f for an automorphic form f satisfying (SAl-3) to 9J1( e, To (1)1) )(C)
is determined by the function 1'(z) = f'(£z,Az,C z ) satisfying (4.71) because

of (4.2) and (4.9) (combined with the action of 8 0 (1)1) given by c). Taking
C = (8 n E G(A(ex;)) with cO n F = e, we have To,c(l)1) = To (e,I)1), and
we find the tuple U:(z)}c giving {lcI A1 fc(z)}c for the corresponding Hilbert
modular form f : G(A) --+ C as in (SAl-3).
By the cuspidality condition, l' is exponentially decreasing towards each
cusp, since the q-expansion matches with the Fourier expansion substituting
exp(Tr( ~ z)) for qf;. Thus we find a canonical embedding of
(as e runs over Cl[2] = Clil(Cli)2) into 8",(1)1,10; C) as in (SAl-3). A function

f satisfying (SAl-3) gives a holomorphic section of the analytic line bundle
C!d.~~ induced by C!d.""c:jc. Then by GAGA (see [Se] and [ALG] Appendix B),
we have (from the quasi-projectivity of 8h(G, X)(C))
(4.73)
Thus the two definitions are compatible over <C.

For each y E L'l0(1J1), we can decompose
(4.74)
u,t
for finitely many u E u(i) and t E Tc(A((X))) with det(t) = det(y) (see [IAT]
Chapter 3 or [MFG] 3.1.6). Since E~ is well-defined over L'l0(1J1) and coincides
with Z H E+(z'l1)E2(det(z('l1))) on Z(A((X))) n L'lo (1J1) , we may extend it to the
subgroup of G(A((X))) generated by Z(A((X)))L'lo(lJ1) (which contains L'l0(1J1)')
so that ~(z6) = E+(z'l1)E2(det(z('l1)))E~(6) for 6 E L'l0(1J1) and z E Z(A(oo)).
We then have
u,t h
(4.75)
The second identity (*) follows from det(t) = det(y), (ut)' = det(t)(ut)-l and
multiplicativity of ~, and by this, the sum is independent of the choice of u and
t as long as det( t) = det(y). It is easy to verify that the operator defined by
(4.75) preserves the space G" (1J1, E; q and 5" (1J1, E; q by confirming (SAl-3)
for fl[5 0 (IJ1)y'50 (1J1)].
The above description (4.75) of the Hecke operator [50 (lJ1)y'50 (1J1)] looks
different from the earlier definition in Sections 4.2.5 and 4.2.6 (especially
(4.44)), because in the earlier definition, we used x H X· h- 1 (x E 5h(G, X))
instead of 9 H 9 . ut (g E G(A)). By applying the involution ", we can take
h = (ut)'. We claim that the two definitions are the same (for the choice
v = 00 in (4.44)). Since the central characters are equal, the two actions coin-
cide for each central element z E Z (A (00)). Thus in order to show that the two
operators coincide for all [5 0 (IJ1)y' . 5 0(1J1)] with y E L'lo (1J1) , we may assume
that y = (6 ~q). For such y, we have t = y' or t = Y in (4.74). If t = y', u
runs over a complete representative set for U ( 0 / q) in U ( 0 q ), and if t = y, we
may take u = 1. We take the decomposition
As before, we write [z,g] for a point of 5h(q represented by (z,g) E

3 x G(A((X))). We have [z,'Ig] = ['I- 1 (z),g] for 'I E G(Q)+. Thus taking
Lc = 0*E9C 1 and putting.C = 0*+c- 1 z for z E 3, we have XzUC) = eel lC.
By computation, if CWq = ~C' for c' in the representative set for Clt, we find
[z,cuy'] = [z,'Iu'ly'cs u] = b;,l'1,;-l(z),cs u]
for Su E 5 0(1J1) and a finite idele c' with c' 5 n F = c/ , where 'Iy' = (~n
and'lu = (61) with v E c running through the congruence classes in c/qc.
Similarly we have
[z, cy] = b;l(Z), C5]

with 5 E 5 0(1)1) and 'Y E G(Q)+. Since
ro( c, I)1h;l ro( c', 1)1) = ro( c, I)1h;l u Uro( c, I)1h1/,,:;-1,
u
X~-I(Z)
ry
and XTyL- I JU- I ( Z ) (u E U(O/q)) give all the quotients of Xz by cyclic
subgroups of order q. Indeed for, = (~ ~) E ro( c, I)1h;l ro( c', 1)1), we have a
natural inclusion £z c (cz + d)£,(z) = L, (f) and an exact sequence
(CZ+d)-1
0--+ (cz + d)£,(z)/ Lz --+ Xz ) X,(z) --+ 0
for (cz + d)£,(z)/ £z ~ O/q. Thus the two sets of abelian varieties,
coincide, where Yh is as in (4.44) made from A = X z . One can also check that
this bijection extends to include the level structures, the polarizations and
the differentials in (4.44) for KgK' = 50 (l)1)y50(1)1). Thus the new action of
[50 (l)1)y'50(1)1)] for y E Ll o (l)1) coincides with the one [50(I)1)y50(1)1)] defined
in (4.44). By linearity, this action of double cosets extends to the ring action
of the double coset ring R(50 (1)1) , Ll o(I)1)).
Each member f of 5,,(1)1,10; C) has a Fourier expansion of the following
form ([MFG] Theorem 3.10 and [SGL] Sections 2.3-4),
J(t ~) = lylA L aoo(~y, J)(~Yoo)-"ledi~yoo)eA(~x). (4.76)

O«~EF
Here y t-+ aoo(y, J) is a function defined on y E F-;

only depending on its
finite part y( 00) , and (Yo ~) t-+ y;;'''1 e F(iyoo) is the restriction of the canonical
Whittaker function of GLt (lR) to matrices of the form (o?) (whose Mellin
transform gives the optimal r-factor of the standard L-function of J). The
function a oo (y, J) is supported by the set (3 x F(0) n F-; of integral ideles.
The determinant factor lylA = Idet (t ~ )IA. in front of the summation LO«';EF
comes from the twist we mentioned earlier.
In view of the decomposition
5 0(1)1) (~q n 5 0(1)1) = (6 ~q ) 5 0(1)1) u U

uEOq/q
(~q ~) 5 0(1)1)
for a prime q f 1)1, we can directly verify by computation (see [MFG] Corol-
lary 3.11) the following Archimedean counterpart of (4.65),
if q f 1)1,
if q 11)1.
(4.77)
We call an idele y E F;integral if y(oo) E O. For each Ql-algebra R C C

containing the values of characters EJ and K, on Tc(Ql), we have
By definition, writing fy for f9 in (S2) for 9 = (y<;) ~), we see easily that
(4.78)
Then by the q-expansion principle, we find S,.,(SJ1, E; R) = HO(Shso('Jl) ' lJd.",c/ R)'
and hence the definition is compatible with the earlier one. We restate (4.43)
slightly more generally for a W-algebra R C Ql with ioo : ij '---+ C:
We recall the embedding ip : ij '---+ ijp. Then for any Qlp-algebra R, we find,
consistently with (4.43),
By linearity, (y, f) r-+ aoo(y, f) extends to functions on F; x G,,(SJ1, E; R) with

values in R. Then we rediscover from (4.78) the p-adic q-expansion coefficients
ap(Y, f) of f E G,,(SJ1, E; R) by the following formula,
(4.79)
The formal q-expansion of f has values in the space of functions on

T(A. (CXJ)) with values in the formal monoid algebra R[[q~ll~EF: of the mul-
tiplicative semi-group F;, which is given by
f(y) = L ap(~y, f)q~. (4.80)
~»O
This is the p-adic analogue of the Archimedean Fourier expansion (4.76). In

particular, if ap(Y, f) = 0 for all integral ideles y, the modular form f vanishes.
In other words, the q-expansion: y r-+ f(y) determines f uniquely (for any
algebra R for which the space of R-integral modular forms is well-defined).
By the q-expansion principle, for any p-adically complete W-algebra R in
ijp (the p-adic completion of ijp), we conclude
As we have already seen in Section 4.2.5, these spaces have geometric mean-
ing as the space of R-integral global sections of a line bundle of the Hilbert
modular variety of level SJ1, that is, S,,(SJ1, E; R) ~ HO(ShJ;~('Jl),lJd.""c/R) whose
198 4 Hilhert Modular Varieties
right-hand side has meaning even for W-algebras not in ijpj so, we define
8 K (I)1, cj R) for any p-adically complete W-algebra by the right-hand side.
We hereafter assume
either pll)1 or [K] 2': 0, (4.81 )
in order to ensure the stability of the spaces of R-integral modular forms under
Hecke operators (see Theorem 4.28).
Proposition 4.33 Let R be an algebra for which K : Tc(R) ---+ R X and

c : Tc(Z) ---+ R X are well-defined. Suppose that the space of cusp forms
8 K(1)1, cj R) is well-defined. Write H = hK (1)1, cj R) and 8 = 8 K(1)1, cj R). Let
V be an H -module and V' be an R-module of finite type with an R-bilinear
product ( , I : V X V' ---+ R. Then we have
(1) The following formal q-expansion for v E V and wE V',
f(v ~ w)(y) = 2)vITp(~Y)' wlq~

~»o
gives a unique element of SK (1)1, cj R).

(2) The map v~w r---t f(v~w) gives an R-linear map ofV~R V' into 8 with
f((vITp(Y)) ~ w) = f(v ~ w)ITp(Y). If furthermore, V'is an H-module
and (vlh, WI = (v, wlhl for all v E V, wE V', and h E H, then the map
f induces an H -linear map V ~H V' ---+ 8 K (IJ1, cj R).
(3) Suppose that 11' is an R-algebra direct summand of H, and put V(1I') = 1I'V
and 8(11') = 11'8. If V (11') is 11' -free of finite rank and HomR (V (11'), R) is
embedded into V' by the pairing ( , I, then the map f : V (11') ~ H V' ---+ 8(11')
is surjective.
The formulation of this proposition was suggested by the expression of the
theta correspondence given in [Sh8] II, Theorem 3.l.
Proof. We have an isomorphism L : HomR(H, R) ~ 8 given by ap(Y, L(¢)) =
¢(Tp(Y)), which is an H-linear map (i.e., L(¢ 0 h) = L(¢)lhj see the proof of
Theorem 4.30). Since V is an H-module, h r---t (vlh, WI gives an element of
HomR(H, R) and hence an element in 8. The element has the expression as
in (1) by the above explicit form of L. The assertion (2) is then clear from
(1). As for (3), by the isomorphism HomR(V(1I'),R) y V', each element
of Hom(1I', R) ~ 8(11') is a finite R-linear combination of h r---t (vlh, WI for
v E V (11') and w E V' j so, the surjectivity follows. 0
4.3.2 Jacquet-Langlands-Shimizu Correspondence
In order to create an example of the module V in Proposition 4.33, we study

cohomology groups on quaternionic Shimura varieties. See [62a], [63c], and
[68c] in [CPS] and also [H94] and [H88] for more details of such cohomology.
Let D be a quaternion algebra over F, that is, a central simple algebra of

dimension 4 over F (see [BNT] for generality of simple algebras and [AAQ]
for the theory of quaternion algebras). We write G D for the algebraic group
defined over Q such that GD(R) = (D 01(l1 RY for each Q-algebra R. Thus if
D = M 2(F), we have G = G M2 (F). Let d(D)2 be the discriminant of D; so,
d(D) is the product of primes q such that Dq = D 0F Fq is a division algebra.
We assume that p t d(D) and define subsets JD and JD of J by
D lR ~ {M2(lR) if (J E JD (4.82)
0F,a - lHI if(JEJ-JD=JD,
where lHI is the Hamilton quaternion algebra over R Finite places dividing
d(D) and Archimedean places in JD are called the ramified places for D IF . We
can create such a quaternion algebra as follows: Choose a quadratic extension
K of F so that Kq = K 0F Fq is a field for each prime qld(D) and that
Ka = K0F,alR = <C for (J E JD. Find 6 E F such that 6 E NKqIFq(KqX) if
and only if the place q is outside d(D) and JD. We find such a 6 if and only if
the number of ramified places of D is even (see [BNT] IX.3 and XIII.6), and
D ~ {(-8bP d'p) E M 2(K)la,b E K}, where p is the generator of Gal(K/F).
By our choice of 6, we find 0 =I- x = (-8bP d'p) E Dq with N(x) = aa P +6bbP = 0
if and only if the place q is outside JD and d(D). Thus Dq is a division algebra
if and only if q divides d(D) or in JD. The map N : D ---+ F and the trace
of the matrix (-8bP d'p) are called the reduced norm and the reduced trace
Tr: D ---+ F of D, respectively. By our construction, D '-+ M 2(K) induces an
isomorphism D 0F K ~ M2(K) which sends b 0 k to kb.
We fix once and for all an extension of (J : F '-+ Q to (J : F ~ Q for an
algebraic closure F / F. We take a quadratic extension K/ F inside F so that
K 0F,a lR ~ lR x lR as F-algebras for (J E JD , K 0F Fp ~ Fp x Fp for primes
pip and D 0F K ~ M2(K). We can always choose such a quadratic extension
K as long as p t d(D). These conditions imply K 0F lR ~ <C for (J E JD.
We identify D 0F K with M 2 (K) by the above isomorphism. We fix max-
imal O-orders OD and OK of D and K, respectively, and we suppose that
OD 00 OK C M 2(OK). Here an O-order OH cHin a finite-dimensional
semi-simple F-algebra H is an O-subalgebra of H such that for every x E OH,
the commutative subalgebra O[x] cHis an O-module of finite type and
OH 0z Q = H. If OH is an O-order, its conjugate aOHa- 1 for a E H is
another O-order. There are finitely many nonisomorphic maximal O-orders
inside H up to conjugation.
For each prime ideal [prime to d(D), we fix an isomorphism OD,[ ~ M 2(O[)
so that for the p-adic place pip induced by ip 0 (J, this isomorphism coincides
with the one, OD '-+ M 2(OK) tpM) M2(Op). By means of this isomorphism,
we identify D[ with M2(F[). For an integral ideal 1)10 of F prime to d(D),
putting 1)1 = 1)10 d(D), we define
rf (1)1) = {x E 8~ Ixmo = (~ ~) with c E 1)100mo} , (4.83)

- - D -1 -D
where OD = OD ®z:2:, and 0<)10 = I1!1<)1o Ot. We set So (91) = 8 To (91)8 for
8 = ( d(d 6D)) ~) with a finite idele d such that dO nF = D. Similarly we define
Llcf(91) c D ®<Ql A(oo) so that it is the product of local components Ll t which
coincide with the local components of .1 0 (91) as long as [f d(D) and Ll t = OD,t
if [ld(D). Again we can think of the double coset ring R(Sf(91), Llcf(91)) ~
R(Sf(91), Llcf(91Y) (XL = N(x)x- 1 for the reduced norm map N : D --t
F). We have T(t) = sf (91) (~{ nSf (91) and T(t, t) = sf (91)wtSf (91) in
R(Sf(91), Llcf(91)L) for [ f d(D), because the local component Llcf (91) t at [
of Llcf(91) is identical to Ll o(91)r. For [ld(D), we take at E OD,t so that its
reduced norm generates [Ot. Then we define T(t) = -Sf(91)atSf(91) for
[ld(D), and we have
R(Srf (91) , Llcf(91)) ~ R(So(91), .10(91)). ( 4.84)
The above isomorphism brings T(t) and T(t, t) to the corresponding elements
in the right-hand side. As an operator on the space of automorphic forms,
T(t, t) induces the central action (t) of wt·
For a given ring R, we consider the following module L(",*; R) of the mul-
tiplicative semi-group M2(R). Let ",* = ("'1 + 1, "'2) and put n = "'2 - "'1 - 1 E
:2:[1], which is the restriction of ",* E X(Tc) to T c G 1 , and we con-
firm (",IT)* = k - 21 = n. The correspondence '" --t ",* is the one for
G = ResF/<QlGL(2) we discussed in Section 1.1 (see Section 5.3.2 for more
details of this fact). We suppose that n ~ 0 (i.e., na ~ 0 for all (J E 1), and
we consider polynomials with coefficients in R of (Xa, Yo- )aEI homogeneous of
degree na for each pair (X a, Ya). The collection of all such polynomials forms
an R-free module L(",*; R) of rank I1a(n a + 1).
As before, we write v for the fixed place p or 00; so, the base ring B is
W if v = p and CC if v = 00. Suppose that R is a B-algebra. Then iv((J(ov))
(which we write simply (J(ov)) for 0 E GD(A) can be regarded as an element in
M 2 (R). Take a Neben character c as in (exl-4) of Section 4.2.6 with cIT(z)
factoring through (Oj91 o)x. We define cJ (j = 1,2) as in (ex1) and extend
C to Llcf(91) by c15(o) = c2(N(0))C(a) if 0<)10 = (~~). Since 8 --t c15(8)
and Z r-t c+(z) coincide on Z(A(oo)) nSf(91), we may extend c to a character

c'fy : Sf(91)Z(A(oo)) --t R X by c'fy(zu) = c15(u)c+(z) for z E Z(A(oo)) and u E
Sf (91). We let Llcf(91) and Z(A)Sf(91)G D (l!{)+ act on L(",*; R) as follows.
O.p((~:)) =c15(O)-lN(Ov)"lP((JWOvY)(~:)), (4.85)

(zu) . p ((~:)) = c'fy(z(oo)u(oo))-l N(ZVUV)'<lp ((J(S(ZVUV)L) (~:)) .
Here Z E Z(A), U E Sf (91)G D(l!{), and OL = N(O)O-l and So = 808- 1 for 8

given just below (4.83). We write L(",* c; R) for the module L(",*; R) with this
(Llcf (91), Z(A)Sf (91)G D (l!{) )-action.
By computation, Z E Z(A) acts on L(",*c; R) through scalar multiplication
by c+(Z)-lz~1+K:2-I; in particular, E E O~ c Z(A) acts trivially on L(",*c; R).
If S is a sufficiently small open compact subgroup so that S n Z(Ql) C O~ in

G(A(CXJ)), central elements in rx
= xSx- 1 n GD(Ql) act trivially on L(f1,*E; R).
Assume that v = 00. We let 9 E GdlR) with N(g) »0 act on 3D = SjID
by the linear fractional transformation of gu = O"(g) E GL 2 (K Q9K,u lR) =
GL 2 (lR) component-wise. We put C u + for the stabilizer of A in the Lie
group (D Q9F,u lR)X and define a maximal compact subgroup cf C GD(lR)
modulo center by TIUEID Cu+ x TIUEID liP, which is the stabilizer in GD(lR)
of i = (A, ... , A) E SjI D. Thus we have 3D = SjID ~ GD(lR)+ Icf by
g(i) H 9 for the identity-connected component GD(lR)+ of GD(lR). Since G
and G D have the common center Z = ResF/IQiG m , we use the same symbol
Z to indicate the center of G D. For any open compact subgroup S C Sr? UTI),
we think of the complex manifold associated with the Shimura variety:
Y(S) = yD(S) := GD(Ql)\GD(A)IZ(A)S· cf.
We write simply YOD(sn) for Y(Sr?(sn)).

Since we have already defined the space of Hilbert modular forms over C
in (SAl-3), assuming that D is a division algebra, we state a definition of
automorphic forms on GD(A). For each f1" we define f1,D E £::[J]2 by f1,D =
(f1,P,f1,?) for f1,If = LUEID f1,j,uO"· Thus f1,D is the projection of f1, to £::[J D]2.
Similarly, we define f1,D by the projection of f1, to £::[JD]2 c £::[J]2.
With each f1, E £::[J]2 = X(Tc), we associate an automorphic factor,
(4.86)
for 9 E GD(lR) and Z E 3D. We write f1,*,D for the projection of f1,* to £::[J D].
Then, if JD T 0, we define S~ (sn, 10; q to be the space of functions J on
GD(A) with values in L CiD(f1,*,D,q satisfying the following conditions.
(SB1) We have the following automorphy
J(exxuz) = E+(Z)E~(U(CXJ))u;;;,l . J(x)J;;(UCXJ,i)-l

for all ex E GD(Ql), z E Z(A), and u E Sr?(sn)Ci , where GD(lR)+ is the
identity-connected component of GD(lR). Here J(x) 1--7 UCXJ' J(x) is the
action of the JD-component u~ ofuCXJ on LCD (f1,*,D; C);
(SB2) Choosing u E G D (lR) with u D = 1 and u(i') = z for each z E 3 D,
define a function Jg : 3D -+ C by Jg(z) = J(guCXJ)J"(uCXJ' i) for each
9 E G (A (CXJ)). Then Jg is a holomorphic function on 3 D for all g.
When JD = J (s- JD = 0), the variety yD(S) is a finite set of points; so,
the condition (SB2) is empty. However, writing S~ (sn, 10; q for the space of
functions satisfying (SB1) in this definite case, we need to take S~ (sn, 10; q
to be the following quotient: S~ (sn, 10; q = S~ (sn, 10; q I Jv(sn, 10; q, where
Jv(sn, 10; q is the subspace made up of functions in S~ (sn, 10; q factoring
through the reduced norm map N : GD(A) -+ T(A). If f1,2 ~ f1,1 T J or 10
is nontrivial for some x E GD(A) with N(x) = 1, Jv(sn, 10; q = 0; so, no
modification is necessary. Decomposing sf? (sn)yL sf? (sn) for y E f1{5 (sn) into
Uh hSf? (sn), we shall define the action of R(Sf? (sn), f1{5 (sn)) on S~ (sn, E; q
by
(4.87)
h h
We state the theorem of Jacquet-Langlands and Shimizu in the following way.
Theorem 4.34 Suppose sn = snod(D) for an integral ideal sno prime to d(D).
Identify R(So(sn), f1o(sn)) and R(Sf?(sn), f1{5(sn)) as in (4.84). Then we have
an R(So(sn), f1o(sn))-linear embedding S~(sn, E; q y S,,:(sn, E; q for all dom-
inant regular"" with ""? - ""p :.0- I D. The image of this embedding only depends
on d(D) and is made up of cusp forms in S,,(sn, E; q new at all primes qld(D).
In particular, if d(D) = 1, the above morphism is a surjective isomorphism.
Since known proofs of this theorem require harmonic analysis and a good
knowledge of representation theory of adele groups (especially the theta cor-
respondence), the proof is beyond the scope of this rather algebraic book; so,
we quote some references where one can find a proof. The above formulation
of the theorem was given in [H81] Section 2 and [H88] Theorem 2.1, where
one can find an exposition of how to deduce this result from the original result
of Jacquet-Langlands (whose exposition can be found in [AAG] Section 10).
We hope to give an exposition of this result in a forthcoming book (through
a line closer to Shimizu's treatment).
4.3.3 Integral Correspondence
Since the isomorphism of Theorem 4.34 is noncanonical, we rigidify it using

some ideas of Shimizu and Shimura. Although we can treat the general case
in a uniform manner, introducing the cuspidal cohomology group, in order
to avoid technicality, we assume at the end of this section that IIDI -:; 1,
which is the case we need later. For the moment, we do not impose any con-
dition on D; also, v can be any of the places p and 00. Suppose that S is
sufficiently small so that the image ['s,x of rs,x = xSx-1Gi;(JR) n GD(Q) in
PG D (JR) acts freely on S/ D for all x E G D (A (CXJ)) and that the action of r S,x
on L(""*E; R) := LCD (""*c; R) factors through ['s,x. As already remarked,
the central action of E E Z(Q) n S· GD(JR) on L(",,*c; R) is the multiplica-
tion by E+(E(CXJ))-lE"1+"2- I , which has to be trivial. Then we can define an
etale space over Y(S): £(""E; R) = G D (Q)\ (G D (A) x L(""*E; R)) /Z(A)S· Cp,
where ,(x, p)zu = (rxzu, (ZU)-l . p)) for u E S . C jD , z E Z(A), and
, E GD(Q). This etale space gives rise to a sheaf L(""*E;R)/y(s) of locally
constant sections. We write Hq(y(S), L(""*E; R)) for the cohomology group of
Y(S) with coefficients in the locally constant sheaf L(",,*c; R).
Since Y(S) '::'"' Ux['x \SJID for finitely many x with Xv = 1, we have a
canonical isomorphism (cf. [LFE] Appendix and [H94] page 470),
where the right-hand side is the direct sum of the group cohomology of the r x-
module L(/'i,*c; R). Even if rs,x does not act freely on the module L(/'i,*c; R),
r
we still have Y (S) ~ Ux x \JjI D for finitely many x with xp = 1, and we
define the left-hand side of (4.88) by the right-hand side of (4.88).
If S c S', we have the trace map Trsl IS (i.e., the transfer map in group
cohomology; see [MFG] 4.3.1) and the restriction map Ressl IS:
TrsIIS: Hq(yD(S),L(/'i,*c;R)) --+ Hq(yD(S'),L(/'i,*c;R)),

Resslls: Hq(yD(S'), L(/'i,*c; R)) --+ Hq(yD(S), L(/'i,*c; R)).
We choose S sufficiently small so that [Sf (91) : S] is prime to p (this is a

condition on p). Again we have Tr and Res:
Tr : Hq(yD(S), L(/'i,*c; R)) --+ Hq(YOD(91), L(/'i,*c; R)),

Res: Hq(YOD(91), L(/'i,*c; R)) --+ Hq(yD(S), L(/'i,*c; R)).
Since Tr 0 Res is the scalar multiplication by [So(91) : S] (cf. [MFG] 4.21), we

have
We can always choose a multiple 91' = 91q (by a prime q) of 91 so that O,x (91') r
acts freely on JjID.
We have a natural action of the ring R( sf (91), L1~ (91)) on the cohomology
group Hq (YOD (91), L(/'i,*c; R)). The definition of the action is quite similar to
the one (4.37) given for the coherent cohomology groups HO(ShK,id.~), and
the operator induced by the double coset SgS for S = Sf (91) is given by
[S8S]v = N(8 v )-1<1 (TrSI(Osns) 0 [8] 0 ResSI(snsO)), (4.89)
Here [8] : L(/'i,*c; R) --+ L(/'i,*c; R) is induced from the action of 8 on L(/'i,*c; R).
If q = 0, decomposing Sf(91) (g n Sf(91) = U"" wSf(91), we have
(4.90)
Suppose ID = 0. Then yD(S) is a finite set, and C j = GDOR). For a W-

algebra R, we may identify H O(Y D (S), L( /'i,* c; R)) with the space of functions
f : G(A) --+ L(/'i,*; R) satisfying
(SB) We have the automorphy f(axuz) = c+(z)cJ5(u(oo))u;l . f(x) for all
a E GD(Q), Z E Z(A), and u E Sf(91)CF.
Since f satisfying (SB) is left invariant under GD(lR), we may regard it as a
function on G(A(oo)) = G(A)/G(lR). Recall Iv(/'i" c; R) which is made up of
functions as in (SB) factoring through the reduced norm N : G(A) -+ T(A).

We have, if ID = 0,
(4.91)
If R is p-torsion (so, B = Wand v = p), Iv(91, 10; R) could be nontrivial even

if K,2 - K,1 i- I, because K,2 - K,1 - I could be trivial as a character of T(R)
(even if it is nontrivial as a character of the algebraic group T).
in (S~(p91) (Po" n
We assume that R = Wand v = p. We have pg right cosets (g = [F : Q])
S~(p91)) /S~(p91), and each coset acts on f E Iv(91,c;R)
by f(x) H f(xp p), regarding f as a function of T(/li/ oo »). Since factually
factors through a finite quotient of T(A(oo»), for a high power U(p)n = U(pn)
of Up(p), we have fIU(pn) = png f; so, Iv (91, 10; R) is killed by the p-ordinary
projector e. Thus we have
(4.92)
where the superscript or the subscript n.ord indicates the image of the projec-
tor e = limn-+oo U(p)n!. Since T(p) == U(p) mod p if K,2 - K,1 ::::: I and [K,] ::::: 1,
the same assertion holds for 91 prime to p:
S~·oTd(91,c;R) = H~.oTd(YOD(91),L(K,*c;R)) if K,2 - K,1 ::::: I and [K,] ::::: 1.
Hereafter we assume that IIDI ~ 1 and F i- Q. Write q = IIDI. In this case,

D is a division algebra, and Y(S) is a compact manifold (cf. [Bo] 1.5). Let
V be the image of Hq(Yrf(91), L(K,*c; W)) in Hq(YOD(91), L(K,*c; W @;z Q)).
By the Eichler-Shimura isomorphism (described in the following sUbsection)
between the space of cusp forms on GD(A) and the cohomology group (e.g.,
[H94] Proposition 3.1 and (10.4)) combined with Theorem 4.34, the above
cohomology group V is a module over the Hecke algebra hI< (91,10; W).
We consider the duality pairing [ , ] on L(K,*c; R) (for Q-algebra R) intro-
duced in [H94] Section 5,
where n = K,2 - K,1 - I, G)

= TIaEI G:),
and, for example, Xl = TIaEI X~a.
As So(91)Z(A)-modules, this pairing satisfies:
(4.94)
where N : Dp -+ Fp is the reduced norm map and E+(Z) = c+(z)z;1<1-1<2+I

(the p-adic avatar of 10+).
Define "'K, = (I - K,2, 1- K,1). Then the pairing [ , ] induces So(91)Z(A)-
equivariant pairing [ , ] : L(K,*c; R) x L("'K,c 1; R) -+ R. Recall q = IIDI =
dime SJI D. Then by the compactness of YoD un), the cup product pairing in-
duces ([H94] (5.3)), a nondegenerate pairing
Thus we obtain from Proposition 4.33 and Th80rem 4.36 the following result.
Proposition 4.35 Let V c Hq (YoD ('J1), L(K,*c; W ®z Q)) be as above Jor
q = lID I ::; 1. Suppose F oj Q. Let V* be the dual W -lattice oj V in the
cohomology group Hq (YOD ('J1), L(f:Zc 1; W ®z Q)) under the Poincare duality
pairing ( , ) as above. Then we have an h" ('J1, c; W) -linear map J : V ®w V* --+
S,,('J1,c;W) defined by the q-expansion: J(v ® w) = Lt;»o(vITp(~y),w)qt;,
where we regard the space V ®w V* as an h,,('J1, c; W)-module through the left
Jactor V.
A similar fact for the matrix coefficients of T(y) in place of (vIT(y),w) has
been proven in [Sh8] II, Theorem 3.1 by analytic means without using the
Jacquet-Langlands correspondence.
4.3.4 Eichler-Shimura Isomorphisms
In this subsection, we assume B = C and v = 00. We relate S;; ('J1, E; C)

and Hq(YOD('J1), L(K,*c; C)). The isomorphism is called the Eichler-Shimura
isomorphism. When lID I = 0, (4.91) gives a direct relation; so, there is nothing
left to do. Thus we assume IIDI = 1 and F oj Q. Then YOD('J1) is a Riemann
surface, which is a disjoint union of ro,x ('J1) \SJ for a suitable finite set of
x E GD(A(oo)). Thus Shimura's theory expounded in [IAT] Chapter 8 applies,
which yields the following canonical isomorphism,
S,,(ro,x('J1), c; C) EEl SK(rO,X('J1), c; C) ~ H O(ro,x('J1), L(K,*c; C))

~ H O(ro,x('J1)\SJ, L(K,*c; C)), (4.95)
where SK(ro,x('J1), E'; C) is made up of complex conjugates of functions in

the space SK (ro,x ('J1), E'; C), and SK (ro,x ('J1), c; C) is the collection of holo-
morphic functions J : SJ --+ LC(K,*,D, C) satisfying the following automor-
phy condition: Jh(z)) = C1(x-l/'x)r . J(z)J;;h, z) for all /' E ro,x('J1).
The fact S;;('J1,E;C) ~ EBxSK(rO,x('J1),c;C) ifYoD('J1) = Uxro,x('J1)\SJ fol-
lows by the computation similar to the one done just below (4.71) using
(SBl-2) in place of (SAl-3). The last isomorphism in (4.95) is a general
non-sense (see [LFE] Appendix). The first isomorphism in (4.95) sends J E
SK(rO,X('J1), c; C) to a 1-cocycle cfh) = Jz'(z) J(z)(Xa+zYa)nadz E L(K,*E; C)
and J E SK(rO,X('J1),E';C) to a 1-cocycle cfh) = Jz'(z) J(z)(X a +zya)nadz E
L(K,*E; C), where ID = {oj. A detailed proof of this fact is given in [IAT]
Chapter 8 (see also [CPS] I, [62a] and [63c] and [LFE] Chapter 6); so, we omit
details of the proof of the following theorem.
Theorem 4.36 (Shimura) Suppose IIDI = 1 and F =1= Q. Then we have a

canonical isomorphism of R( So (91), ,10 (91)) -modules,
S;z (91, c:; q EB S~ (91, t'; q S:! cP

H O(Y (91), L( /'i,* c:; q),
where S~ (91, t'; q is made up of complex conjugates of elements in S;z (91, t'; q
with complex conjugate character t'.
4.3.5 Constant Dimensionality
Recall the Neben character c: = (C:1,C:2,E+) with values in W X as in (exl).

Here E+(Z) = c:+(z(P»)z;1<1-1<2+ 1 for Z E Z(A.too»). We define E= (E1,E2,E+)
with values in W X by the Teichmiiller lift of the reduction modulo p of the
characters c:. We prove the following theorem.
Theorem 4.37 If /'i,2 - /'i,1 ?: I, then dim S;Z,n.ord(p91, c:; q depends only on
E and d(D) and 91. Moreover if /'i,2 - /'i,1 ?: 21, we have
rank SM2(F),n.ord(p91
K, ,
C", W) = rank SM2(F),n.ord(91
K, ,
C", W)
for 91 prime to p.
If the assertion of the theorem holds for 91 containing a sufficiently large
integer N, the assertion holds for all 91, because by the theory of primitive
forms [MFM] 4.6 and [H88] Section 3 (or the strong multiplicity one theorem
[AAG] Sections 4 and 10), one can recover the dimension of each space with
lower level. Here we give a proof under the assumptions either that p ?: 5
and pis unramified in FjQ or that YoD(91) is smooth and the sheaf L(/'i,*C:; R)
is well-defined over YoD (91). The latter assumptions are valid for 91 having a
sufficiently large integer N.
Proof. We choose an open compact subgroup K C So(91) so that yD(K) is
a smooth manifold and that the sheaf L( Ii* c:; R) is well-defined over Y D (K).
We write Y = yD(K) and Y' = yD(K n So(p)). When Yl(91) is a smooth
manifold and L(Ii*C:; R) is well-defined over Yl(91), we take K = Sf(91). By
the same argument as in the proof of Theorem 4.9, the evaluation at (1,0)
of polynomials in L(Ii*C:; IF) gives rise to a morphism L(/'i,*C:; IF) ---+ L('KZ, IF)
of sheaves on Y', where K, = (0,1) and E is as above. Composing with the
restriction map, again by the same argument as in the proof of Theorem 4.9,
we get the following isomorphism, if /'i,2 - /'i,1 ?: 21,
H~.ord(y,L(/'i,*c:;lF)) S:! H~.ord(Y',L(Ii*c:;lF)) S:! H~.ord(Y',L(K,E;lF)).
If 1i2 - 1i1 > I, we only have H~.ord(Y" L(Ii*C:; IF)) S:! H~.ord(Y" L(K,E; IF)). If
1i2-1i1 = I, L(Ii*C:; IF) = L(K,E; IF) (although L(Ii*C:; W) might be different from
L(K,E; W)), and we have H~.ord(Y" L(Ii*C:; IF)) S:! H~.ord(Y', L(K,E; IF)). Thus we
need to choose K so that the dimension dim S;Z,n.ord(91, C:; q only depends
on dimJF H~.ord (Y, L( Ii* c:; IF)). We specify K later.
By Theorem 4.34, dime Sf! (1J1, c; C) is independent of D as long as the

quadruple (d(D),/'\:,c,lJ1) is fixed; so, we may assume that q = lID! s 1,
because, for a given finite set of primes S, we can always find a quaternion
algebra D with d(D) = I1 qE s q and 9 - lID! = IIDI == lSI mod 2 by a Hasse
principle (see [IAT] (9.2.4-6) and [BNT]). Since the projectors e and eo are
defined taking the p-adic limit, they give idempotents of hK (1J1, c; W). However,
every idempotent of hK (1J1, c; W) comes from an idempotent of hK (1J1, c; W) if
W is sufficiently large, since hK (1J1, c; W) = hK (1J1, c; W) ®w W. Thus these
idempotents are well-defined over Wand over C; so, H~.ord = e· H~.ord makes
sense for any coefficient ring.
By definition, Sf! (1J1, c; R) is a quotient of H O(YOD (1J1), L( /'\:* c; R)) when
q = 0, and as we have seen in (4.92),
s;;,n.ord(lJ1, c; R) ~ H~.ord(YOD(IJ1), L(/'\:*c; R)) for all /'\: and c if q = 0.
Similarly, by Theorem 4.36, we have
Since by the universal coefficient theorem (e.g., [SFT] Theorem II.18.3), we

have from the flatness of C and W over W that
H~.ord(YOD(IJ1), L(/'\:*c; C)) = H~.ord(YOD(IJ1), L(/'\:*c; W)) ®w C,

H~.ord(YOD(IJ1), L(/'\:*c; W)) = H~.ord(YOD(IJ1), L(/'\:*c; W)) ®w W.
Thus we need to prove the constancy of rankw H~.ord (Yl (1J1), L( /'\:* c; W))
and that H~.ord(Yl(IJ1), L(/'\:*c; W)) is W-free. Take a prime q plJ1 such that
p f (N(r)2 - 1). As we have already remarked in the proof of Theorem 4.10,
there exist infinitely many such primes [ if p ;::: 5 and p is unramified in F /Q.
We take K = So(lJ1) n SHr). Then HQ(Yl(IJ1), L(/'\:*c; R)) is canonically a
direct summand (as Hecke modules) of Hq(y, L(/'\:*c; R)) by the restriction
map for all W-algebras R, because the restriction followed by the transfer
is a multiplication by an integer factor of [So(IJ1),K]I(N(r)2 -1) (cf. [MFG]
Proposition 4.21), which is invertible in R. Thus we only need to prove these
facts for H~.ord(Y' L(/'\:*c; W)).
We may further choose [so that the image ['x of Tx = x . K X-I n G D(Q) in
GDd(Q) = GD(Q)/Z(Q) is torsion-free for all x E GD(A(oo)). Indeed, if'Y E Tx
is torsion in ['x, Fb] is a quadratic extension of F with 'YN = E E Ox for the
order N of the image of'Y in ['x' Write M for the extension of F obtained
by adding Vu for all u E Ox. Since [Ox: (OX)2] = 29, [M : F] S 29. In
M, ( = 'YJE- I has finite order N; so, ( is a root of unity in Mb]. Since
E == 1 mod [, we may assume that ( == 1 mod..c for a prime factor ..c of [
in Mb]. This shows either ( = 1 or £IN for the residual characteristic £ of L
This never happens if £ is sufficiently large, because N is bounded. If ( = 1,
then'Y has two eigenvalues ±yiE; so, yiE == -yiE == 1 mod [, which happens
only if £ = 2. Thus we have plenty of choice of [ as above. Under the above
choice, Tx c Z(Z) if q = 0 and Y is smooth if q = 1. Since Z(Z) = Ox acts

on L(K*E; R) trivially, L(K*E; R) is a well-defined sheaf over y.
For simplicity, we write H~.ord(Y' L(K*E; R)) as H~ord(L(K*E; R)). By the
pm
long exact sequence attached to L(K*E; W) Y L(K*E; W) --» L(K*E; IF) for the
residue field IF of W, we have another exact sequence:
H~.ord(L(K*E; W))®wWm Y H~.ord(L(K*E; W m )) --» H~~;d(L(K*E; W))[pm].
When q = 0, H~.ord(L(K*E; W)) = 0 because dimY = O. Thus we get

H~.ord(L(K*E; W)) ®w Wm ~ H~.ord(L(K*E; Wm)),
and H~.ord(L(K*E; W)) is W-free.

Suppose q = 1. Then H2(L(K*E; W)) is the dual of HO(L(K,c I ; W*))
by Poincare duality, where W* is the Pontryagin dual Unp-nWjW of W.
Here the adjoint operator of [KgK] acting on HO(L(K,C\ W*)) is given by
[Kg-"K] = [K· N(g)-lK][KgK] for the reduced norm map N : G D -+ Z.
Since the action of the center [K ·N(g)K] is the multiplication by a p-adic unit,
we have the duality between the nearly ordinary part of the two cohomology
groups H~.ord(L(K*E; W)) and H~.ord(L(K,CI; W*)).
We now show H~.ord(L(K,CI; W*)) = O. Since we have
we only need to show that T(p) or U(p) is nilpotent on HO(L(K,C I ; WI)). We

regard HO(L(K,C I ;Wd) as a space of locally constant functions on GD(A)
with values in L(K,; Wd satisfying the automorphy property as in (SB). Choose
an integer n > 0 so that pn == 1 mod L Decomposing t = T(pn) = Uh hK,
we have from (4.90) that Jlt(x) = p-nKl I:h h . J(xh-"). We may assume
det(h) = wp and that h is of the form (a ~) with af3 = pn for u running over
representatives modulo aOp . By the strong approximation theorem, connected
components of Yare indexed by T(A)jN(K)T(F)T(JR)+ = ClF(t) for the
reduced norm map N : G D -+ T. Since J is constant on each connected
component of Y and T(pn) or U(pn) does not move the connected component
(i.e., because the ideal (pn) is generated by a totally positive element pn == 1
mod [ in F), J(xh-") is a polynomial P(X, Y) independent of h.
By the definition of the action, for Ph (X, Y) = J (xh -"), we have p-nKl h .
Ph(X, Y) = P(f3X - uY, aY) for r + s = n. Therefore writing P as a linear
combination of II,. P"(X(J) Y,,), we compute h· P at each piece P"(X,,, Y,,).
Writing Pu(X", Y,,) = p,,(pr Xs - uY",psYs ) = p,,(pr Xs - uY", 0) if s > 0 and
P'(X", Y,,) = P,,(O, Y s ), we see I:u p,,(pr X" - uY", 0) = 0 in L(K,E- I ; Wd.
For U(p), only those p,,(pr X" - uY", 0) appear; so, U(p) = 0 on the co-
homology. Since U(p) = 0, after applying T(p), we only get a monomial
P,,(O, Y s ) of Y". When we consider T(p), p is prime to 1J1. So P(X, Y) is
invariant under Uo(lJ1)p = GL 2 (Op), but a nonzero monomial of Y" is never
4.4 Appendix: Fundamental Groups 209
invariant under GL2(Op); so, P"(X,, , Y,,) does not involve a monomial of Y",
that is, P,,(O, Y,,) = O. This shows the desired vanishing. In this proof of
H~.ord(Y' L("'*E; R)) = 0, we have used the strong approximation theorem
to show that T(pn) and U(pn) for a suitable n > 0 preserve each connected
component of Y; so, obviously the argument does not work if ID = 0 where
the strong approximation theorem for G D ,l(A(CXJ)) (for the derived subgroup
G D,l = Ker(N: G D --+ T) c GD) does not hold.
We find H~.ord(L("'*E; W)) ®w Wm = H~.ord(L("'*E; W m )) by the vanish-
ing of H~.ord(L("'*E; W)). Again by the exact sequence
and the vanishing of H~.ord' we find that H~.ord(L("'*E; W)) is torsion-free if

q = 1. This finishes the proof.
4.4 Appendix: Fundamental Groups
This section gives a sketch of the theory of algebraic fundamental groups

which has been and will be used in the proof of the global reciprocity laws
(for example, Theorem 4.14 and Theorem 6.26). At the end, we summarize
group-theoretic tools also used in the proof of the reciprocity laws.
4.4.1 Categorical Galois Theory
We give a brief exposition of how one can axiomatize Galois theory (the idea
is due to Grothendieck, and the main reference is [SGA] VA). We assume a
minimal knowledge of category theory and terminology which one can find in
[HAL] Chapter II, [GME] Sections 4-5, and [MFG] Section 4. This subsection
interprets the result in Section 2.3.1 in terms of category theory.
We recall some terminology in category theory. Let C be a category. A
morphism X ~ Y in this category is called an epimorphism if for any third
object Z and morphisms Y J!+ Z, 9 0 f = h 0 f {:} 9 = h (so f is a
h
right unit). In other words, for any other object Z in C, the pullback map
Homc(Y, Z) ~ Homc(X, Z) is injective if f is an epimorphism. Similarly a
morphism X ~ Y in this category is called a monomorphism if for any third
object Z and morphisms Z l4 X, fog = f 0 h {:} 9 = h (so f is a left unit).
h
Let SETS be the category of sets. Recall that a covariant (resp. contravari-
ant) functor F : C --+ SETS is called representable if there is an object Fin C
such that we have a system of functorial bijections ¢x : F(X) --+ Homc(F, X)
(resp. ¢x : F(X) --+ Homc(X, F)). Here the word "functorial" means that
f. 0 ¢x = ¢y 0 F(f) in the covariant case and 1* 0 ¢y = ¢x 0 F(f) in the
contravariant case for any morphism X ~ Y in C, where f* (g) = fog for
9 E Homc(F, X) and f*(h) = h 0 f for h E Homc(Y, F). The object in C

representing a functor is unique up to isomorphisms (a characterization by a
universal property; see Section 3.1.5 in the text and [GME] Lemma 4.1).
If a covariant functor Fo sending every object in C to a singleton {o}
is representable in C, the object representing the functor is called an initial
object in C. Similarly if a contravariant functor F O sending every object in C
to a singleton {o} is representable in C, the object representing the functor is
called a final object in C.
If the covariant functor T f-7 Home (X, T) x Homc(Y, T) for two given
objects X, Y of C is representable by an object of C, the object is written as
XEBY and is called the direct sum, and we have inclusions ix : X -t X EB Y and
iy : Y -t X EB Y such that Homc(X EB Y, T) ~ Homc(X, T) x Homc(Y, T) is
given by ¢ f-7 (¢oix, ¢oiy) (see [MFG] 4.4). If the contravariant functor: T f-7
Homc(T, X) x Homc(T, Y) for two given objects X, Y of C is representable by
an object of C, the object is written as X x Y and is called a direct product,
and we have projections px : X x Y -t X and py : X x Y -t Y such that
the isomorphism Home (T, X x Y) ~ Home (T, X) x Home (T, Y) is given by
¢ f-7 (Px 0 ¢,py 0 ¢) (see [MFG] 4.4).
For a given object Sin C, we write CIs for the category of arrows X .L. S.
Morphisms from X .L. S into Y !4 S are given by morphisms h : X -t Y
with go h = f. If we apply the above construction to CIS, the direct product
in CIs if it exists is written as X x S Y and is called the fibered product of
X and Y over S (see [MFG] 4.2.1). So in this case, the contravariant functor
T f-7 Homels(T, X) x Homels(T, Y) is representable by X Xs Y in CIS (see
[GME] 5.2). If C = SETS, we see easily that
X Xs Y = {(x,y) E X x Ylf(x) = g(y)}.
Dualizing the above construction, for an object S of C, we consider a

category CS whose objects are arrows S .L. X and ¢ E Hom(S .L. X, S !4 Y)
is a morphism ¢ : X -t Y with ¢ 0 f = g. Then X EB s Y in CS if it exists
represents the covariant functor T f-7 Homes (X, T) xHomes(Y, T). The object
X EB s Y is called the fibered sum of X and Y under S. If C = SETS, we see
easily that X EB s Y is the quotient of the disjoint union X U Y under the
relation that x "-' y ~ x = f(s) and y = g(s) (see [MFG] 4.2.1).
An epimorphism X .L. Y in C is called a strict epimorphism if for any
other object Z of C
Im(f* : Homc(Y, Z) -t Homc(X, Z)) = {g E Homc(X, Z)lprg = p;g},

where PJ : X x y X -t X are the two projections. If this is the case, we
Pl,P2 f
call the following sequence exact: X Xy X ::::t X -'-+ Y. The notion of
strict epimorphisms is stronger than that of epimorphisms. For example,
i : Spec(Q) -t Spec(7!..) is an epimorphism in SCH (because it is dominant).
Since Ql ®z Ql = Ql, if i were a strict epimorphism, the empty set Homalg(Ql, Z)

has to be identical to the singleton Homalg(Ql, Ql). This is a contradiction, and
hence i is not a strict epimorphism. A faithfully fiat morphism X -+ Y of
finite type is a strict epimorphism (see [ECH] Theorem I.2.17) in the category
of schemes. In the category SETS, the notion of strict epimorphisms and that
of epimorphisms coincide and are equivalent to surjectivity.
Let G C Autc(X) be a finite subgroup for an object X of C. Write
Homc(X, T)G for the set of morphisms X .4 T such that 1 0 9 = 1 for
all 9 E G. If the covariant functor T -+ Homc(X, T)G is representable
by an object Y of C with a unique morphism Jr : X -+ Y inducing
Jr* : Homc(Y, T) ~ Homc(X, T)G, we call Y the categorical quotient of X
by G and the morphism Jr : X -+ Y the quotient morphism (see [GME] 8.2).
Here is an example of C with fibered products and fibered sums. We start
with a profinite compact group II. A typical example for II is the Galois group
Gal(K/k) of a (possibly infinite) Galois extension K/k. We denote C(II) for
the category of finite sets on which II acts continuously (from the left). Thus
for each object X in C(II), the stabilizer in II of each point x E X is an open
subgroup of II. For X, Y in C(II), a map 1 : X -+ Y is a morphism of C(II)
if I(Jr(x)) = Jr(f(x)) for all x E X and Jr E II. The collection of all such
maps is written as Hom17(X, Y). Let 0 be the set of a single point on which
II acts trivially. Then Hom17(X, 0) is made of one map taking everything in
X to O. In other words, 0 is the final object. The empty set 0 is considered
to be an object in C(II), which gives an initial object in C(II). Since C(II) is
a subcategory of SETS, the fibered product X xs Y for X .4 Sand Y -'4 S
is given by X xs Y = {(x, y) E X x Yll(x) = g(y)} with two projections Px
and py induced by two projections X x Y onto X and Y. The fibered sum
of S .4 X and S -'4 Y is given by the set-theoretic fibered sum X EfJs Y
already described, and in particular, the direct sum is the fibered sum for
an initial object S and is Xu Y with two obvious inclusions ix and iy. If
G C Autc(17) (X) is a finite group, the set-theoretic quotient X / G gives the
categorical quotient in C(II). If X .4 Y is a morphism in C(II), it factors as
X -» Im(f) y Y. Here p : X -» Im(f) is a surjection by definition and hence
is a strict epimorphism, and i : Im(f) y Y is a monomorphism inducing a
direct sum decomposition Y = (Y - Im(f)) U Im(f). Thus we have verified
the following three properties for C = C(II).
(C1 ) C has a final object and a fibered product of any two morphisms
X -+ Sand Y -+ S in Cover S.
(C2) C has an initial object and a direct sum of any two objects in C.
For each object X of C and each finite subgroup G of Autc(X), the
categorical quotient X/G exists.
I !' I"
(C3) Each morphism X --'--7 Y in C factors as X -'--+ X' -'---t Y so that l' is
a strict epimorphism and I" is a monomorphism.
A category e satisfying the three conditions (Cl-3) is called a Galois category.

A covariant functor F : e --+ e' is called left exact if it takes monomorphisms of
e to monomorphisms of e' and F(X Xs Y) = F(X) XF(S) F(Y). We note that
F(X) xF(S) F(Y) = {(x,y) E F(X) x F(Y)iF(f)(x) = F(g)(y)} for X l-t S
and Y 4 S if e' = SETS. Similarly a covariant functor F : e --+ e' is called
right exact if it takes epimorphisms of e to epimorphisms of e' and commutes
with fibered sums. A left exact covariant functor F : e --+ e' is called exact if it
preserves monomorphisms, epimorphisms, and exact sequences. The inclusion
functor F : e(II) '--+ SETS obviously satisfies the following three conditions:
(F1) F is left exact and has values in the full subcategory of finite sets in
SETS;
(F2) F(X EB Y) = F(X) U F(Y) for objects X, Y in e; F(f) is a surjection
if F : X --+ Y is a strict epimorphism; and if X 2t Y is the categorical
quotient in e by a finite group G c Aute(X) , F(X) F(7r\ F(Y) is
a categorical quotient in SETS by the same group G sitting inside
Aut(F(X)), which is just a set-theoretic quotient;
(F3) a morphism X l-t Y in e is an isomorphism if F(f) : F(X) --+ F(Y)
is an isomorphism.
For a Galois category e, a covariant functor F : e --+ SETS satisfying the
three conditions (Fl-3) is called a fundamental functor.
We quote the following theorem of Grothendieck from [SGA] V.4.
Theorem 4.38 If a pair (e, F) of a category e and a covariant functor F :
e --+ SETS satisfies the six conditions (Cl-3) and (Fl-3), there exists a
unique profinite group II (up to isomorphisms) such that e is equivalent to
the category e(II).
The group II is called the fundamental group of (e, F) and sometimes written
as 'lTl (e, F). We refer to [SGA] V.4 for the proof. We call an object X in e
connected if it is not a direct sum of two non initial objects. Here we list steps
of the proof and give a brief description of each step.
1. A morphism f E Home(X, Y) is a monomorphism if and only if F(f) is
an injection.
In fact px : X x y X --+ X is an isomorphism if and only if f : X --+ Y is
a monomorphism. Indeed, if h, g E Home (Z, X) satisfy fog = f 0 h = 1',
regarding Z as an object in elY by 1', we have g Xy hE Hom(Z, X Xy X).
Thus f being a monomorphism is equivalent to the fact that the two
functors Z f--+ Home( Z, X) and Z f--+ Home (Z, X x y X) are isomorphic.
This implies by the uniqueness of the object representing a functor (see
[GME] Lemma 4.1) the above statement that X Xy X ~ X. Then by
(C1), (F1), and (F3) combined, we conclude the assertion. 0
2. A sequence Xl A X 2 A X3 -+ .,. -+ X J -+ X of monomorphisms fJ

for an object X in C stabilizes after the finite steps; that is, for sufficiently
large j, fJ are all isomorphisms.
This assertion is obvious in the category of finite sets. Since F preserves
monomorphisms and brings C into the category of finite sets, it follows by
(F3) from the result for the category of finite sets. 0
3. There is a connected partially ordered set I and a projective system
{P,,1f'J: Pi -+ PJ (i > j)}iEl in C such that
a) for each object X in C we have a unique morphism <Pi : P, -+ X for
each sufficiently large i E I compatible with the projective system;
b) there exist a universal element 'P = ~ ,'Pi E ~ 'El F(Pi ) and an
isomorphism LX : F(X) ~ lli!?iE1Homc(P" X) induced by LXU) =
~ ,F( <Pi)( 'Pi).
We hereafter write Hom(P,X) for lli!?iE1Homc(Pi,X), and consider P =
~iElPi in the procategory of C. The object P is called the fundamental
pro-object of C with respect to the fundamental functor F.
4. P, can be chosen so that
a) Pi for all i is connected;
b) P, is not an initial object for all i;
c) 1ftJ is an epimorphism for all i > j.
For an axiomatic proof of (3) and (4) which is based on the assertions (1)
and (2) besides the axioms (Cl-3) and (Fl-3), see [FGA] Expose 195.
We describe what is going on by examples. Suppose that C is the oppo-
site category of the category of separable finite-dimensional semi-simple
commutative algebras over a given field k. In other words, C is made of
Spec(K) for separable semi-simple extensions Kjk with dimk K < 00. Let
fl (resp. fl sep ) be an algebraic (resp. separable) closure of k (Spec(flse p)
is a fundamental pro-object). Then F : Spec(K) H Homk(K1fl sep ) =
Homc(Spec(flSe P ) , Spec(K)) gives the covariant functor. One dan easily
check that (C, F) is a Galois category. In this case, we may take I to be
the set of all finite Galois extensions in the fixed algebraically closed field
fl containing k. The order of I is induced by the inclusion of two fields.
If C is the category of finite etale coverings of an irreducible variety S
of finite type over a field k, we fix a geometric point S E S(fl) and take
F(X) = HomSCHIS(S, X). We often write this category as ET/ s . In this
case, I is the category of all etale irreducible Galois coverings with ordering
X -+ Y on I given by S-morphisms. For each X E I, its function field is a
finite separable extension of the function field of S. Since the function field
determines X over S (because epimorphisms in the category of schemes
are dominant morphisms), the category I is actually a set.
5. If X is not an initial object and Y is connected, any morphism f :X --t Y

in C is a strict epimorphism.
In the above example of the two categories (of semi-simple separable finite-
dimensional commutative algebras over k and of etale coverings over S),
this assertion is obvious. However this is just a consequence of the axioms
(C3) and (F2-3). An axiomatic proof is in [SGA] V.4.e.
6. The following conditions are equivalent.
a) The natural injection: Homc(P" P,) --t Hom(P, P,) ~ F(P,) is sur-
jective;
b) Aut(P,) acts on F(P,) transitively;
c) Aut(P,) acts on F(P,) simply transitively.
We call P, of Galois type if one of the above equivalent properties is
satisfied by P,.
This is again obvious for our two examples of C, because End(P,) is the
monoid of field endomorphisms of a finite extension over a given base field
k. To prove this axiomatically, we apply the functor F and prove the fact
in the image of F by using the previous assertions (1), (4) and (5).
7. Every object X in C is covered by an object P, of Galois type so that the
unique map P --t X in 3.a) factors through P,.
8. Let Ig al be the subset of I made of all indices i with P, of Galois type.
Then we have Hom(P, P) = Aut(P) = ~ 'EIgal Autc(P,), which is set-
theoretically in bijection with F(P) = ~,F(P,). We put II = Aut(P).
9. Define a functor G : C(II) --t C so that
Homc(G(Z),?) ~ Homn(Z, Hom(P, ?))
as functors with variable X in C.

For each connected X in C(II), the group II acts transitively on X by its
connectedness; so, X ~ II / H for a stabilizer of a point x EX. Since H
acts faithfully on P, for sufficiently large j in Ig al , if i > j, 1T,] induces
an isomorphism P,/ H ~ p] / H of the categorical quotients, which gives
P/H (independently of i > j). We put G(X) = P/H. Here the existence
of the categorical quotients P,/ H in C is guaranteed by (C2). If X is
not connected, we can write X uniquely as X = Uz Z for connected
components Z. Then we define G(X) = EBz G(Z) which gives the inverse
G of F. See [SGA] V.4.h for more details of the argument.
In our example of Galois extensions, one can check that Spec(K)/H =
Spec(KH) gives the categorical quotient of Spec(K) by a finite group
He Gal(K/k). For the category of etale coverings of a base scheme S, the
existence of categorical quotients is explained in [GME] Proposition 8.4.
10. The two functors F and G give the equivalence of categories.

In our two examples, this point is plain. 0
We now quote two more propositions from [SGA] V Propositions 5.5-6.

Proposition 4.39 Let {Q,hEi be a pro-object in C = C(ll) (i.e., a projective
system of objects in C). Suppose that Q, is not an initial object for any i and
that Q, is connected for all i. Define a covariant functor G : C -+ 5 ET 5 by
G (X) = Home( Q, X). Then Q is isomorphic to II j H for a closed subgroup
H of ll, and the functor G is isomorphic to the functor E H EH for the fixed
point subset EH of E under H.
The subgroup H is given by the stabilizer of one point q E Q. Since Q, is
connected, the action of II is transitive, and Q ~ II j H via 7rq +-+ 7r H. 0
Proposition 4.40 Let (C, F, P) be as above (a Galois category, a fundamen-
tal functor, and the fundamental pro-object with respect to F). Let pi =
~ J EJ p; be a pro-object in the procategory of C, and define F' : C -+ 5 ET 5
by F'(X) = Hom(pl,X) = ~JEJHomc(p;,X). Then the following three
conditions are equivalent.
1. pi ~ P and also F' ~ F;
2. pi is fundamental and also F' is fundamental;
3. F' brings direct sums to direct sums, and F(X) =I- 0 if X is not an initial
object.
We may identify C with C(ll). For C(ll), the fundamental pro-object P is
isomorphic to II as a profinite set with continuous ll-action. Thus if pi is
another fundamental pro-object, we find the stabilizer of a single point p of
pi has to be trivial. Since pi is connected, the action of II is transitive;
so, 7r H 7rp gives an isomorphism II ~ pi ~ ll'. This basically shows the
equivalence as above. 0
Remark 4.41 If C is the category of etale coverings of a connected variety

S of finite type over a field k and F(X) = Fa(X) = HomSCH/ s (8, X) for
a geometric point 8 of 5(D), we write the fundamental group for (C, F) as
7rl (5,8). If one changes a point 8 to another geometric point I, FI is still a
fundamental functor (see [SGA] V.7); so, we have an isomorphism (5 : FI ~ Fa
which induces an isomorphism again denoted by (5 of the fundamental pro-
objects PI ~ Pa associated with FI and Fa, respectively. Then 7rl (5, I) ~
7rl (5,8) by T H (5T(5-1. We call an isomorphism (5 : FI ~ Fa a path from 8 to
I.
Remark 4.42 Suppose that 5 is reduced and irreducible. If we choose t to
be the generic point ry of 5, a connected etale Galois covering X -+ 5 gives an
irreducible Galois covering of 5; so, it is determined by its function field k(X)
as a Galois extension of k(ry) = k(5). Thus 7rl(5,'fj) is canonically isomorphic
to Gal(k(P)jk(5)) for a fundamental pro-object P in ET/ s .
4.4.2 Algebraic Fundamental Groups
Let X be a locally Noetherian connected scheme. We fix a geometric point x in

X(Sl). Then the category ET/ x of etale faithfully flat coverings of X and the
fundamental functor Fx: ET/ x -t SETS given by Fx(Y) = RomscH/ x (x, Y)
supply us with a Galois category and give the profinite group 7r1 (X, x) so that
Fx gives the identification of ET/ x with C(7r1 (X, x)). The group 7r1 (X, x) is
called the algebraic fundamental group of X with base point X. We briefly
expose functorial properties of the algebraic fundamental group.
When X is a smooth variety over IC, we can think of the topological funda-
mental group 7riOP (X, x) for x E X(q defined as the homotopy classes of paths
starting from x ending up at x. Then taking the universal cover p : U -t X,
any etale covering Y of X is canonically isomorphic to U j r for a subgroup
r = 7r1(Y, y) (for a point y E y(q over x) of finite index in 7riOP (X, x). Then
Fx(Y) ~ 7ri OP (x,x)jr and thus we get
(4.96)
where r runs over all normal subgroups of finite index in 7ri OP (X, x); in other
words, the algebraic fundamental group 7r1(X,X) is the profinite completion
of the topological fundamental group 7ri OP (X, x). If we pick another point
x' E X(IC) and choose a path IJ from x to x', we find 7riOP (X, x) ~ 7ri OP (X, x')
by T f-t IJTIJ-I, and IJ induces a unique isomorphism of functors Fx ~ Fx"
If we have a morphism f : X -t Y, the pullback 1*T = X Xy T for
an etale covering T/ y is an etale covering of X, because a pullback of a
faithful flat morphism is faithfully flat (see [ECR] Corollary I.3.6). If X and
Yare connected, obviously we have Fx 0 1* ~ FfCx) for any geometric point
x E X(Sl), because an etale covering of a connected scheme X is determined
by its fiber at any given geometric fiber ([ECR] I.3.12 and [SGA] I.5.5). Then
1* is an exact functor from ET/ y into ET/ x .
Suppose now that X and Yare connected locally Noetherian. Then
Fx 0 1* gives a covariant functor from ET/ y -t SETS. By definition,
7r1(X,X) = Aut(Fx) , and each IJ E 7r1(X,X) induces an automorphism of
Fx 0 1* ~ FfCx). Thus Aut(Fx 0 1*) = Aut(Ff(x)) = 7r1 (Y, f(x)), and we
have a homomorphism of profinite groups, 7r(f) : 7r1 (X, x) -t 7r1 (Y, f(x)). In
other words, the association (X, x) f-t 7r1 (X, x) is a covariant functor from the
category of pointed schemes into the category of profinite groups.
Using this, we compare fundamental groups of the fibers of X 1-t
Y and
those of X and Y. We always suppose that X and Yare connected locally
Noetherian and that f is proper, surjective and separable with f*(Ox) = Oy
(i.e., each geometric fiber of f is connected; see [ALG] III.11.3 and [SGA]
X.2). A scheme over a field k is called separable if for any field extension Kjk,
its base change to K is a reduced scheme (no nontrivial nilpotents in the
structure sheaf). A morphism X 1-t Y is called separable if f is flat and the
fiber Xy = X Q9y Spec(k(y)) is separable for all points y with residue field
key) (see [SGA] X.I).
We choose a point y in the image of f in Y and a geometric point 'fj over y
so that its field keY) is an algebraic closure of key). Take a point x in X with
f(x) = y and a geometric point x over x so that k(x) is an algebraic closure
of k(x) and f(x) = 'fj. Then we take a geometric point a E Xy = X Q9k(y) key)·
We writei: X y "-+ X for the inclusion morphism. By the functoriality of 7r1,
we have the following sequence,
(4.97)
which is induced by the sequence of functors ET/ y ~ ET/ x ~ ET/ x y •

Theorem 4.43 Suppose that X and Yare connected locally Noetherian and
that f is proper, surjective, and separable with f* (Ox) = Oy. Then the se-
quence (4.97) is an exact sequence of profinite groups.
Here is a sketch of the proof (see [SGA] X.1.3 for more details).
Proof. If Y' !4 Y is a connected etale finite covering, then f*Y' / X is a finite
etale covering and has to be connected because f* (Ox) = Oy. This shows
that Fx(f*Y) = Fy(Y), and hence f* induces a surjection of the universal
pro-object of ET/ x onto that of ET/ y . Since the universal pro-object is in
bijection with the fundamental group, we find that 7r1 (X, x) ~ 7r1 (Y, y) is a
surjection.
Since the fiber g-l (y) of Y' over'fj is a finite set of closed points, i* 0 f* (Y')
is a disjoint union of X y indexed by points of g-l (y). Since X y is the final
object of ET/ Xy ' we find that Im(i*) c Ker(f*).
To prove the reverse inclusion, Im(i*) ~ Ker(f*) , we need to show that
for any finite etale Galois covering X' ~ X, we have an etale Galois covering
Y' /Y so that X' ~ Y' Xy X if and only if X~/ Xy has a section (i.e., a
disjoint union of copies of Xy). The only nontrivial point is the "if" part.
By the properness of f and the finiteness of h, A = (f 0 h)*(Ox') is a sheaf
of Oy-algebras finite over Oy ([EGA] II1.3.2.I). Define Y' = Spec(A) (cf.
[GME] 5.4). By the separability of f, Y' /Y is an etale covering (see [EGA]
III.7.8.10 and [SGA] X.2). Then the assertion is plain. 0
By Hensel's lemma, if A is a Henselian local ring with residue field K" the
category of finite etale algebras over K, is equivalent to that over A by taking
fibers over K, (see [ECH] 1.4.2 and 1.4.4). We have the following generalization
of this fact to proper smooth schemes by A. Grothendieck and M. Artin:
Theorem 4.44 Let A be a henselian local ring and X/A be a smooth proper
scheme with special fiber X o/",. Then we have 7r1 (X, x) ~ 7r1 (Xo, xo) canoni-
cally, where Xo is a geometric point in the special fiber, and x is the geometric
point of X whose schematic closure contains the image of xo.
See [ECH] Remark 1.4.6 for an indication of references containing the proof.
4.4.3 Group-Theoretic Results
We gather here group-theoretic results we have used in the proof of the global
reciprocity law. Here we use the notation introduced in the proof of Theo-
rem 4.14. In particular, E is a proper subset of the set of rational primes.
In the proof, we have used the following facts for G 1 = ResF/IQJSL(2) and
PG = ResF/IQJ(PGL(2)).
(8) For open compact subgroups S,08 c G1(A,(17oo)), if a : S ~ IYS is an
isomorphism of groups, replacing S by an open subgroup, a is induced by
g(a) E Aut (Fl 17oo ) jA,(17oo)) ~ PG(A,(17oo)): IYs = g(a)sg(a)-l.
We may modify a by 9 E Aut(Fl17OO ) jA,(17oo)) ~ PG(A,(17oo)) so that its pro-

jection aF to Aut (Fl 17CXJ ) jA,(17CXJ)) is the identity map. Then this assertion (8)
follows from the following three assertions for a with aF = l.
(81) For open subgroups Se and lYSe of G1(Qle) (for every prime fI), an iso-
morphism ae : Se ~ lYSe, after replacing Se by an open subgroup of Se,
is induced by the conjugation s H ge(a)sge(a)-l for ge(a) E G(Qe).
(82) For a sufficiently large rational prime fI, we have
Aut(G1(Ze)) = Aut(FeJQle) ~ PG(Ze)

and Aut(G1(Qle)) = Aut(FeJQle) ~ PG(Qe),
where PG is the quotient of G modulo its center.
(83) For sufficiently large distinct rational primes p and fI, any group homo-
morphism ¢: G1(Zp) --+ G1(Ze) is trivial (i.e., Ker(¢) = G1(Zp)).
The assertion (81) follows directly from a result of Lazard on fI-adic Lie groups
(see [GAN] IV.3.2), because automorphisms of the Lie algebra of S (hence of
IY S) are all inner up to automorphisms of the field (in the cases we need). In
the case of classical groups (including SL(2)), this can be proven easily (see
the argument after Proposition 4.47).
8ince the assertions (82-3) can be proven for more general linear algebraic
groups G 1 smooth over the localization Z(pe) C QI at two primes p and fI, we
introduce some terminology in order to prove them for more general groups
including SL(2). For a semi-simple group Glover a field k, we take a maximal
split torus TSP C G 1 defined over k (so TSP is an algebraic subgroup of G 1
maximal among those isomorphic to G~), and take a maximal torus T con-
taining TSP. The group G 1 is split if TSP = T and is quasi-split if the derived
group of the centralizer Z(TSP) of TSP in G 1 is trivial (so, Z(TSP) = T; see
[Ttl). The only thing we need to prove (81-3) is the fact that the structure
of quasi-split groups is determined by its root system and a small amount
of additional data (see [Ttl); particularly, we need the fact that the group is
generated (integrally) by unipotent elements.
For a finite set of rational primes E', we write Z(17') C QI for the localization
outside E'; so, Z(17') = Z [~Ip E E']' We suppose that G 1 / 7L (E') is a (smooth)
semi-simple classical group; so, G 1/1QJ and G 1/ lFp for all p 9' E' are semi-simple.
Since the conditions (S2-3) concern only large primes, we may assume
(QS) G 1 / Zp is smooth quasi-split for all primes p outside E' of rational
primes (i.e., all fibers: G 1/ lFp and G 1/lQJp are quasi-split for all p 9' E').
We now prove the assertion (S3). Let ¢ : G 1(Zp) --+ G 1(Ze) be a homomor-
phism. Since G 1(Zp) is quasi split, G 1(Zp) is generated by unipotent elements
(see Lemma 4.46), and its unipotent radical U is generated by an additive
subgroup Ua corresponding to a simple root D'.
If G 1 = SL(n)/z, for example, we may assume that Ua is made of
diag[l J , (6 Y), 1n - J - 2 l with u E Zp for an index j (with 1 :s: j :s: n).
In general, U and Ua are p-profinite. We consider the normalizer N(Ua )
and the centralizer Z(Ua ) of Ua in G 1 . Then by conjugation, N(Ua)jZ(Ua )
acts on Ua . Since ¢ is a group homomorphism, the quotient N(Ua)jZ(Ua )
keeps acting on the image ¢(Ua ) in G 1(Ze) through conjugation by elements in
¢(N(Ua )). If p oJ e, every element of ¢(Ua ) is semi-simple (because unipotent
radical of G 1 (Ze) is e-profinite). Thus the centralizer (resp. the normalizer) of
¢(Ue) is given by Z(Ze) (resp. N(Ze)) for a reductive subgroup Z (resp. N)
of G 1 . Then N(Ze)jZ(Ze) is a finite subgroup of the Weyl group WI of G 1 /([
which is independent of e.
For example, if G 1 = SL(n)/z, and if ¢(Ua ) is made of diagonal matri-
ces diag[(11ml' (21m2' ... , (r 1m r l for generically distinct (J, Z is given by the
subgroup
SL(n) n (GL(ml) x GL(m2) x ... x GL(m r )),
where GL(md x GL(m2) x ... x GL(m r ) is embedded in GL(n) diagonally.
The quotient N j Z in this case is isomorphic to the subgroup of permutation
matrices preserving Z.
If ¢(Ua ) is nontrivial, the image of N(Ua)jZ(Ua ) in Aut(¢(Ua )) grows at
least on the order of pas p grows. In the above example of G 1 = SL(n)/z,
if ¢(Ua ) ~ Zjpmz for m > 0, all elements in Aut(¢(Ua )) ~ (Zjpmz)x come
from N(Ua)jZ(Ua ) ~ Z;. This is impossible if p » IWll. Thus ¢(Ua ) = l.
Since G 1 (Zp) for large enough p is generated by U a for all simple roots D', ¢
has to be trivial for p large enough. 0
Now we prepare several lemmas and propositions in order to prove (Sl-2)
for SL(2). Let K be a number field, and consider V = K2. We consider the
A-linear endomorphism algebra EndA(VA). Let Z(A) be the center of the K A -
linear endomorphism algebra EndKA(VA) C EndA(VA). Thus KA ~ Z(A) by
sending f E KA to the scalar multiplication by f.
Lemma 4.45 The centralizer in EndA(VA) of UA = {( 6~) Ib E K A }
n
and an
element (;j E M2(K) with 1 oJ r E K X is given by Z(A).
Proof. By a regular representation p : KA y EndA(KA), we can embed i :

End KA (VA) y EndA(VA). Identifying EndA(VA) = M 2d (A) and End KA (VA) =
M 2 (KA ), we have ~ (~~) = (~i~j ;i~~)· Let N(A) = {(g 8) Ib E A} for a ring
A. We consider the centralizer
ZI = {u E EndA(VA)lzn = nz for all n E N(A)}.
If u = (~ ~) E ZI, by computation, we find p(n)d = a· p(n) and c· p(n) = 0

for all n E KA; so, by taking n = 1, we have a = d and c = O. Since we can
choose p(n) with d distinct eigenvalues, we find that a E p(KA). Thus we find
ZI = { (p~a) pta)) la E KA and bE EndA(KA)} .
We then consider Z = {z E Zllz (p\;y) n n

= (P~,) z for 1 =I- 'Y E KX}. Then
by computation, we find p('Y)b = b; so, b = O. This shows that Z is the center
Z(A) of End KA (VA). 0
Lemma 4.46 Let K be afield. Then SL 2(K) and PSL 2(K) are generated by
unipotent elements. If IKI > 2, the commutator [GL 2(K), GL 2(K)] is equal
to SL 2(K), and [PGL 2(K), PGL 2(K)] = PSL 2(K).
Proof. Let H be a subgroup generated by all unipotent elements in SL 2(K)

and H be its image in PSL 2 (K). If u is a nontrivial unipotent element, we find
an eigenvector v =I- 0 of V = K2 fixed by u. Taking a base (w, v) of V over K,
we find u(w, v) = (aw+bv, v) = (g ~). Taking the determinant, we find a = 1;
so, u is a conjugate of (5 ~). This shows that gHg- I = H. Thus H is a normal
subgroup of PSL2(K). Since the upper triangular unipotent group acts by
translation on the affine space A I (K) = K, it acts transitively on A I (K).
The opposite unipotent subgroup tu acts transitively on pI (K) - {O} ~
AI(K), and H acts transitively on pl(K). Thus for any g E PSL 2(K), we
find h E H such that h(oo) = g(oo); so, h-Ig E B, where B is the subgroup
of upper triangular matrices. Thus if T c H for the diagonal torus T, we find
T· H = PSL 2(K).
Since C~~) = (5 X~l) (~-"b-l) (5 x~l), we have j(x) = (~-"b-l) E H.
Since j( -x)j(l) = (X~l ~), we find that T c H. Thus SL 2(K) and PSL 2(K)
are always generated by unipotent elements.
Write [g, h] = ghg-Ih- I . Then we have [( 0~), (5 n] = (~(a-l~-I)X).
Thus if IKI > 2, we can choose a, bE K X with a =I- b. Thus [GL 2(K), GL 2(K)]
contains all unipotent elements; so, [GL 2(K), GL 2(K)] = SL 2 (K). By the
same argument, we find [PGL 2(K), PGL 2(K)] = PSL 2(K). 0
Fix a continuous section s : KX -+ GL 2 (K) of the determinant map sat-

isfying x = det(s(x)). Let Autdet(GL2(K)) be the automorphism group of
the group GL 2 (K) preserving the section s up to inner and field automor-
phisms, that is, 0' E Autdet(GL2(K)) satisfies a-(s(a)) = h . T(s(a))h-I for
a field automorphism T E Aut(K) and h E GL2(K). More generally, for a
subgroup H containing SL 2(K), we define Autdet(H) for the automorphism
group of H preserving the section s of det : H -+ K X up to field and inner

automorphisms of K. The inner automorphisms of GL 2 (K) give rise to an
inclusion PGL 2(K) c Autdet(GL2(K)). A field automorphism 0" E Aut(K)
can be considered to be a group automorphism of GL 2 (K) by its action entry
by entry on matrices. Here is (S2) for SL2(K).
Proposition 4.47 Let K be a non- A rchimedean local field or a global field
(i.e., a number field). Then the automorphism groups of the groups SL 2(K),
PSL 2(K), and PGL 2(K) are given by Aut(K) ~PGL2(K). If H is a subgroup
of GL2(K) containing SL 2(K), Autdet(H) is given by
{(r,g) E Aut(K) ~ PGL 2(K)lr(det(H)) = det(H)}.
Proof. We first study Aut(PGL 2(K)). We have an exact sequence,
where i(x)(g) = xgx~l. We write B (resp. U) for the upper triangular Borel
subgroup (resp. the upper triangular unipotent subgroup) of G L2 (K). Their
image in PGL 2(K) is denoted by Band U. Let U' be a subgroup of GLn(K)
isomorphic to the additive group K; so, we have a : K '?" U'. Consider the
image a(l) of 1 E K in U'. Replacing K by a finite extension containing an
eigenvalue 0' of a(l), let Ve, C V be the eigenspace of a(l) with eigenvalue 0'.
Then a(~) acts on Va: and a(~)m = a(l) = 0' E End(Va:). Thus we have an
algebra homomorphism K[x]j(x m -0') -+ EndK(Va:) for all 0 < m E Z. If K is
a p-adic local field, nm nm
(KX )m! = {I}. By using this, we find (Kx)m = {I}
for a global field K. Thus if K is a non-Archimedean local field or a global
field, we find that 0' has to be l. Thus U' is made up of commuting unipotent
elements; so, by conjugation, we can bring U' onto U.
Since O"(U) for 0" E Aut(PGL 2(K)) is isomorphic to K in PGL 2(K),
O"(U) is again a unipotent subgroup of PGL 2(K). Since B is the normalizer
of U, again O"(B) is the normalizer of O"(U); so, O"(B) is a Borel subgroup.
We find g E GL 2(K) such that dB) = gBg~l. Thus we may assume that
0" fixes B. Applying the same argument to U, we may assume that 0" fixes
U. Let 0" (6 i) = (6 ~). Then changing 0" by i (a:~' ?) 00", we may assume
that 0" (6 i) = (6 i ). Then 0" induces an automorphism (f of T = B jU. Fix
an isomorphism t : K X '?" T and u : K '?" U. Since 0" sends the center to
the center, we find t(a)u(b)t(a)~l = u(ab). Then applying 0" to the above
formula, we get (f(a)O"(b) = dab) for a E K X and b E K. Taking b = 1,
we find (f(a) = O"(a). Thus O"(ab) = O"(a)O"(b), and 0" : K -+ K is a field
automorphism. Modifying 0" by an element of Aut(K), we may assume that 0"
induces the identity map on B. Thus 0" induces an automorphism of pl(K) =
G L 2 (K) j B fixing CXJ and commuting with all affine transformations z f-7 az+b
of pl(K) for a E K X and b E K. Then 0" is the identity map on pl(K) and,
hence, on GL2(K). We find Out(PGL 2(K)) 2" Aut(K) and Aut(PGL 2(K)) =
Aut(K) ~ PGL2(K).
We apply the same argument to Aut(PSL 2(K)). Let Bl = B n PSL 2(K)

and 7\ = Tn PSL 2(K). We find 9 E PSL 2(K) such that a-(Bd = gB1g- 1.
Thus we may assume that (J leaves Bland U stable. Further modifying (J by
conjugation of an element in GL 2(K), we may assume that a-(u(l)) = u(l).
Note that in this case, t(a)u(b)t(a)-l = u(a 2b). Then we have 0'(a 2)(J(b) =
a-(a 2 b); so, a-(a 2 ) = 0'(a 2 ) = 0'(a)2. Thus O'(a) = ±(J(a). Thus if K is a field of
characteristic 2, we find that (J = 0' is a field automorphism of K.
Suppose that K has a characteristic different from 2. This implies
(J(a 2) = 0'(a 2) = 0'(a)2 = (±a-(a))2 = (J(a)2,
(J((a + b)2) = a-(a 2 + 2ab + b2) = (J(a)2 + 2(J(ab) + (J(b)2 and a-((a + b)2) =
O'(a + b)2 = (J(a + b)2 = a-(a)2 + 2(J(a)(J(b) + (J(b)2. Thus if K has char-
acteristic different from 2, we find (J(ab) = (J(a)(J(b); so, (J E Aut(K).
Thus for a character E : K Xj(KX)2 -+ {±1}, we have O'(x) = E(xk(x);
so, a-(t(a)) = t(O'(a)) = ±tk(a)) = t((J(a)) in PSL 2(K). This shows
Aut(PSL 2 (K)) = Aut(K) ~ PGL 2(K) = Aut(PGL 2(K)).
We now study Aut(SL 2(K)). Since physically SL 2(K) = PSL 2 (K) for a
field of characteristic 2, we may assume that K has a characteristic different
from 2. We have an exact sequence
1 -+ {±1} -+ SL 2(K) -+ PSL2(K) -+ l.
Since Aut(K) ~PGL2(K) c Aut(SL 2(K)), each automorphism (J of PSL 2(K)

lifts to an automorphism of SL 2 (K), which in turn induces an automor-
phism of {±1}. Since Aut({±l}) = {I}, we find that Aut(SL 2(K)) =
Aut(PSL 2(K)) = Aut(K) ~ PGL2(K).
Now we look at Autdet(H). Since (J E Autdet(H) preserves the section s
up to field and inner automorphisms, modifying (J by such an automorphism,
we may assume that (J fixes s( K X). Then (J is determined by its restriction
to SL 2(K) cHand, hence, comes from and element in Aut(K) ~ PGL 2(K)
preserving H. This shows the last assertion. 0
Here is a direct proof of (Sl) for SL 2 (K) for a finite extension K of Qp,
although the proof in [GAN] is a highly polished and generalized vers.ion
of this argument. The exponential exp(X) = Ln>o -;,~ and the logarithm
10g(1 + X) = Ln>l (-1 )n+l ~n converge p-adically on a sufficiently small
neighborhood of 0 of M 2 (K) and 1 of GL 2 (K), respectively. If (J : S ~ S' is
an isomorphism of open compact subgroups of SL 2 (K), by shrinking Sand
S', we may assume that the log induces an isomorphism log: S ~ 10g(S)
and S' ~ 10g(S'). Then (J induces an isomorphism of the Lie algebra (J :
10g(S) ~ 10g(S'), which extends to an automorphism of the Lie algebra.s[2(K)
of SL 2 (K) by scalar extension. Since (J takes the maximal abelian Lie algebra
N(K) ~ K of .s[2(K) into its conjugate, changing (J by (J composed with a
conjugation by an element of GL 2(K), we may assume that (J takes N(K)
onto N(K). Identifying N(K) with K by sending each matrix in N(K) to its
upper-right corner element, as in the proof of the above proposition, we have

CJ(ab) = CJ(a)CJ(b) for a, b E K. Thus by Lemma 4.45 (or its proof), we may
assume that CJ E End(s[2(K)) is K-linear. Then by the Lie algebra version of
Proposition 4.47, it is a conjugation by an element of GL2(K). This shows
(Sl) for SL 2 . This argument proving (Sl) can be generalized to quasi-split
semi-simple classical groups more general than SL(2).
Here is (S2) for SL2(0).
Proposition 4.48 Suppose that K is a finite extension of Qlp with integer
ring O. Then the automorphism groups of the groups PSL 2(0), SL 2(0), and
PGL 2(0)) are given by Aut(O) I>< PGL 2(0). If H is a subgroup of GL 2(0)
containing SL 2 (0), Autdet(H) is given by
{(T,g) E Aut(K) I>< PGL 2(0) Hdet(H)) = det(H)}.
Proof. Let K, be the residue field of O. We study Aut(PGL 2(0)). We have

Aut(K) I>< PGL 2(0) c Aut(SL2(0)). Let U(O) = Un SL2(0). For the max-
imal ideal m C 0, we define r(mn) = {a E SL 2(0)la == 1 mod mn}. Since
Pl(O) = PI(K), all Borel subgroups of SL 2 (0) are conjugates of each other.
Since BI(O) = SL 2(0) n B is a semi-direct product of TI(O) and U(O), all
unipotent subgroups are conjugates of each other. By the same argument in
the case of the field, we may assume that CJ E Aut(PGL 2(0)) leaves U(O)
stable. We have t(a)u(b)t(a)-I = u(ab). Then applying CJ to the above for-
mula, we get J(a)CJ(b) = CJ(ab) for a E K X and b E K. If CJ(l) E m, then
J(a)CJ(1) = CJ(ab) Em. However, {CJ(a)la E O} = 0; we find 0 c m, a contra-
diction. Thus CJ(l) = a is a unit. Then t(a) E GL 2(0), and modifying i(a),
we may assume that CJ(l) = 1. Proceeding in exactly the same way in the case
of the field, we find that Aut(PGL 2(0)) = Aut(O) I>< PGL2(0). From this,
again we obtain all the desired assertions. 0
Our argument proving Propositions 4.47 and 4.48 can be easily generalized
to a semi-simple algebra K of finite dimension over Ql or Qlp. We only state
the result leaving the proof to attentive readers.
Proposition 4.49 Assume that K is a semi-simple commutative algebra of
finite dimension over Ql or Qlp (or the direct sum of such). Then the auto-
morphism groups of the groups PGL 2(K), PSL 2(K), and SL 2(K) are given
by Aut(K) I>< PGL2(K). If K is a semi-simple Qlp-algebra, we again have
Aut(PGL 2(0)) = Aut(PSL 2(0)) = Aut(SL 2(0)) = Aut(O) I>< PGL 2(0),
where 0 is the integral closure of Zp in K.

The group Autdet(H) for a subgroup of GL2(K) and GL 2(0) containing SL(2)
has the same expression as in Propositions 4.47 and 4.48.
A similar argument proves (S2) for symplectic and unitary groups, for
whose proof we refer to [H03bj Section 3. Let hn = (I~ -In), I2n = (IOn \:J),
and 12n + 1 = ( 00 0I In)

0 . We take a finite extension field K of <Qlp and its
In 0 0
semis imp Ie quadratic extension M; so, M is either a field extension or isomor-
phic to K EB K. We write R (resp. 0) for the integral closure of Zp in M (resp.
K). Write c for the generator of the ring automorphism group Aut(M/ K).
Define G(A) for A = K and 0 by one of the following groups,
G(A) = {a E GLn(A00 R)ialn' ta c = v(a)In} '

(4.98)
G(A) = {a E GL2n(A)iahn' ta = v(a)hn} '
where v : G(A) -+ A x is a group homomorphism (called the similitude

map). We call G / K a quasi-split unitary group if it is defined for M / K
and In. We write G = GU(In) in this case. The group defined for hn is
called the symplectic group. We write G = GSp(2n) in this case. We have
the determinant map det : G(K) -+ M X for unitary groups G and write
7r for det x v : G(K) -+ M X x KX. We then define PG by G modulo its
center and write G I for Ker(7r) or Ker(v) accordingly as G = GU(In) or
G = GSp(2n). Fixing a section s of? = 7r or v, we define Aut?(G(A)) simi-
larly to Autdet(GL2(A)). We assume that p > 2 is unramified in M/<Qlp.
Theorem 4.50 Under the above notation and assumption, we have
Aut?(G(A)) = Aut(PG(A)) = Aut(GI(A))

= {Aut(K) ~ PG(A) if G = GSp(2n),
Aut(M) ~ PG(A) if G = GU(In),
where? = v or 7r accordingly as G is symplectic or unitary.
As a corollary, we have
Corollary 4.51 If H is a subgroup of G(O) containing GI(O), Autv(H) is
given by
{(T,g) E Aut(K) ~ PG(O)hv(H)).= v(H)} ,
and Aut 1r (H) is given by
{(T,g) E Aut(M) ~ PG(O)h7r(H)) = 7r(H)}.

5
Generalized Eichler-Shimura Map
Let G/z(p) for Z(p) = Qnzp be a connected reductive group split over Zp (see
Section 4.4.3 for the definition of split algebraic groups). We first prove semi-
simplicity of the commutative Hecke algebra acting on the nearly ordinary co-
homology group H~.ord(X(S), L) inside the interior cohomology H!q(X(S), L)
for a modular variety X(S) associated with an arbitrary p-power level open
compact subgroup S of G(A 00), where H? is the image of the compactly sup-
ported cohomology group inside the standard cohomology group. Here the
locally constant or coherent sheaf L of Q-vector spaces on X(S) is associ-
ated with a rational representation of G twisted by a finite-order character.
After dealing with topological cohomology groups, we relate as Hecke mod-
ules the topological and the coherent cohomology groups via the generalized
Eichler-Shimura map (which shows the semi-simplicity of the Hecke algebra
acting on the coherent cohomology and topological cohomology). Although
we have assumed that G is split over Zp, the argument works equally well for
a connected reductive G smooth quasi-split over Zp-
We write B (resp. N) for a fixed Borel subgroup of G (resp. the unipotent
radical N of B). We have the torus To = BIN. In this chapter, the character
U is often used to denote a unitary group.
5.1 Semi-Simplicity of Heeke Algebras
We describe the semi-simplicity of the Hecke algebra acting on topological

cohomology groups. Since in this book we do not touch upon representation
theory of p-adic reductive groups, purely representation-theoretic results are
quoted from appropriate sources. A main point is to prove, assuming repre-
sentation theoretic results, that the nearly p-ordinary part of the cohomology
(where certain Hecke operators at the primes p dividing the level S act by
p-integral automorphisms) is semi-simple as Hecke modules. Thus the part
of the cohomology of level S (where the group G(Qp) acts through nearly

226 .5 Generalized Eichler-Shimura Map
p-ordinary representations) is semi-simple as Hecke modules (over an appro-

priate Hecke algebra).
5.1.1 Jacquet Modules
We summarize here known results On Jacquet modules from representation

theory. For a p-adic Lie group 9 and a (right) representation V with coeffi-
cients in a field K of characteristic 0, we call V smooth if for each v E V,
we find an open subgroup S c 9 that leaves v invariant. If furthermore,
HO(S, V) is finite-dimensional for all open subgroups S, V is called admissi-
ble. Let 7r be an admissible semi-simple representation of G(Qp) On a vector
space V = V(-7r) over a field K of characteristic O. Contrary to the tradition,
we always suppose that V is a right G(Qp)-module.
We have a Haar measure du of N(Qp) with fN(71 p ) du = 1. We define
V(B) = V(B,7r) = {v - v7r(n) E V(7r)lv E V(7r), n E N(Qp)} as a Tc(Qp)-
module, and put VB = VB(7r) = V/V(B), which is called the Jacquet
module. By this definition, V f--t VB is a right exact functor. For each
W = v - v7r(n) E V(B), we take a sufficiently large open compact subgroup
Uw C N(Qp) containing n. Then we see that fu v7r(u)du = 0 for every open
subgroup U with Uw cUe N(Qp). If fu v7r(u)du = 0 for every sufficiently
large open subgroup U of N(Qp), for the stabilizer U' of v in U, we find
which is in V(B). Thus V(B) is the collection of v with fu v7r(u)du = 0 for

every sufficiently large open subgroup U of N(Qp). By this fact, the functor
V f--t VB is left exact, and we conclude that the association is an exact functor.
Later we give a canonical splitting VN ~ VB E9 V(B)N as Hecke modules (due
to Bernstein and Casselman), where V N = HO(N(Zp), V).
For a smooth representation V' of Tc(Qp) (regarding it as a B-module via
the projection B --» T c), a smooth induction from B of V'is defined by
Ind~(V') = {J: G(Qp) -+ V'lf: smooth, f(gb) = g(g)b Vb E B(Qp)}, (5.1)
on which we let G(Qp) act by f(x)g = f(gx). Here the word "smooth" meanS
that for each f E Ind~ V', we find an open compact subgroup S such that
f(kx) = f(x) for all k E S. Thus Ind~ V'is automatically smooth.
Since the smooth induction preserves admissibility ([BeZ] 2.3), V =
Ind~ V' has composition series {O} C VI C V2 C ... c V stable under
the action of G, and hence its semi-simplification (Ind~ V')SS = EBJ Vy+1/Vy
is well-defined as an admissible G(Qp)-module. The beauty of the theory of
admissible representations is its purely algebraic nature; so, we do not need to
assume any analytic assumptions; in particular, our representations are often
5.1 Semi-Simplicity of Heeke Algebras 227
not unitary as is clear from our main result (Theorem 5.3) in this section.
Hereafter all representations of G, B, and Tc are supposed to be admissible.
The following results are due to Jacquet and Bernstein-Zelevinsky and are
well known (see [BeZ]),
1. (Frobenius reciprocity; [BeZ] 1.9) HomB(VB, V') ~ Homc(V, Ind~ V');
2. If 1[" is absolutely irreducible, then dimK VB :s; IWI, where W is the Weyl
group of Tc in G (Bernstein-Zelevinsky; [BeZ] 2.9);
3. If 1[" is absolutely irreducible and VB =f. 0, then we have a surjective linear
map Ind~ A -t> V of G(Qp)-modules for a character A : Tc(Qp) --+ K X
(Jacquet; [BeZ] 2.4-5);
4. (Ind~ >:) ss ~ (Ind~ AW) ss for w E W (Bernstein-Zelevinsky; [BeZ] 2.9),
where ss indicates "semi-simplification", AW (t) = A( wtw- 1), and>: = 5if2 A

for the module character 5B of B: fN(Qp) ¢(u)du = 5B (b) fN(Qp) ¢(b-1xb)du
(for all ¢). We have the following corollary of the above facts, which is a key
to our semi-simplicity result.
Proposition 5.1 Suppose that 1[" is irreducible and that VB [>:] =f. 0, where>: =
5if2 A for the module character 5B on B. Then 1[" is a quotient of Ind~ig:i i
If Aw(t) for wE Ware all distinct, VB C EB wE ® AW as Tc(Qp)-modules.
°
We say a representation 1[" is regularly induced (at p) if for each irreducible
component V of 1[";..!!e have a nontrivial >:-eigenspace VB [>:] =f. for a character
A of Tc(Qp) and Aw(t) (w E W) are all distinct.
Proof. Since the algebra in EndK(VB ) generated by the action of Tc is a

finite-dimensional c<2..mmutative algebra, the >:-eigenspace is nontrivial if and
only if the maximal A-quotient is nontrivial. Thus, we have a morphism of Tc-
modules, VB -t> V(>:). Since we have (Ind~>:)SS ~ (Ind~Aw)SS, by Frobenius
reciprocity, all eigenvalues AW can show up as a quotient of (Ind~ >:)B whose
dimension is bounded by IWI. Thus if all characters AWare distinct, we have
c- rv
VB c (Ind B A)B = EB wE ® A . Since V f-c> VB is exact, this is enough to
-;;;
conclude the assertion. D
5.1.2 Double Coset Algebras
We consider the double coset algebra made of formal linear combinations of

double co sets of a subgroup in a semi-group. This type of algebra is con-
sidered in [IAT] 3.1 and often called a Hecke ring. We use the terminology
"double coset algebra" to avoid confusion with Hecke algebras (which is the
image of the double coset algebra in the endomorphism algebra of a space of
automorphic forms).
Let D = {x E Tc(Qp)ixN(Zp)x-1 :::) N(Zp)} which is called the expand-
ing semi-group in Tc(Qp). Since xN(Zp) :::) N(Zp)x (x ED), iJ. N = N . D . N
228 5 Generalized Eichler-Shimura Map
for N = N(Zp) is also a multiplicative semi-group. When G = GL(n)/zp' it

is easy to see that D is generated by T c (Zp) and diag [pe 1 , pe 2 , ••• , pen] for
sequences of integers el ::; e2 ::; ... ::; en. Write B = B(Zp) and N = N(Zp)
for simplicity. Define the so-called Iwahori subgroups by
So(r) = {u E G(Zp)lu mod pT E B(ZjpTZ)} , (5.2)

Sl(r) = {u E G(Zp)lu mod pT E N(ZjpTZ)}.
These subgroups S have the Iwahori decomposition S = N'T' N ~ N' x T' x

N for open compact subgroups T' c Tc(Zp) and N' in the opposite unipotent
W = W(Zp). Each xED shrinks W: xWx- l C W. From this, .1s = S·D·S
is again a multiplicative subsemi-group of G(Qp) (this statement includes
.1B = B· D· B = .1so(oo))' We call .1s the expanding semi-group with respect
to (S, N). When G = GL(2), .1s for S = So(r) (r > 0) is almost equal to
the p-part of the semi-group .1o(pT91) introduced in Section 4.2.5 (strictly
speaking, we have .1o(pT91)p = .1s modulo center).
By the Iwahori decomposition, we have, for X = N(Zp), B(Zp), and an
Iwahori subgroup S,
X~X= u X~u= u
uEN\f.Nf.-l
Xu~. (5.3)
By this fact, the double coset algebras generated additively over Z by double
cosets of X in .1x have the following homomorphic relations as algebras:
via N~N H B~B H S~S for ~ E .1N' These algebras are commutative:
T(~)T(TJ) = for T(~) = N~N and ~,TJ E D. See [IAT] Chapter 3 and
T(~TJ)
[MFG] 3.1.6 for this type of results dealing with GL(2). Here we give a proof
of these facts for G = GL(n). The same argument works well for classical
Chevalley groups quasi-split over Zp replacing aj below by the image of p
under simple coroot a positive with respect to D (i.e., a(p) ED).
Lemma 5.2 Let the notation and the assumption be as above. Assume that
G = G L( n) /zp' The algebras R(S, .1s) for Iwahori subgroups S are com-
mutative, and if S :::) B(Zp), they are all isomorphic to the polynomial ring
Z[h, ... , tn, t;;-l] (with tn inverted) fort J = SajS with a J = diag[ln-J,p1J]' If
S:::) N(Zp), we have R(S, .1s) ~ Z[TS][tl,"" tn, t;;-l] for the quotient group
Ts = Tc(Zp)SjS ofTc(Zp), where Z[Ts] is the group algebra ofTs .
Proof. For simplicity, we write B for B(Zp). For ~ E D, we consider the double
coset B~B. Decompose B = U1)E.S'(f.)(~-l B~ n B)TJ. Multiplying by ~-l B~
from the left, we get ~-l B~B = U1)E.S'(f.) ~-l B~TJ {:} B~B = U1)E.S'(f.) B~TJ. If
~ = diag[pa 1 , ••. ,pan], we have
5.1 Semi-Simplicity of Hecke Algebras 229
We may choose the subset 3(0 inside N(Zp) so that
3(0"3 (T/2J) r-+ (T/2j mod paJ-a'Zp)J>i E EBZjpaJ-a,z

J>i
is a bijection. Then we have B~B = U1)E3(';) B~T/ and a formula of the index:
(B : ~-1 B~ n B) = p[';] with [~] = LJ>i(aj - a2). Writing deg(B~B) for the
number of left cosets of B in B~B, we find deg(B~(B) = deg(B~B) deg(B(B),
because [~(l = [~] + [(] for ~,( E D. Since B~B(B => B~(B, if we can
show that deg(B~B(B) = deg(B~(B), we get B~B(B = B~(B and (B~B) .
(B(B) = B~(B in the double coset ring R(B, 41 B ), which in particular shows
the commutativity of R(B, 41B). To see deg(B~B(B) = deg(B~(B), we note
B~B(B = U1)E3(O U1)'E3() B~T/(T/'. This implies
deg(B~B(B) ~ deg(B~B) deg(B(B) = deg(B~(B),
and hence we get the identity deg(B~B(B) = deg(B~(B). Since the ajs give
independent generators of D jTc(Zp), the monoid algebra Z[D jTc(Zp)] is
isomorphic to a polynomial ring with n variables Z[a1' ... ,an, a;;-l] with an
inverted (because an E Z(Qip) is invertible in Z(Qip) C D). The association
a J r-+ T( a J ) = Baj B therefore induces a surjective algebra homomorphism
Z[DjTc(Zp)] ---+ R(B,41 B ), which can be easily seen to be an isomorphism.
Replacing DjTc(Zp) by Dj(Tc(Zp) n S) = Ts X (DjTc(Zp)) in the above
argument, the same proof works well for any S with So(r) => S => Sl(r) and
yields R(S, 41s) ~ Z[Ts][DjTc(Zp)] ~ Z[TS][t1, ... , tn, t;;-l], where Z[Ts] is
embedded into R(S, 41s) by sending t E Ts to StS. 0
We let R act on v E V N = HO(N(Zp), V) by
vIT(O = vl[N~N] = L
uE.;-lN.;\N
V7f(~u) = 1
';N.;-l
V7f(u)7f(Odu, (5.4)
and similarly for v E VB in place of v E VN; then the projection: VN ---+ VB

is R-linear. To see the last identity of (5.4), it is sufficient to recall that we
have normalized the measure du so that JN du = 1.
Let E be the set of maximal (proper) parabolic subgroups P => B. Decom-
posing P = MpNp for a reductive part Mp => Tc (called a Levi factor; see
[RAG] ILl.S) and the unipotent radical N p, we can identify the set E with
the following set of cocharacters,
{ap : G m ---+ Glap(p) generates Zp(Qip) n D modulo Zp(Zp)Z(Qip)} ,
where Zp (resp. Z) is the center of Mp (resp. G). Then {~'" = a(p)}",a

generates DjTc(Zp)Z(Qip), and R ~ Z[Tc(Zp)Z(Qip)][T(~",)]"" where as be-
fore Z[Tc(Zp)Z(Qip)] is the group algebra of the discrete group Tc(Zp)Z(Qip).
Similarly we have RB ~ Z[~i~;?][T(~",)]",. If G = GL(n),
230 5 Generalized Eichler Shimura Map
and the cocharacter (XJ corresponds to the two-block parabolic subgroup P =

{(g:i)} with j x j-matrices a (and (n - j) x (n - j)-matrices d).
For e = Il,EE ~Q' we have N(Qp) = U~oeJ Ne- J • Thus writing T(~) =
[N~N] for ~ E D as an operator on V N = HO(N, V), we see easily from
(5.4) that T(e) = T(e)1 and for each finite-dimensional subspace Xc V(B),
T(e)lx is nilpotent on XN by (5.4).
For any 'R.-eigenvector v E V N with vt = >.(t)v (t E Tc(Qp), v = v
mod V(B)), we get
vl[NxN] = [N : x- 1 Nx]>.(x)v = Idet(AdN(x))lp>'(x)v, (5.5)
where I Ip is the standard absolute value of Qp such that Ipl;;l p and

Ad = Ad N is the adjoint representation of Tc on the Lie algebra of N.
Let us prove a canonical splitting VN = VB EB V(B)N as 'R.-modules. We
have by definition, VN = VN(Zp) = Ur v S l(r). The subspace Vr = v S l(r) is
finite-dimensional and stable under 'R.. By the Jordan decomposition applied
to T(e), we can decompose uniquely that Vr = Vro EB vml so that T(e) is
an automorphism on Vro and is nilpotent on V m1 . We may replace T(e) by
T(e a) = T(e)a for any positive a in the definition of the above splitting. Since
T(e) is nilpotent over any finite-dimensional subspace of V(B), Vro injects into
VB; so, dim Vro is bounded by dim VB ::::: 12171. For any Tc-eigenvector v E VB,
lift it to v E V. Then for sufficiently large j, V1r(e- J ) is in VN. Since V1f(e- J )
is a constant multiple of v, we may replace v and v by V1f(e- J ) and V1f(e- J ),
respectively. Then for sufficiently large k, w = vT(e k ) E Vro, and wT(e)-k
is equal to v for the image w in VB. This finishes the proof when the action
of Tc on VB is semi-simple. In general, take a sufficiently large r so that Vr
surjects down to VB. We apply the above argument to the semi-simplification
of Vr under the action of the Hecke algebra. Thus VO = Ur Vro ~ VB, and this
finishes the proof of V N = VB EB V(B)N(Zp) as'R.-modules.
5.1.3 Rational Representations of G
We now fix a finite extension Kover Qp. Let W be the p-adic integer ring
of K. We write S = So(r) (r > 0). Recall the Iwahori decomposition S =
N'Tc(7l,p)N. We consider the space C(R) = {¢: SjN(7l,p) ---+ R} made up of
continuous functions for R = Wand K. Here the continuity is with respect
to the p-adic topology of the source and the target. We would like to make
C a left L1;;l-module for the opposite semi-group L1;;1 of L1p = L1s. This
L1;;l-module C can be considered as the p-adic continuous induction of the
right regular representation of T c (7l,p) on the space of p-adically continuous
functions on Tc(7l,p) with values in R (thus, this is the third induction process
we encounter in this section). For that, we first define a left action of L1p on
Ys = SjN(7l,p). Since S acts on Ys = SjN(7l,p) from the left, we only need to
define a left action of D. Pick yN(Zp) E Y s and by the Iwahori decomposition,

we may assume that y E NITe(Zp) C S and consider yN. Then for d E D,
dyNd~l = dyd~ldNd~l C dyd~lN(rfJp) and dyd~lN(rfJp) is well defined in
G(rfJp)/N(rfJp). Since conjugation by d E D expands N(Zp) and shrinks N I,
we have dyd~ 1 E S, and the coset dyd~ 1N (rfJp) n S = dyd~ 1S is a well-defined
single coset of N, which we define to be the image of the action of d E D.
We now let .1;1 act on C by d¢(y) = ¢(d~ly). In this way, C becomes a
.1;l-module. Note that this action factors through D /Z(rfJp); so, if the group
G has an involution d r--+ d L with dd L E Z(rfJp) , d¢(y) = ¢(dLy) = ¢(d~ly).
We consider the algebrogeometric induction module:
Le(K;K) = {¢ E HO(G/N,Oe/N)I¢(yt) = K(t)¢(y) 'Vt E Te}, (5.6)
where Oe/N is the structure sheaf of the scheme G/N, and we regard its
global section ¢ as a function on G(rfJp)/N(rfJp) with values in K. We let G
act on L(K; K) = Lc(K; K) by g¢(y) = ¢(g~ly). Then L(K; K) = ind~ K~l
(following the normalization of induction as in [RAG] 1.3.3), which is the
induction in the category of scheme theoretic representations (i.e., rational
representations). We call K dominant if L(K; K) I- O. We write a matrix form
of this representation as PI< = pC: : G r--+ GL(L(K; K)).
All dominant weights in X(Te) ~ zr form a cone
if G is semi-simple (see [RAG] 11.6.6). Here the generators w J are called

fundamental weights. Take a weight W of Te. Since the derived group G 1
of G is semi-simple, its restriction to T = Te n G 1 can be written as
wiT = m1w1 + ... + mrWr for fundamental weights wJ of G 1. The weight
W is called regular if m J > 0 for all j, and we sometimes call a weight W
sufficiently regular if m J » 0 for all j (without specifying how large m J is).
We restrict functions in L(K; K) to Ys = S/N(Zp) and regard L(K; K) C
C(K). Multiply L(K;K) by a character E: Tc(Z/prz) = SO(r)/Sl(r) r--+ W X
(regarding it as a function in C(W)). Since rfJ; = Z;
x pZ, we can decompose
Te(rfJp) = Tc(Zp) x (pZr for the rank r ofTe, and we can extend E to Te(rfJp)
requiring it to be trivial on (pzr. In this way, we get the twisted .1;l-module
L(KE; K) = EL(K; K) c C. The action preserves L(KE; W) = L(KE; K) n C(W)
but the original PI< may not be. For ~ E D, the action of ~~1 E .1;1 is given by
K(~)~lE(Opl«C1). For a W-module R, we define L(KE; R) = L(KE; W) ®w R,
which is a .1;l-module under the modified action.
We recall that the p-component of the semi-group .1o(plJ1) introduced in
Section 4.2.5 is a subsemi-group of the semi-group .1p. The definitions of
the Hecke operator in (4.89) and (5.8) below are compatible, because we let
5 E .1o(plJ1)' act on L(KE; K) by x r--+ 5x in the definition of the Hecke operator
(4.89) and 5¢(y) = ¢(5 Ly) = ¢(5~ly) for the main involution & of GL(2)
as already remarked. This modification (in order to preserve p--integrality)
does not show up for coherent sheaves for G = SL(2), because Lc(K; K)
is one-dimensional due to the fact that C = T for G = GL(2). A similar
remark applies to Lc(K; K) when G = G 1 = ResF/Q1SL(2). However, in the
Hilbert modular case of G = ResF/Q1GL(2), the induction of K* = (K1' K2) E
X(Tc) for the diagonal torus Tc of G is isomorphic to the tensor product
det"'! 0L(K2 - K1 - I; K) for the restriction K2 - K1 - I of K to T = Tc n G 1
(G 1 = ResF/Q1SL(2)). The p-integral action created through this process is
the action on LC(K2 - K1 - I; K) removing the "det" factor.
5.1.4 Nearly p-Ordinary Representations

We now study global cohomological automorphic representations. Let G(Z) c
G(A(oo)) denote a maximal compact subgroup hyperspecial everywhere (by
abusing notation; see [Ttl] for hyperspecial maximal compact subgroups). We
assume that the p-component of G(Z) is given by G(Zp). For split groups over
Q (like GL(n) and GSp(2n)), they are just maximal compact subgroups. We
take G(Z) = GLn(Z) if G = GL(n) and G(Z) = GSP29(Z) if G = GSp(2g).
Hereafter we assume that K is dominant in X (Tc) and write c for a char-
acter of TC(ZjpTZ). Let S be an open subgroup of G(Z). We consider the
associated modular variety X(S) = G(Q)\G(A)j(S· Co) ~ Ur X r , where r
runs over finitely many discrete subgroups G(Q) n t(S . G(JR)+)C 1 in G(JR)+
for a coset decomposition G(A) = Ut G(Q)t(S . G(lR)). For the symmetric
space X = G(JR)jCo, we have written Xr for r\x. For any W-module R,
we then define a right action of u E SCo on L(Kc;R) by ¢Iu = pI<E(U;;l)¢
(assuming that Sp C So(r)).
We define the covering space X(S) of X(S) by
X(S) = G(Q)\(G(A) X L(Kc; R))j SCo ~ UXr , (5.7)

r
where I(X, ¢)u = (rxu, ¢Iu) for I E G(Q) and u E SCo, and we form the
quotient space Xr = r\(X x L(Kc; R)) by the diagonal action. We use the
same symbol L(Kc; R) for the sheaf of locally constant sections of X(S) over
X(S). Writing [~] for the action of CIon L(Kc; R) for ~ E Ll;;l x G(A(p)),
we define as in (4.89) (and (4.37))
[S~S] = Trs/(s<ns) 0 [~] 0 Ress/<sns. (5.8)

The choice of the schematic induction action as the action of ~;1 is standard
in the theory of automorphic representations (with coefficients in a character-
istic 0 field). We write T(O for the Hecke operator [S~S] under this choice of
the action. When we study the p-adic Hecke operators, we choose the modified
p-integral action. We write Tp(O for the operator [S~S] under this p-adically
integral action of ~ ELls. Note that the factor I det(g)IA in (4.37) is missing in
this definition of topological Hecke operators. Actually this factor (which co-
incides with I det(AdN(~))lp when ~ E Llo(poosn) and G = GL(2)) is absorbed
by the Eichler-Shimura map as we show later in (5.40).
We consider the limit, shrinking S,
(5.9)
Here H,q(X(S),L(lic;R)) for R = W,K is the image of the compactly sup-

ported ~ohomology Hg(X(S), L(lic; R)) in Hq(X(S), L(lic; K)). The group
H,q(X(S), L(lic; R)) is called the interior cohomology of X(S) with coeffi-
cients in L(lic; R).
On the space .c(K), the group G(A (0) acts from the right via a smooth
representation, which is completely reducible. In particular, we have an action
on HO(S, o (lic; K)) = O(lic; K)S of the double coset algebra
R(S, G(APOO) x Ll p) ~ R(S(p), G(APOO)) 0 RB,
where S = Sp x S(p) and we have assumed that Sp = So(r).

For ~ E D, T(~) acts on O(lic; W)N (N = N(7l,p)) through the Ll;l_
module structure of L(lic; W). We write this operator as Tp(~). On the other
hand, T(~) acts on .cq(lic; K)N through the action of G(A (0). The correspond-
ing operator is written by the same symbol T(~). Then the two operators Tp(~)
and T(~) are related on the image of .cq(lic; W)N by
(5.10)
This follows from the fact that the normalized action of ~-1 E Ll;l is given by
1i(~)-lc(~)p",(~)-l, while as an element of G(A(oo»), ~-1 acts via c(~)p",(~)-l.
Suppose that G admits Shimura variety Shs of PEL type defined over
its reflex field E (see Chapter 7 for generality of Shimura varieties), whose
complex points give X(S). When Ii = 0 (the identity character), the action
of the Hecke operator is motivic and is (canonically) induced by the Hecke
correspondence T(~) c Shs x Shs, and in this case, Tp(~) = T(~) is the
action of the correspondence on the Betti cohomology of X(S). If Ii > 0,
we may relate cohomology groups of the sheaf L(lic; K) as a part of the
cohomology group with constant coefficients of a certain self-product Z of
copies of the universal abelian scheme over X(S). The Hecke operator then
has an interpretation as an isogeny action on the universal abelian scheme,
and it can be regarded as the action induced by the Hecke correspondence
in Z x Z. The action of Tp(~) and T(~) uses different action of Ll;l. This
action of Ll;l determines the part of the cohomology group over Z identified
with the cohomology group over X(S) with locally constant (but nonconstant)
coefficients. Thus the motivic realization of the two operators T(~) and Tp(~)
could be actually different, and the operator Tp(~) may not even have any
motivic realization (as in the Hilbert modular case of nonparallel weight). For
example, in Scholl's construction [Sc] of the Grothendieck motive associated
with an elliptic Hecke eigenform J, if one changes the action of congruence
subgroup r of SL 2 (7l,) by a power of determinant character, the physical sheaf
over Xr(C) obtained is the same, but its rational structure (including the
Galois action) is different. In the elliptic modular case (treated in [Sc]), we can
construct the motive associated with the standard p-adic Galois representation
PI of f and its Tate twists PI(m) as the etale realization of motives directly
realized over a self-product of the universal elliptic curve.
For a Hilbert-Hecke eigenform f, we could twist PI locally at each p-adic
place by a power of the p-adic cyclotomic character, but this twist may not
extend to a global twist because the exponent of the cyclotomic character
depends on the p-adic place. In particular, if f is of nonparallel weight, the
process of defining Tp(~) corresponds to untwisting PI to reach a p-ordinary
Galois representation at each p-adic place p, which cannot be performed glob-
ally; so, the operator Tp(~) may not have a motivic interpretation.
Unfortunately, for a Hilbert-Heeke eigenform f for F =I- Q, the motive of
f realized from the Hilbert modular variety does not produce a rank 2 motive
whose etale realization gives rise to the two-dimensional Galois representation
of f (see [BlRJ, [TaJ, and also [MFG] 5.6.1 for a description of such Galois
representations). Therefore to obtain such a rank 2 motive, we need to re-
sort to some tricky construction either moving to Shimura curves treated in
Section 4.3 (see [H81]) or extracting a rank 2 motive from a rank 4 motive
(which can be constructed from some Shimura surfaces; see [Bl]). Despite such
an effort, this problem is not completely solved yet.
We take ~ E D such that N(Qp) = UJ ~J N(Zp)~-J. The product e =
rL,EE ~n (or its power ILEE ~~a with en > 0) has this property. As seen at
the beginning of Section 3.3, the p-adic (or profinite) limit e = limn-+oo Tp(e)n!
exists as an operator on Hq(X(S), L(K,E; R)) for R = Wand K (and is in-
dependent of the choice of e), because Hq(X(S), L(K,E; R)) is an R-module
of finite type. Thus the limit e exists as an endomorphism of .cq(K,E; R)N for
R = Wand K. It is easy to see, if Sp :J N,
(5.11)
We write .c~.01"iK,E; R) for e (0 (K,E; R)N). An irreducible representation 7r of

G(Aoo)), which is a sub quotient of .cq(K,E; K), is called nearly p-ordinary of
p-type K, if eV(7r)N =I- 0 for the representation space V(7r) of 7r.
5.1.5 Semi-Simplicity of Interior Cohomology Groups
Let 7r ~ 0e7re be a cohomological automorphic representation of p-type K,.

Suppose that its p-component 7rp (acting on V(7rp )) is a sub quotient ofInd~ >:
(this is automatic if 7r is nearly p-ordinary). Then we find a character >.
Tc(Qp) --+ K X with the above property such that VB [>:] =I- 0 and
The equality holds if and only if 7r is nearly p-ordinary (in this case, automat-
ically VB =I- 0 and Ind~ >: --» 7rp because V N ~ VB EB V(B)N as R-modules).
.5.1 Semi-Simplicity of Hecke Algebras 235
For the moment, suppose that G(Qp) = GLn(Qp) and factor the character
A as A(diag[tl, ... , t n ]) = rr=l A,(t,). Define the Hecke polynomial (at p) by
H,,(X) = TI~=l (1- A,(p)X), and write L1N for the Newton polygon of H11:(X)
(e.g., [NAZ]). Define the Hodge polygon L1H of 7r to be the Newton polygon
of TI~l (1 - (r;,cp),(p)X). Then the above inequality implies L1N :::: L1H and
the two extreme ends of the two polygons match.
We return to a general group G and assume that 7r is nearly p-ordinary.
By definition, fN(Qp) ¢(u)du = oB(b) fN(Qp) ¢(b-1xb)du. This shows that
(5.12)
By definition, 2p = det 0 Ad N is a sum of positive roots, and p is a sum of

fundamental weights with respect to E. This shows
(5.13)
Since the dominant r;, is nonnegative with respect to E, r;,p > 0; that is, /'i,p is
in the interior of the Weyl chamber of E. This shows that if wi-I,
(5.14)
because W acts simply transitively on Weyl chambers and each element in

the interior of the chamber of A has the maximum p-adic absolute value in its
conjugates under W. In particular, we get
Theorem 5.3 Let 7r be an irreducible nearly p-ordinary representation of p-
type r;,. Then there exists a character A : Tc(Qp) ---+ K X such that>: y
V(7r p)B y EBWEWAw and IAlp = Ipr;,lp, where p is the sum of fundamental
weights with respect to E and I Ip is the absolute value on K. Moreover,
eHO(N(Zp), V(7r p)) is one-dimensional, on which T(~) = s~s for ~ E D acts
by scalar Ip(~)lpA(~).
Now suppose that S = Sp x G(Z(p)) with Sp = So(r) for r > O. From the
above theorem, we get the following semi-simplicity of the Hecke algebra (for
cohomological nearly p-ordinary cusp forms of p-type r;,) from the fact that
the spherical irreducible representation of G(Qe) has a unique vector fixed
by any (given) hyperspecial maximal compact subgroup (e.g., [BlRl] 1.12),
because of the Iwasawa decomposition of G(Qe).
Corollary 5.4 Let the notation and the assumption be as above. Then the
Hecke module eH,q(X(S), L(r;,c; K)) is semi-simple.
The projector e = ep is defined over .cq(r;,c;Q)N. Thus the above semi-
simplicity remains true on epH,q(X(S), L(r;,c; Q)) for ep = TIeEP ee with a
finite set P of primes (where G is quasi-split over Zp) and an open compact
subgroup S hyperspecial outside P and Iwahori at e E P. For such nearly
P-ordinary automorphic forms, semi-simplicity as in Corollary 5.4 holds.
5.2 Explicit Symmetric Domains

We describe unitary Lie groups UW(lR) associated with Hermitian forms 'IjJ
and their symmetric domain XW in an elementary way. Since each connected
component of XW is isomorphic to '[) = UWCIR)/Cw for a maximal compact
subgroup CW of UW(lR), we describe an explicit form of'[) as a classical Her-
mitian domain, which is useful in describing /1,* from a given /1, E X(T). We
often follow the treatment by Shimura in [EPEl. In this section, Rr;: for a ring
R denotes the set of all m x n matrices with entries in R.
5.2.1 Hermitian Forms over C
Let V be a d-dimensional vector space over C. We suppose having a Hermitian

form cp : V x V ---+ C. Thus we have
(UI) cp(>.v + Ilw, x) = >.cp(v, x) + IlCP(w, x) for v, w, x E V and >',11 E C;
(U2) cp(x,y) = cp(y,x) for X,y E V.
A Hermitian form cp is called positive (resp. negative) semi-definite if cp(x, x) ~
o (resp. cp(x, x) :::; 0) for all x E V. We write cp ~ 0 (resp. cp :::; 0) if cp is
positive (resp. negative) semi-definite. The form cp is called positive definite
if it is positive semi-definite and cp(x, x) = 0 implies x = O. In this case, we
write cp > O. If cp - 'IjJ > 0, we write cp > 'IjJ. This is a partial ordering on the
set of Hermitian forms on V. Similarly we define negative definiteness.
Take a base v = {VI, ... ,Vd} of V, and identify V with the d-dimensional
row vector space Cd by ~v : V = L: XjV J r-+ (Xl, ... , Xd) E Cd. We
have cp(v,w) = L:i,Jx;yJCP(v"vJ) for ~v(w) = (YI, ... ,Yd). Set <P = <Pv:=
(cp(Vi,V J )) E C~. We get cp(v,w) = ~v(v)<P~v(w)*, where X* = tx for a ma-
trix X with coefficients in C. By (U2), we see <P* = <P. Conversely, for a given
<P E C~ with <P* = <P, if we define cp(x, y) = x<Py* for x, Y E Cd, obviously cp is
a Hermitian form. Thus the knowledge of Hermitian forms is equivalent to the
knowledge of Hermitian matrices. We call <P positive definite if cp(x, y) = x<Py*
is positive definite. We write Sd(C) for the set of Hermitian matrices of degree
d and define Si(C) C Sd(C) to be the subset of positive definite Hermitian
forms.
When we change a base {VI, ... , Vd} into {vi, ... , v~}, then we find a
matrix A = (aiJ) E Md(C) such that v~ = L: j azJvj. Thus Av' = v,
where we have written v in the form of a column vector v = t( VI, . . . ,Vd).
Since ~v(v)v = v, we see ~v/(v)Av = ~V/(V)V' = V = ~v(v)v and we get
~v/(v)A = ~v(v). This shows that ~v/(v)A<Pv(~V/(W)A)* = ~v(v)<Pv~v(w)* =
cp( v, w) = ~V' (v )<PVI ~V' (w)*, and A<PvA * = <PV/
For x E C, we write Tr(x) = x + x. We now compute <P choosing a nicer
base. Let N = Rad(V) = {v E Vlcp(v, V) = O}. Obviously N c V is a
subspace. This space is called the null space or radical of V. We have cp(v +
n, w + m) = cp(v, w) + cp(v, m) + cp(n, w) + cp(n, m) = cp(v, w) for n, mEN.
Thus cp is well-defined on the quotient V = V/N, on which Rad(V) = {O}.
5.2 Explicit Symmetric Domains 237
First suppose that Rad(V) = 0, which is equivalent to det <P =I- O. Then we
define the unitary group of rp by
U'P(JR) = {T E EndC(V)lrp(xT,yT) = rp(x,y) \fx,y E V}
~ {T E Md(c)IT<PvT* = <Pv} = {T E Md(c)IT*<P~lT = <p~l}. (5.15)
By definition, U'P(JR) = U>''P(JR) for any 0 =I- A E R

For every x E V, there exists y E V such that rp(x, y) =I- O. Then we have
rp(x + y,x + y) = rp(x, x) + Tr(rp(x,y)) + rp(y,y). If Tr(rp(x,y)) = 0, then
<p(x, y) = -rp(x, y). Replacing y by iy with i = A, we may assume that
Tr(rp(x,y)) =I- O. Thus at least one of {rp(x+y,x+y),rp(x,x),rp(y,y)} is non-
zero. Therefore, we can find Vl with rp(vl,vd =I- 0 and rp(vl,vd E JR by (U2).
Let W = (CV1).l = {v E Vlrp( v, vd = O}. Then W is a subspace of V not
containing Vl; that is, dim W < dim V. For any v E V, w = v- t.p'P«v,v tl ) Vl E W;
Vt,Vl
V = W +CV1. Since dim W < dim V, this shows V = WEB CV1. By induction,
we find a base V2, ... ,Vd of W such that rp( Vi, V)) = 0 if i =I- j and 0 i=- rp(Vi' Vi).
Reordering v" we may assume that rp(Vl' Vl), ... , rp(vm' v m ) are positive and
rp(vm+1' Vm+l), ... , rp(vm+n' vm+n ) (d = m + n) are negative. The numbers
(m, n) are called the signature of rp. Since m is the dimension of maximal
subspace of V on which rp is positive definite, (m, n) depends only on rp.
Changing rp by -rp if necessary (this does not change U'P), we may assume
that m 2': n. Hereafter we always assume this and write m = n + t with t 2': O.
When Rad(V) =I- 0, we take a base v = {Vl' ... ,v)} as above of V = V / N
and lift them to vectors Vl, ... ,Vj of V. Set Vo = CVl + ... + Cv). Then we
have V = N EB Vo. Supplementing v = {Vl, ... ,v)} with a base Vj+l, ... ,Vn
of N, we have <Pv = (4'0" g) E sn(C), where <Py E M) (C) for j = dim V.
Theorem 5.5 We have the following assertions which are equivalent.
(1) Let rp be a Hermitian form on a complex vector space V of dimension d.
Then there exist subspaces V+, V_, and N of V such that
and rp is positive (resp. negative) definite on V+ (resp. V_). Moreover,

we can find bases of V+ and V_ so that V+ = CVl + ... + CV m and
V_ = CVm +1 + ... + Cv m +n , with rp(Vi,Vi) = ±1 and rp(v"Vj) = 0 for
i =I- j.
(2) Let <P be a Hermitian matrix of degree d. Then there exists a matrix X E
GLd(C) such that X<PX* = (~ _OB 8) and A and B are positive definite
°°°
of degrees m and n, respectively. Moreover, we can take X so that A = 1m
and B = In.
Hereafter we assume that rp is nondegenerate (i.e., det <P =I- 0). There is
another realization of a given Hermitian form if it is isotropic. A Hermitian
form is called anisotropic if rp( v, v) = 0 implies v = O. If rp is not anisotropic,
it is called isotropic. A subspace I of V is called totally isotropic if I.l n I =

{x E II<p(x, I) = O} = I. There exists a maximal totally isotropic subspace
of V (because dim V < (0). If <p is isotropic, the maximal totally isotropic
subspace has positive dimension.
Let J be a subspace of V with J c J.l; that is, J is totally isotropic. We
claim that we have a totally isotropic subspace J' such that J' ~ Homc(J, CC)
by <po We prove this by induction on dim J. Let 0 =I VI E J. Then as we have
seen, we always have w E V such that Tr( <p( w, vd) =I O. We consider the
equation for x E JR,
0= <p(w + XVI, w + xvd = <p(w, w) + xTr(<p(w, vd) + <P(Vl' vdx 2

= <p(w, w) + xTr(<p(w, vd). (5.16)
Since Tr( <p( w, vd) =I 0, we can solve this equation and define v~ = w + XVI
for the solution X. If dim J = 1, we set J' = CCv~, which proves the claim.
Suppose now that dim J > 1. We take a subspace J 1 of J so that dim J 1 =
dim J - 1. Since J is totally isotropic, J 1 is totally isotropic. By the induction
assumption, we find J{ so that <p induces the duality between J{ and J 1 .
Then any V E J{ n J 1 is orthogonal to every w E J 1 ; so, J{ n J 1 = O. We put
X = Jf + J 1 , on which <p is nondegenerate. Thus V = X EEl X.l. Let J o be the
projection of J to X.l, which is totally isotropic and of dimension 1. We find
inside X.l a subspace J~ dual to J o under <p, and J' = J{ EEl J~ does the job.
Theorem 5.6 We have the following assertions.
(1) If J is a totally isotropic subspace of V, we can find a totally isotropic
subspace J' of V such that <p induces a CC-antilinear isomorphism i : J' ~
Homc(J, CC). Here "CC-antilinearity" means that i(Av) = >:i(v) for A E cc.
(2) If J is a totally isotropic subspace of V, we can find a base v = {VI, ... , Vd}
of V such that
(a) v, for i <S j span J;
(b) <p(v" Vn-J+k) = i5,k for 0 <S k <S j;
(c) <p(v"vk)=Oifj<k<sd-j andi<Sj ori>d-j.
(3) If J is a totally isotropic subspace of V, we can find a base v = {V1' ... , Vd}
of V such that the V,s for i <S j span J, and 1>v = (1~ t 18 ) .
(4) If 1> is isotropic, then we find an integer j > 0 and X E G L d(CC) such that
X1>X* = ( ~ ~ 10)
1) 0 0
for an anisotropic !Jr.
5.2.2 Symmetric Spaces of Unitary Groups
Let Im,n = (10 _ t ), and set for d = m + n

(5.17)
and 1: = {X E c~IX*Im.nx = (~_OB) A> O,B > o}, which is nonempty

by Theorem 5.5. Then we can let (g, h) E GLm(C) x GLnUC) act 1: on the
right and a E Um,n on the left by X ~ aX (g ~). We now prove
Proposition 5.7 Let V(m,n) = {z E ce~lz*z < In} and X(z) = (~':' 1:)
faT' z E ce~. Then V(m, n) x GLm(C) x GLn(C) ~ 1: by (z, g, h) ~
X(z)(g~).
Proof. Write X E 1: as (~ ~) with a E ce~. Then
X *Im,n X = (a*a-c*ca*b-c*d)
bOa_doc bOb-dod
°)
= (AO-B
for positive definite A and B. Thus a*a > c*c ::" 0, because x*c*cx =
(cx)*(cx) ::" O. Thus if au = 0 for u E cem, then u*(a*a)u = 0 and hence
u = O. This shows the linear map cem -+ cem given by u ~ au is injec-
tive and hence invertible. That is, det a i- O. Similarly, b* b < d* d implies that
det d i- O. Then we define z = bd- 1 . The fact d* (d-*b* -ca- 1 )a = b*a-d*c = 0
tells us z* = (bd- 1 )* = ca- 1 . Thus X = (~':' tn) (o~) = X(z) (o~). Since
0> b*b - d*d = d-*(1- z*z)d, we have z*z < In ¢} 1m > zz* because the
signature of Im,n is (m, n). This shows that z E V(m, n), and the map is sur-
jective. Suppose (~':' tn ) = X(z) = X(z') (0 ~) = C;c z~d). Then c = d = 1
and hence z = z', which shows the injectivity. 0
We now show that X(z) is invertible. It is enough to show that X(z)x = 0

implies x = 0 for x E ce d . Write x = (~) with x E ce m . Then we have
x + zy = 0 and z*x + y = O. Replace y in the first formula by y = -z*x. We
get x - zz*x = (1m - zz*)x = O. Since 1m - zz* < 0, det(lm - zz*) i- 0 and
hence x = O. Then y = -z*x = O.
We now define z -+ a(z) for a E Um,n and z E V(m, n) by
aX(z) = X(a(z)) °
(h(OO,Z) ° )
J(OO,z)
with h(a, z) E GLm(C) and j(a, z) E GLn(C). Since we have
X((aj3)(z))
°
(h(OO{3,z) ° ) = aj3X(z)
J(oo{3,z)
= aX(j3(z)) (h({3,Z)
° ° )
J({3,z)
= X(a(j3(z))) (~h(-OO,-;;{3'-(Z~))h;-;(-;;-{3,--"'Z)
° °
J(oo,{3(z))J({3,z)
)
for a, 13 E Um,n, we conclude
h(aj3, z) = h(a, j3(z))h(j3, z) and j(aj3, z) = j(a, j3(z))j(j3, z), (5.18)

(aj3)(z) = a(j3(z)), (5.19)
because X (z) is invertible. If a = (~ ~), we have aX (z) = (~:~~: ~::~) and
h(a,z)=b·tz+a, j(a,z)=cz+d, a(z) = (az+b)(cz+d)-l. (5.20)
Let 0 be the zero matrix in D(m, n). Then a(O) =0 implies that
aIm,n -- I m,n °
(h(a,O) ° ),
J(a,O)
since X(O) = Im+n. Thus b = c = O. Then alm,na* = Im,n implies a E Um(lR)

and d E Un(lR) for the classical definite unitary group Un(lR) of degree n:
Un(lR) = {x E GLn(<C)iXX* = In}. Thus we have
(5.21)
via a = (g ~) H (a, d) E Um x Un.

For each X E X, writing X* Im,nX = (~ _OB)' we choose u E GLm(<C)
and v E GLn(<C) so that A = u*u and B = v*v (by Theorem 5.5). Then
X ( u~l V~l) E Um,n, and hence Um,n(lR) acts transitively on the quotient
X/(GLm(<C) x GLn(rC)). We have D(m, n) ~ Um,n(lR)/(Um(lR) x Un(lR)) by
Proposition 5.7, and Um,n(lR) acts transitively on D(m, n); that is, for any
z E D(m, n), there exists a E Um,n such that a(O) = z. The symmetric domain
D(m, n) is bounded in C:::" and is called a bounded realization of Um,n/CO.
Corollary 5.8 We have D(m, n) ~ Um,n(lR)/Co via 9 H g(O), where g(z) =
(az+b)(cz+d)-l for 9 = (~~) with a E C~. The maximal compact subgroup
Co is made up of (~ ~) with u E Um (lR) and v E Un (ll~.).
Since X(w)* Im,nX(z) = (~-;;::'~: WZ'-:;;::'l)' replacing z and w by z+Llz and

z, we get
).( )*] (1-9(W)g(z)"

d·lag [th( g,Z,]g,z Llg(z) ) d·lag [h( g,Z,]g,Z
Llg(z)* g(w)*g(Z)-1 ).( )] =
diag[th(g, Z), j(g, Z)*]X(g(z))Im,nX(g(z + Llz)) diag[h(g, z), j(g, z)]

= X(Z)*g* Im,ngX(Z + Llz) = e~~;* W;;~l) .
From this, we conclude, for dz = (dz'J kj,
dg(z) = th(g, z)-ldz. j(g, z)-l. (5.22)
We now exhibit an unbounded realization. We consider
cp= ( ° °-'In)
° =(O<P)
°° °
06
d n 1[1
with rp E C~. Since 'P has signature (m, n) (m 2: n) if () is positive definite, it

is equivalent to Im,n, and hence
U'P(JR) = {x E GLm+n(C) Ix<px* = <p}

= {x E GLm+n(CC)IX*<p~lx = <p~l} ~ Um,n'
The isomorphism is given by U'P 3 x r-+ XxX~l for X E GLm+n(C) with

X <pX* = 1m,n' Let
Write Y E ~ as
Y = ( adebe)
9 h J
f =
(VI fC)
V2
W J
so that a, j E C~. Computing Y*<p~l Y, the n x n matrix at the bottom right

corner is ij*c + j*e~l f - ic*j < O. Thus 0 ::; j*e~l f < i(c*j - j*c). If
cu = 0, then iu*(c*j - j*c)u = 0 and hence u = O. Thus c E GLr(CC) and
similarly j E GLr(C). We put Xl = cr 1 and y' = fj~l. Then Zl = (~;)
satisfies i((x')* - Xl) > (yl)*e~ly'. Put
3= 3(n,e) = {z = (~) Ix E C~, Y E c:~n, R(x* - x) > y*e~ly}.

We have Zl E 3 and (v*,w*)<p~l(~) > 0 for v = (~~) E C;::::. We want to
show that v is invertible. It is enough to show that vu = 0; then u = O. We
have u*(v*, W*)<p~l (~) u = u*(w*lJrv + v*1>w)u = u*w*lJrvu + (vu)*1>wu = 0,
because vu = O. Thus u = O.
Define by s (xo* }!;e) s v.
= Then we have
Let Z = (xo* }!;e ~ ). Then Z*<p~l Z = ( ~ ~OB) with A > 0,

p q lr
B > O. Looking
at the lower left corner, we have ix* - ix*p = 0 and hence p = In because x is
invertible (as we have shown c and j are invertible). Similarly looking at the
middle block of the lowest row, we get iy* - iy* - ix*q = 0 and hence q = O.
Thus Y = B (z) (5 ~) for
(5.23)
for z = (~) E 3.
Proposition 5.9 Let <p be as above. Then ~ ~ 3x GLm(C) x GLn(C) by
B(z)(o~) H (z,a,b).
Proof. The argument before the proposition shows the surjectivity of the map
3 x GLm(C) x GLn(C) --7 ~. Suppose B(z) = B(z') (8 ~). Writing = e
( a ~,e) we have
In 0 '
242 5 Generalized Eichler-Shiroura Map
From this we conclude that d = 1 and Bc = B. Since B is invertible, c = 1.

Thus the map is injective. 0
By the proposition, we define a(z) E 3, 'l9(a, z), f,L(a, z) for z E 3 and a E

U'I'(JR) by the following formula,
aB(z) = B(a(z)) (19(a,Z) 0 ).

o /-L(a,z)
Again, for a,(3 E U'I'(JR), we have, similarly to (5.18),
'l9(a(3, z) = 'l9(a, (3(z))'l9((3, z) and f,L(a(3, z) = f,L(a, (3(z))f,L((3, z), (5.24)

(a(3)(z) = a((3(z)). (5.25)
Corollary 5.10 Choosing T E GLd(C) with TcpT* = Im,n (by Theorem 5.5),
the map T : 3 -t V(m, n) given by TB(z) = X(T(z)) (h(~,Z) J(~'Z») with
h(T, z) E GLm(C) and j(T, z) E GLn(C) is a holomorphic isomorphism.
Writing i = ('k) E 3, we have U'I'(JR)jCj ~ 3 by g H g(i), where C j = {g E
U'I'(JR)lg(i) = i}. Furthermore, if we choose T so that T(i) = 0 (possible by
transitivity), a H TaT- I induces an isomorphism Cj ~ Um(JR) x Un(JR).
Proof. By Propositions 5.7 and 5.9, z H T(z) is a well-defined isomorphism.
By definition, a H TaT- I is a surjective isomorphism of Lie groups from
U'I'(JR) onto Um,n' In particular, we have T(a(z)) = TaT-IT(z) by defini-
tion. If we can prove the transitivity of the action of U'I'(JR) on 3, we get
another proof of the transitivity of the action of Um,n on V(m, n). To show
the transitivity under U'I'(JR), we define ~ : 3 -t S+,(C) and TJ : 3 -t S+.(C) by
B(Z)*cp-1 B(z) = (~(Z) 0 ). (5.26)

o 7)(Z)
By definition, we have
f,L(a, w)*TJ(a(w))f,L(a, w) = TJ(w). (5.27)
We want to find a with a(i) = z for a given z. We assume that the last n rows
of a are (0,0, d) with d E GLn(C). Then aB(w) = B(a(w)) (0 ~) because the
lower right corner of B(w) is In. Thus f,L(a, w) = d. Since B(i)*cp-I B(i) =
°
(6 -'?2)' we have TJ(i) = 2 and d*TJ(z)d = 2. Since TJ(z) > by the definition of
3, we can find a E GLr(C) such that i(x* -x) _y*()-Iy = TJ(z) = 2aa*. Then
for z = t(tx, ty) E 3,
a = p(z) = In,y*O-l x-wa*) (a 0 0)

U'I'(JR) (5.28)
( 0 I",-n
o 0
Y
In
0 I",-n
0 0 a-*
0 E
satisfies a(i) = z, because

aB(i) = (Xa~~_~.2W ~ie ~a:-=':) = B(z) diag [( _"~~ly 1rr~-n) ,a-*].

Thi;:; show;:; tran;:;itivity. The last isomorphism follows from Corollary 5.8. 0
We define
5(z) = Tn det(7](z)). (5.29)
The factor 2- n in the above definition is a normalization factor to make

det(5(i)) = 1. We then have by (5.27) the following transformation formula,
5(a(z)) = 5(z)[ det(p,(a, z))[-2 (for a E U'P(lR)). (5.30)
Let J be a totally isotropic subspace of V. We define
P(lR) = Pj(lR) = {g E U'P(lR)[gJ c J}.
Writing r.p = Tv and taking 1= 'Lr;'=-+::+l CvJ , we see easily that p(z) C Pj (lR).
Thus the above proof of Corollary 5.10 shows
Corollary 5.11 We have U'P(JR) = Pj(lR)C for any maximal totally isotropic
subspace I c V and any maximal compact subgroup C of U'P(lR).
The fact that U'P (lR) = Pj (lR)C holds for any totally isotropic subspace J.
Proof. The assertion follows from Corollary 5.10 for C = C j . Since any other C
is a conjugate of C j , the assertion is clear. This fact follows from the topological
fact that any compact subgroup has a fixed point on U'P(lR)jC j = 3. 0
5.2.3 Invariant Measure
We now compute the measure on 3 invariant under the action of U'P(lR). For
that, we first compute several formulas:
(5.31)
with ~(Zl'Z) = C('j'~;,tX) _~ty) and 7](Zl'Z) = i(xi - x) - y~O-ly. Thus

7](z) = 7](z, z) = i(x* - x) - y*O-l y. We have by computation
B(z) (
00)
,e-1 l
10 = (-''7(Z)
0
y'
-,e y . x)
-1 01 0 0 1
From this we get
det(B(z)) = det( -iO) det( -i7](z)). ( 5.32)
Since we have aB(z) = B(a(z)) (19(~,z) J.1.(~,z)), we get

det(a) det(B(z)) = det(B(a(z)))det('I9(a, z)) det(/L(a, z)). (5.33)

From /L(a, z)*ry(a(z))/L(a, z) = ry(z), we have for a E U'P(IR),
det(a) det(-iB) det( -iry(z))

= det(a) det(B(z)) = det(B(a(z))) det('I9(a, z)) det(/L(a, z))
= det( -iB) det( -iry(a(z))) det('I9(a, z)) det(/L(a, z))
= det( -iB) det( -iry(z) det(/L(a, z))-l det(/L(a, z))-l det('I9(a, z)) det(/L(a, z))
= det( -iB) det( -iry(z)) det(/L(a, z))-l det('I9(a, z)).
This combined with det(a)-l = det(a) shows
det(a) det('I9(a, z)) = det(/L(a, z)). (5.34)
By replacing z and Zl by a(z) and a(zd, we get from
aB(z) = B(a(z)) (19(a,Z) 0 )

o /1-(a,z)
that
t'l9(a, zd(a(z) - a(zl))/L(a, z) = z - Zl, (5.35)
/L(a, zd*ry(a(zd, a(z))/L(a, z) = ry(Zl' z), (5.36)
'I9(a, zl)*~(a(zd, a(z))'I9(a, z) = ~(Zl' z). (5.37)
To compute the Jacobian matrix of z t-+ a(z), in (5.35), replace Zl by
z+dz. We get a*dz = t'l9(a,z)-ldz/L(a,z)-l. Writing
r+t r
dz = /\ /\ dZhk /\ azhk,
h=lk=l
we have
la*dzl = Idet('I9(a, z) Q9/L(a, z))1- 2 Idzl

= Idet('I9(a, z))1- 2n l det(/L(a, z))1- 2m ldzl = lJ(a, z)I- 2n - 2m ldzl·
Since 6(a(z)) = 6(z)lj(a, z)I- 2 , we see that 6(z)-m- n ldzl is the invariant
measure.
Proposition 5.12 The measure 6(z)-m-n (Q9h=l Q9k=l Idzhk Q9 azhkl) gives
an invariant measure on 3 under the action of U'P(IR).
5.3 The Eichler-Shimura Map

Before starting a detailed study of the nearly ordinary part of coherent coho-
mology groups, we make explicit a generalized Eichler-Shimura map for uni-
tary and symplectic groups and hence the association of the weight: Ii t-+ Ii*
so that Hgusp(Xr,r,fd."') '---t Hd(Xr, L(Ii*; C)).
5.3 The Eichler-Shimura Map 245
5.3.1 Unitary Groups
Recall U = Um,n(IR) = {g E GLm+n((C)lgIm,ng* = Im,n}, where Im,n =

(10' _t ). We have made explicit the quotient space U/ Co as the bounded
symmetric domain V = V(m, n). As we have seen, the complexification C of
Co is G Lm (C) x G Ln (C). The functions t h -1 and j correspond to the standard
representation of GL(m) and GL(n), respectively.
For the upper-triangular subgroup Bm C GL(m), we write Tm for the
diagonal torus in Bm. For a dominant weight K, of the diagonal torus T =
Tm,n = Tm x Tn of Um,n, writing Pit for the rational representation into
GL(Lc(K,; C)) of GL(m) x GL(n) of weight K" we have the automorphic factor
Um,n(IR) x V(m, n) 3 (g, z) H PItCh- 1(g, z),j(g, z)),
with which we associate a vector bundle !!,lit on V(m, n) as we described in Sec-
tion 1.1. Take a reductive group G admitting the Shimura variety whose sym-
metric domain X+ is analytically isomorphic to a product I1O"EE V(mO", nO")
(for an index set E). Let T be the torus of G giving rise to the product
I1O"EETm",n". We have the vector bundle Q90"!!,l1t" over X+ associated with
K, E X(T). For a congruence subgroup r C G(Q)+, we want to embed
HO(Xr,!!,lIt) into Hq(X r , Lc(K,*; C)) as Hecke modules for a suitable ratio-
nal representation Lc(K,*; C) of G with highest weight K,* E X(T). By the
Kiinneth formula (cf. [CDR] 1.1.3 and VIL5.2), K,* = EO" K,;, so we may as-
sume that lEI = 1; thus, we drop the subscript (J from our notation.
We have seen in (5.34) that det(h(g, z)) = det(g)-l det(j(g, z)). This also
can be shown as follows. On the diagonal torus Tee U (m) x U (n), for
9 = diag[h, ... , tm, tm+1, ... , tm+nJ, t J satisfies tj = tj1 and j(g, z) = cz+d =
diag[t m +1, ... , tm+nJ (resp. h(g, z) = a + b . tz = diag[tl\ ... , t;;,l]). The
representation 9 H th(g, 0)-1 (resp. 9 H j(g, 0)) corresponds to the standard
representation of GLn(C) (resp. the standard representation of GLm(C)); so,
the corresponding highest character, after applying "det", is
m+n
IT t
m
diag[t1, ... , tm+nJ H J (resp. IT tj).

j=l j=m+1
This relation well explains the above formula (5.34). We thus embed the prod-
uct U(m) x U(n) into GL(m) x GL(n) by 9 H J(g) = (th(g,O)-l,j(g,O)).
We also write J(g,z) = (th(g,z)-l,j(g,z)).
Writing dz = A,J dz tJ , we get from (5.22) that
dg(z) = det(g)n det(j(g, z))-m-ndz.

Write f..lm,n E X(T) for the character
(5.38)
Suppose that K, 2: /Lm,n, and write K,* = K, - /Lm,n' We consider a polyno-

mial function PI<' : V --+ Homc(Lc(K,*; C), Lo(K,*; C)) in z such that
p(a(z)) 0 P~' (J(a, z)) = ap(z) (p~, = Ind~nc K,*)

for all a E G, where G = GL(m) x GL(n). Since V = GIGo, if it exists, such
a function is unique. We can define it by p(a(O)) 0 P~' (J(a, 0)) = ap(O). If
we change a by au for u E U(m) x U(n), then we have
p(a(O)) 0 P~' (J(a)J(u)) = au· p(O)

~ p(a(O)) 0 P~' (J(a)) = au· p(O) 0 P~' (J(u))-l.
Such a map p(O) with up(O) 0 p~,(J(U))-l = p(O) exists because the group
GL(m) x GL(n) is identified with a subgroup of GLd(C) = U(d)(C) (d =
m + n); thus, it corresponds to the inclusion
ff"") = PI<'
L C (K, * ; IL- C
'---+
. dOL(m+n)
In P
C
PI<' ff"") = PI<'
= L 0 (K, * ; IL- 0
for P = diag[GL(m), GL(n)]B. Take K,*ITc to be the highest weight Wn as-

sociated with the standard representation of GL(n). Then K,* corresponds
to the standard representation of Um,n, and we have PW n (z)(x) = (1) x for
x E en. We easily verify that gpwn(z)(x) = Pwn(z)(j(g,z)x). Thus p(z) is a
polynomial in z in this special case. Similarly to the above, if K,* ITc = Wm cor-
responds to the contragredient of the standard representation of G L( m), then
K,* is associated with the complex conjugate of the standard representation
of Um,n, and we have Pw", (z)(x) = (\; ) x for x E em. Again we verify that
gpw", (z)(x) = Pw", (z)(h(g, z)x), and Pw", (z) is a polynomial in z. For general
K" Lc(K,*;'C) (resp. Lo(K,*; C)) is a quotient of Lc(w n ; C)0 t ® Lc(w m ; C)0 8
(resp. Lo(wn ; C)0 t ® Lo(wm ; C)0 8 ). The general PI<-/1-(z) is a constant multi-
ple of the projected image of the tensor product of copies of Pw, (z) and hence
is a polynomial in z.
For a congruence subgroup r c G(lR) , each global section of ~I< is a
holomorphic function J : V --+ Lc(K,; C) satisfying J(-y(z)) = JI<(-Y, z)J(z)
for JI<(-Y, z) = PI< (J(-y, z)). We define for J E HO(Xr,~7d a holomorphic
differential with values in Lo(K,*; C) by w(f) = PI<' (z)(f)dz. Note that here
Lc(K,;C) = Lc(K,*;C)®Lc(/Lm,n;C) and Lc(/Lm,n;C) is one-dimensional; so,
we can identify Lc(K,; C) with Lc(K,*; C) canonically as vector spaces and,
thus, the above definition is consistent. We verify that a*w(f) = p~, (a)w(f).
Theorem 5.13 Let r be a discrete subgroup oj U(m, n)(lR) with the compact
quotient space Xr = r\V(m, n). Suppose that K, 2: /Lm,n as above, and define
K,*= K, - /Lm,n' Then the association J f--t [w(f)] E Hq(X r , L(K,*; C)) Jor
q = dime V induces an embedding HO(Xr,~I<) '---+ Hq(X r , Lo(K,*; C)), where
[w(f)] is the de Rham cohomology class oj w(f).
When r is a noncocompact congruence subgroup of a global unitary group
G IrQ, the same assertion is valid replacing ~I< by its cuspidal subsheaf ~I<'
The theorem follows from standard harmonic analysis when Xr is compact

(see [CDR], [MaM] and also [LFE] 6.2 for the simplest case of G = SL(2) IQ).
Even in the noncom pact case, the same argument holds for cusp forms, since
cuspidal sections decrease rapidly towards the boundary of the canonical
(topological) compactification of X r . The compactification is called the Borel-
Serre compactification which is a manifold with corners (see [BoS]). Since the
detailed proof of this fact relies on harmonic analysis, which is beyond the
scope of this algebraic book, we admit this result.
5.3.2 Symplectic Groups
We give a brief description of the Siegel modular version of Theorem 5.13.

Let fJ g = {z = t z E qllm(z) > O} be the Siegel upper half space, and
take G = GSp(2g) IQ for simplicity, although this works well for any reductive
group G IQ admitting Shimura varieties whose symmetric domain is isomorphic
to a product of copies of fJg. Thus
G(R) = {a E GL 29 (R)laJg . ta = v(a)Jg with v(a) E RX} .
For z,w E fJ g, we write Y(z,w) = (f '1'), and for 9 = (~~) E GSp(2g) , we

have gY(z,w) = Y(g(z),g(w)) (czt d cwO+d). From this, we get
C~w WOZ) = v(g)-lty(z,W). tgJgY(z,w)

_ ()-l (t(CZ+d) 0 ) ( 0 g(W)-g(Z)) (cz+d 0 )
- v 9 0 t(cw+d) g(z)-g(w) 0 0 cw+d '
which implies w - z = v(g)-lt(cz + d)(g(w) - g(z))(cw + d). Replacing w by

z + dz for dz = (dz,)),,), we get d(g(z)) = v(g)t(cz + d)-ldz(cz + d)-l. We
put dz = 1\"2) dz,) and define a character f.-L : G L(g) x G m -t G m by
dg(z) = f.-L(cz + d, v(g))-ldz. (5.39)
Regard f.-L as a character of the maximal split torus TGSp (2g) of GSp(2g) by
the isomorphism TGL(g) X G m ~ TGSp (2g) given by (t, t') I-t diag[t, tiel].
For a given character K, > 0 of the maximal split torus TGL(g), the auto-
morphic factor SP2g(lR) x fJ g => (g, z) I-t p",(J(g, z)) with J(,,(, z) = cz + d for
9 = (~~) gives rise to a vector bundle i!,2.'" on fJg. Now suppose that K, is a
character of TGSp (2g) , and use the same symbol to denote its restriction to
T Sp (2g) = TGL(g). Suppose also that K, ;::: f.-L. Let K,* = K, - f.-L and consider the
GSp(2g)-module L(K,*; C) on which t E TGSp (2g) acts by ¢(g) I-t ¢(gt). We
want to have a polynomial function p: fJ g -t HomdLdK,*;C),L(K,*;C)) in
coordinates of z E fJ g such that p(g(z))op"'-J1-(J(g, z)) = gp(z) (g E SP2g(lR)).
We show the existence and uniqueness (up to a scalar multiple) of the map.
If we change 9 by gu for u E C for the stabilizer C of i = i1 g (i = A) of
fJ g in SP2g(lR), then the expected formula tells us
p(g(i)) 0 P,,_p.(J(g, i)J(u, i)) = gup(i)

~ p(g(i)) 0 P,,_p.(J(g, i)) = gup(i) 0 P,,_p.(J(u, i))-I.
The homomorphism u r--+ J( u, i) identifies C with the definite unitary group

U(g) C GLg(C). Such a map p(i) with up(i) 0 P,,_p.(J(u, i))-1 = p(i) exists
because GL(g) is identified with a subgroup of Sp(2g)(C); thus, it corresponds
to the identity inclusion P"_p. '---+ Ind~Sp(2g) p,,' = L(K,*; C) for the Siegel
parabolic subgroup P C GSp(2g) with reductive part GL(g) x G m . Then we
extend p(i) to p(z) by p(z) = gp(i) 0 P,,_p. (J(g, i)) -1, taking 9 E SP2g (JR.) with
z = g(i). From the construction of p, it is unique up to a constant multiple.
Take K,* = K, - /L to be the highest weight WI associated with the standard
representation of G L(g). Then K,* corresponds to the standard representation
of Sp(2g), and we have p(z)(x) = (l)x for x E cg. We easily verify that
gp( z) (x) = p( z) (J (g, z)x). Thus p( z) is a polynomial in z in this special case.
For general K" Lc(K,*; C) and Lo(K,*; C) are quotients of the tensor product
of several copies of Lc (WI; C) and Lo (WI; C), respectively. The function p( z)
for general K, is given by the projected image of the tensor product of copies
of p(z) for WI and hence is again a polynomial in z.
Again for a congruence subgroup r of GSP2g (JR.), we may identify a global
section of ~"/rc over Xr with a holomorphic function f : S)g -+ Lc(K,; C) with
f (r( z)) = p~ (J (r, z)) f (z) for all, E r. Then we define a differential with val-
ues in Lo(K,*; C) by w(f) = p(z)(f)dz. Note that L(/L; C) is one-dimensional;
so, we can identify L(K,; C) with L(K,*; C) canonically (just multiplying by the
character /L) as vector spaces, and thus the above definition is consistent. We
verify from the above property characterizing p(z) that ,*w(f) = ,w(f) for
all, E r(N). Thus for cuspidal holomorphic forms, the harmonic analysis in
[CDR] again gives us
Theorem 5.14 Let r be a congruence subgroup of SP2g(Q) and /L be the
weight ofTosp(2g) defined as above. Suppose that K, ;:0: /L, and define K,* = K, -
/L. Then the association sending f to [w(f)] in the de Rham cohomology group
Hq(Xr,L(K,*; C)) for q = dimrcS)g induces the embedding HO(Xr,~,,) '---+
Hq(X r , Lo(K,*; C)), where [w(f)] is the de Rham cohomology class of w(f),
and ~" C ~" is the cuspidal subsheaf of ~".
We admit this fact. A more thorough comparison theorem (with proof) of de
Rham cohomology and the Betti cohomology (with coefficients in locally con-
stant sheaves) of Siegel modular varieties can be found in [DAV] Chapter VI.
5.3.3 Hecke Equivariance
We show that the Eichler-Shimura map is equivariant under Hecke operators

and is compatible with our normalization of Hecke operators. We assume G IrQ
to be a reductive group with G I (JR.) isomorphic either to DUEL' SUm~,n~ (JR.)
or to DUEL' SP2n~ (JR.) for a finite index set E, where SUm,n(JR.) is the derived
group of Um,n(lR). In the former case, we call G unitary (or of type A) and
in the latter case, we call G symplectic (or of type C). Again by the Kunneth
formula, to describe the equivariance, we may assume 1171 = 1; so, we drop
the subscript 0". We write /-L for /-Lm,n given in (5.38) when G is unitary and
for /-L in (5.39) when G is symplectic.
We have normalized the Hecke operator on the topological Betti co-
homology group taking the action of ~ E .dB normalized as jf,f(~-1) =
fi:-1(Op~(C1). Note that PK = p---:®x = PXK for an algebraic character
X : G or C -+ G m .
We normalize again in the same way the action on Id.K taking the action of
P;?(C 1) = fi:-1(~)p~(~-1) in addition to the division by /-L(O. Let 1I' = Tp(O
and also write coset representatives as~; that is, T(O = Ut; N(7Lp)~. Recalling
fi:* = fi:/-L-1, we have
w(f)I1I'B = 2)fi:*(O)-lp~*(C1)p(~(z))(f(~(z)))d(~(z))
t;
= 2: p(z) ((fi:/-L-1 (~)) -1 PK-I" (J(~, z)) -1) f(~(z) )/-L(~) -ldz
t;
~ pC') (M(O-' ~>-' «)p,(J«, z)) ']«(Z))) dz ~ w(Jl1f c). (5.40)
Here we have added the subscripts Band C in order to emphasize the space on
which the operator acts; that is, B implies the topological Betti cohomology
Hq(X(U),Lc(fi:*;C)) and C indicates the coherent cohomology HO(Shu,Id. K )
(which is a part of the de Rham cohomology). In short, the extra modification
of the action of the Hecke operator T(O by the character /-L on the coherent
cohomology is absorbed by d(~(z)) = /-L(O-ldz in the topological cohomology.
Hence the normalization of Hecke operators at p is identical on the left-hand
side and the right-hand side of the Eichler-Shimura map. This is why we do
not have such a factor in (5.8). We also remark that in the case of GL(2), the
determinant factor I det(g)IA in the definition (4.37) of the Hecke operator on
the coherent cohomology is the value /-L(g) -1 as above. Thus we do not have
this factor in (4.89).
6
Moduli Schemes
We recall the construction of moduli of abelian schemes. The theory of moduli

varieties of abelian varieties has been studied mainly by Shimura and Mum-
ford in the years 1950s to 1960s. Shimura proved the existence of the moduli
varieties over a canonically determined number field relative to a given en-
domorphism ring, a level N-structure and a polarization in the late 1950s to
the early 1960s. This gives a modulus over the integer ring of the field with a
sufficiently large number of primes inverted.
Basically at the same time, Grothendieck studied the moduli of subschemes
in a given projective scheme X / s (fiat over S) and also that of the Picard func-
tors. The existence of a moduli scheme, the Hilbert scheme Hilb x / s , of closed
fiat subschemes of X / s enabled Mumford, via his theory of geometric quo-
tients of quasi-projective schemes ([GIT]), to construct the moduli of abelian
schemes with level N -structure over Z [h ].
We briefiy recall here the construction of Grothendieck and Mumford,
limiting to the case which we need later. After that, we recall the construction
of Shimura varieties with a canonical family of abelian varieties in the following
chapter.
6.1 Hilbert Schemes
We describe the theory of the Hilbert scheme that classifies all closed S-fiat
subschemes of a given projective variety XIs' This is a generalization of the
earlier theory of Chow coordinates that classifies cycles on a projective variety.
The theory is due to Grothendieck and the main source of our exposition is
his Expose 221 in [FGA]. One can find an exposition in [NMD] 8.2 on this
subject under milder assumptions.

252 6 Moduli Schemes
6.1.1 Vector Bundles
We recall a construction in [EGA] 1.9.4 of algebrogeometric vector bundles

over a scheme S. Let S be a Noetherian scheme and £ be a locally free
sheaf on S of rank n. For an open subscheme 0 c S, we consider the tensor
algebra T(£(O)), which is a noncommutative graded Os(O)-algebra whose
m
"
degree m component is Tm(£(O)) = £(0) Q9os(O) £(0) Q9 ... Q90s(0) £(0)'
and To(£(O)) = Os(O). The multiplication is induced by tensor product
(x, Y) f--t XQ9Y for x, Y E £(0). Then S(£(O)) is the quotient ofT(£(O)) by the
homogeneous two-sided ideal generated by x Q9 Y - Y Q9 x for all x, y E £ (0).
The (commutative) Os(O)-graded algebra S(£(O)) is called the symmetric
algebra of £(0). The functor 0 f--t S(£(O)) is a contravariant functor from
the category of open subsets of S into graded algebras; so, it is a presheaf of
algebras over S. We write S(£) for the sheafication of the presheaf of algebras
([GME] 1.1.2). Then by definition, S(£) is a graded Os-algebra. We define the
relative spectrum V(£) = Specs(S(£)) as an S-scheme (e.g., [GME] 1.5.4).
When S = Spec(B) is affine, for every multiplicative subset S of B, it is
plain that S(S-l M) ~S(M) for M = r(S,£). This shows that S(£) is the
quasi-coherent sheaf S(M) associated with the Os-module S(M) (cf. [GME]
1.2.3). For any B-algebra R, a B-linear map £ : M --+ R extends uniquely to
a B-algebra homomorphism £ : S(M) --+ R by sending Xl Q9 X2 Q9 ... Q9 Xm to
£(Xd£(X2) ... £(xm). Thus the algebra S(M) represents the covariant functor
ALG /B --+ AB given by R f--t HomB(M, R), where ALG /B (resp. AB) is the
category of B-algebras (resp. abelian groups). Since ALG/ B is antiequivalent
to the category of affine S-schemes AF F/ s , V(M) represents the contravariant
functor AFF/ s --+ AB given by 0 f--t Homos(M,Oo) = Homoo(Mo,Oo),
where Mo = M Q9B R for 0 = Spec(R); in other words, Mo = 1* M for the
structure morphism f : 0 --+ S. See Section 4.4.1 in the text and [GME] 1.4
and 1.5 for representability of functors.
In general, covering S by affine open subsets {O}o, we see that the restric-
tion of S(£) to each affine open subset is quasi-coherent and, hence, S(£) is
quasi-coherent over S. By further shrinking the affine open subsets 0, we may
assume that £10 ~ 0 0 . Then S(£(O)) is isomorphic to the polynomial ring
over Os(O) = R generated by a base Tl' ... ' Tn of £(0) over R. This shows
that V(£) is an S-scheme of finite type and hence Noetherian. Since V(£lo)
represents the functor S' f--t Homoo(£lo,Os') = HomosIU*(£lo),Osl) for
f : S' --+ S, the schemes {V(£lo)}o glue over S to represent the functor
S' f--t Homos(£,Osl) = Homosl(£sl,Osl) for £SI = 1*£ (see Remark 1.5.1
on page 31 of [GME]); in other words, we have, for f : S' --+ S E SCH/s,
6.1 Hilbert Schemes 253
where £ is the Os-dual of £. The last equality in (6.1) follows from the defini-
tion of f: Homos' (£s" Os') = r(S', £s,). Thus V(£) gives the vector bundle
associated with the dual sheaf f.
We consider a contravariant functor S' t-+ Homos' (£s' , F s ') for S' -4 S,
where £ and F are two locally free sheaves of finite rank. Note that
Homos' (£s" Fs') ~ £s' 00 s' Fs' ~ j*(£ 00 s F).
Then (6.1) tells us that the above functor is represented by V(£ 00 s :F):
Homs(S', V(£ 00 s :F)) = Homos' (£s' , Fs')

= r(S', j*(£ 0 F)) for f: S' ---+ S E SCH/s. (6.2)
In particular, S' t-+ Endos' (£s') is represented by V(£ 0 f).

Taking an affine open subset 0 = Spec(R) so that £ ~ 0 0 and fixing an
isomorphism £ ~ 0 0 , each ¢ E Endos' (£s') has matrix expression ¢ = (¢i))'
Then we may define det ¢ E Os,. The construction of det ¢ does not depend
on the choice of the base of £10; so, covering S by affine open subschemes,
they glue together into a functorial homomorphism det : Endos' (£s') ---+ Os'
inducing a morphism det: V(£0£)/s ---+ G a / S by Yoneda's lemma (cf. [GME]
Lemma 1.4.1). We define
(6.3)
Then GL(£)/s is a group scheme (cf. [GME] 1.6) representing the functor
S' t-+ Autos' (£s,)· In particular, we write GL(n)/s = GL(O'S) and we have
GL(n)/z = Spec(Z[Tij,det(Tl))-l]) and GL(n)/s = GL(n) Xz S. (6.4)
We can slightly generalize the above argument. Let F be another locally

free sheaf over S with the same rank as £. We consider the functor from
SCH/ s into SETS given by
Isoms(£, F)(S') = {¢ E Homos, (£s" Fs' )i¢ is an isomorphism} . (6.5)
On a small open subscheme 0 c S over which £ ~ F ~ 0 0 , the func-

tor Isoms(£lo,Flo) is represented by an affine scheme Isoms(£lo,Flo) iso-
morphic to GL(£lo), identifying Flo with £10. Covering S by such open
subschemes 0, the schemes Isoms (£ I0, Flo) glue together into a scheme
Isoms(£, F) affine over S by the uniqueness of the universal scheme. The
resulting scheme Isoms(£, F) obviously represents the functor in (6.5).
6.1.2 Grassmannians
We define a contravariant functor Grass: SCH ---+ SETS by

Grasson,m(S) = [7f: as ---» FI7f surjective, F locally Os-free of rank m] .

As before, the straight brackets [*] indicate the set of isomorphism classes of
the objects * inside the brackets. The quotient 7f : as
----t F is isomorphic to
7f' : as----t F' if we have the following commutative diagram,
0---+ Ker(7f) ---+ as ~ F ---+ 0

II II 11
o ---+ Ker( 7f') ---+ as -+ F' ---+ 0
1f'
with exact rows. For each morphism f : S' ----t S, the pullback j*7f : OSI ---»
j* F gives the contravariant functoriality. The stabilizer of 7f : ----t as as
can
be identified with the maximal parabolic subgroup
P = {(g~) E GL(n)ld is of size m x m}.

As is well known (see below Theorem 6.1), the quotient Grasson,m =
GL(n)/ P is a projective scheme defined over Z. Intuitively, this projective
scheme represents the functor Grasson,m, that is,
functorially. We soon verify this fact in down-to-earth terms after stating the
representability as a theorem (Theorem 6.1). Of course, if m = n - lor 1, we
have Grasson,m = pn-l.
We can generalize this construction slightly. Let S be a scheme, and take
a locally free sheaf [IS of rank n. Then, for each S-scheme S' ~ S, we define
a contravariant functor from S-SCH to SETS by
Grasse,m(S') = [7f : [S' ---» FI7f surjective, F locally OSI-free of rank m] .

Covering S by sufficiently small open subschemes 0, so that [0, ~ 0 we o"
have Grassoo"m represented by Grassmlo, = Grassm x 0,. The gluing data
g'J : Oo,no] ~ [o,no] ~ Oo]no, give rise to a Cech co cycle gij with values
in GL(n). This gluing datum induces a gluing datum of {Grassmlo,}, giving
rise to the scheme Grasse,m over S which should represent the above functor.
Theorem 6.1 For each locally free sheaf [IS of finite rank n, the functor
Grasse,m is represented by a projective scheme Grasse,m over S.
Here we use the word "projective" in a strict sense: "strongly projective"
in Section 6.1.6 that a scheme X over S is projective if we have a closed
immersion over S of X into a projective space pN xzS for a positive integer N,
where pN = Proj(Z[Xo, ... , XN]). Thus if S = Spec(B) and G = GrasSe,l ~
pn-l for n = ranke, OG(l) = [ has to be free of rank n over B; so, [ has
to be free. However, we can always find a surjective morphism of sheaves
0:;+1 ---» [ if S is Noetherian, which induces a closed immersion GrassE,l '---+
Grass O N+l I = pN; SO, by enlarging the dimension of the projective space,
S '
we confirm that Grass£,1 is projective. We reduce in the following proof of
projectivity of Grass£,m to this case of m = 1 by taking Plucker's coordinates
(described in the proof).
The scheme Grass£,m gives a geometric quotient of GL(n) by P when
£ = Os in the sense of Mumford (see [GIT] 0.6 and [GME] 1.8.3).
Proof We repeat a proof given in [EGA] 1.9.7. By the argument preceding the
theorem, we may assume that £ = Os. We choose a base I = {VI, V2, ... , v n }
of Os over Os and consider a subset J of order m of I. We then define a
subset GJ(S') c Grass£,m(S') by
where £Sf = 1*£ for f : S' -+ S. By definition, Grass£,m(S') = UJ GJ(S') for

all S-schemes S'. If SI! !4 S' is an S-morphism and (¢ : £Sf -+> F) E GJ(S'),
by the right-exactness of g*, we find that g* (¢) (L.vEJ 0 Sf( v) = g* F, and
hence GJ : SCHls -+ SCH is a sub functor of Grass£,m'
We now show that GJ is represented by an affine scheme 9J. Let £J =
L.vEJ OSfV. Then ¢ : £JISf -+ F is a surjective homomorphism of a locally
free sheaf of the same rank; so, it is an isomorphism. Let X = V(£ 0 Os)
representing S' t-+ Homosf(£sf,O:sn and Y = V(Os 0 Os) representing
S' t-+ Endosf(OS;) (see (6.2)). Define the morphism of functors O'.Sf,/3sf :
X(S') -+ Y(S') by O'.Sf(V: £Sf -+ OS;) = VOLJ,Sf and /3s f (v) = ido:;;, where
LJ,Sf : O~f = Os; -+ £Sf is given by LJ,sf((av)vEJ) = L.vEJavv, Define 9J by
the following fibered product,
9J=Xx y X ----7 X
1 1<>
X ----7 Y.
(3
Then we see
QJ(S') = {(¢, w) E Homosf (£Sf, 0s;)21¢ 0 LJ,Sf = idO~f}
~ {¢: £Sf -+ Os; I¢("L OSfV) = os;} = GJ(8').

vEJ
Thus the scheme 9 J represents the functor G J. Since X and Yare affine over
8 and X is affine over Y, 9 J is affine over 8.
Take another subset J' c I of order m. We show that G J n G Jf is repre-
sented by an open subscheme of 9J. If ¢ E G J (8'), then v = ¢OLJ,Sf : Os; -+ F
is an isomorphism. Write its inverse as w : F -+ OS;. Then w 0 LSf,J is an
element in Endosl (OS;); so, we have det( w 0 L5 ,J') E G a (S'). The association
1
¢ H det(w 0 LSI,JI) induces a functorial map of GJ into G a ; so, by Yoneda's

lemma ([GME] Lemma l.4.1), we have a morphism of S-schemes <5 : (h -+ Ga.
Regarding <5 as a section of OgJ' we find that G J n G J' (S') = flJ m
(S'). This
shows that the intersection G J n G J' is represented by an open affine sub-
scheme of C;h (and hence of QJ')' and {QJ} J glue into a scheme Grass[,m repre-
senting the functor Grass[,m' For each valuation ring R with f : Spec(R) -+ S
and quotient field K = Frac(R), (fK[ ~ F) E Grass[ m(K), we can ex-
tend F = V uniquely to Spec(R) as follows. Regarding fK : Kn -+ V, we
have M = Rn / (Ker(fK) n Rn). Then defining F = M, we find an extension
f : An -+ F. Plainly this is a unique torsion-free extension of F. Thus by the
valuative criterion of properness ([ALG] II.4.7 and [GME] Theorem l.9.2),
Grass[,m is proper over S.
We give a sketch of a proof of the projectivity of the scheme Grass[,m.
Since the property is local on S, we may assume that S' = Spec(R) and
S = Spec(B) for an algebra B. Then we write [ = An. For each surjection
Rn --» M for a flat R-module M of rank m, further shrinking Sf, we may
assume that M ~ Rm. Having a surjection ¢ : R n --» R m is equivalent
to having an m-subset J of a base I of Rm such that L5',J : R m y R m
composed with ¢ is an isomorphism of R-modules. This in turn is equivalent
to det(¢oL51,J) E RX. This shows that ([51 ~ F) H ((I\m [)51 /\Tn1;) I\m:F)
gives rise to a functorial injection L : Grass[,m(S') y Grass(/\Tn [),1 (S'), where
r+ 1 = rank o S 1\ m [. The point of the projective space associated with a point
[51 ~ F in the Grassmannian is called the Plucker coordinate of [51 ~ :F.
Thus the morphism is an immersion. Since GrassE,m is proper over S, it has
to be a closed immersion (cf. [GME]l.9.1).
As we remarked just after stating the theorem, even if [ is merely locally
free, Grass(/\ Tn [),1 is projective, although Grass(/\ Tn [),1 itself may not be a
projective space. This shows the PlUcker coordinate embedding: Grass[,m y
Grass(/\ Tn [),1 composed with a projective embedding of Grass(/\ '" [),1 gives a
projective embedding of Grass[,m. 0
6.1.3 Flag Varieties
We can further generalize our construction of Grassmannians to flag varieties.

We follow [EGA] 1.9.9. We consider the following functor from S-SCH to
SETS,
Fl (S') [ .[ , --* [ IKer('ifJ+r) C Ker('if)) and ]

~B = 'if). 5 ) [J is locally free of rank n - j (1 :::; j :::; n - 1) .
Here the subscript B indicates a split Borel subgroup of GL(n), since Flag is
represented by a projective scheme FlagB = GL(n)/ B if [ = Os as we show
below (Theorem 6.2). Since [ ~ Os locally, covering S by open subschemes
6.1 Hilbert Schemes 2.57
o over which [ is free, we can glue schemes over the open Os representing
Flag B/ o to get a scheme representing Flag over S. The association ('ll'] : [s' -+
[]) r-+ ('ll'])] induces a functorial map,
which is an injection for all j. Thus if Flag B is representable by a scheme

FlagB/s, it is an immersion. As in the case of GrassE,], we can show that
FlagB / S is proper; so, the morphism i is a closed immersion. Thus FlagB / S is
projective if it is representable ([EGA] 1.9.9.3).
Theorem 6.2 The functor Flag B/ s for a locally free sheaf [ of rank n over
S is representable by a projective scheme FlagB / s.
Proof. By the argument preceding the theorem, we only need to prove the
represent ability assuming [ = Os. Write
v = GrassE,l Xs GrassE,2 XS'" Xs GrassE,n-l .
We have a universal morphism 'IT] : [v -+ E]. Let M = Ker('IT] EEl'ITJ+d.

Then for each x = ('ll'], 'll'J+d E V, dimM ®Oy k(x) = n - j - 1 implies that
Ker('ll']) ~ Ker('ll']+d, otherwise dimM ®Oy k(x) <:: n - j - 2. Then
~ = {x E VI dimM ®Oy k(x) = n - j - I}
is a closed subscheme of the support of M (cf. [GME] Lemma 4.l.3). Thus

Im(i) = n] ~ is a closed subscheme of V. This shows the desired repre-
sentability of Flag B/ s ' 0
We can generalize the construction of flag varieties to vector bundles over

FlagB' We consider the following functor,
Here we understand that Ker( 'll'o) = [, and j runs over all integers between 0
and n - l. If [ ~ Os and S is affine, writing 1 = ('ll'], cP]) for the standard flag
'll'] : Os -+ O~-], projecting column vectors down to lower n - j coordinates,
the upper unipotent subgroup U of G L( n) = Autos (Os) is the stabilizer of
1. Therefore Flag u is represented by Flagu/s = GL(n)jU. In general, cover S
by open affine schemes S, = Spec(8,) so that we have an isomorphism [Is, ~
Os,. On S'] = S, n S]' by the universality, Flagu / s, x S S'] is canonically
isomorphic to Flagu/s J Xs Sq. Thus these schemes glue each other, giving
rise to a scheme Flagu/s representing Flag u ' Obviously Flagu/s is a T-torsor
for the maximal split torus T c G L( n). Here the action of Ton Flagu is given
n
Let W : Flagu -t FlagB be the projection (Wj, rPJ) ~ (Wj). Then for a
character K E X(T), we define a sheaf £1<.(0) = HO(w-1(0),OFlagu[K]) for
each open subset 0 c FlagB' Then £1<. is a locally free sheaf on FlagB' Since
f : FlagB -t S is proper fiat over S, we find that f*£1<. (which we again write
£1<.) is a locally free sheaf on S. In this way, we can associate a K-power £1<.
of the original locally free sheaf £ for each weight K E X (T) with respect to
(B, T), which is non-zero if and only if K is a dominant weight of GL(n) with
respect to the pair (B, T).
6.1.4 Flat Quotient Modules
Let f : X -t S be a projective scheme over a (separated) Noetherian con-

nected scheme S of relative dimension n. Here the word " projective" means
that we have a closed immersion t : XIs Y pfs' Thus X has a very
ample invertible sheaf Ox(l) = t*OpN(l). The sheaf of graded algebras
A = EB:=o f*(Ox(l)n) determines X as X = Projs(A).
For a coherent sheaf M on X, we write M(k) for M ®ox Ox(k) and
define a sheaf of graded modules M = EBk2:0 Mk by Mk = f*M(k). Then M
is a graded A-module of finite type, and we have M = M, where M is the
sheaf on Projs(A) associated with the graded module M (e.g., [ALG] 11.7 or
[GME12.3.3). First removing finitely many graded pieces of M does not alter
M = M. If M is Os-fiat, for each geometric point s = Spee(k(s» E S, there
is a polynomial PM (X) such that
dimX(s)
x(M(n» = L (-l)j dimk(s) Hj (X(s), M(n) ®Os k(s» = PM(n),
J=o
since the Euler characteristic is additive with respect to exact sequences. For
sufficiently large n, the ampleness of 0(1) tells us (a theorem of Serre: [ALG]
III.5.2) that HJ(X(s),M(n) ®os k(s» = 0 if j > O. Thus PM(n) gives the
dimension of HO(X(s), M(n)®os k(s», which is equal to the Os-rank of f*M
(by fiatness of M); so, PM is independent of the choice of s E S (connectedness
of S). When S = Spee(B) , M is associated with a graded B-module, and
PM(n) is the Hilbert polynomial of this graded module (cf. [CRT] Section 13).
For a given coherent sheaf F/ x, we consider the following contravariant
functor defined over locally Noetherian S-schemes SI J:.." S (inducing rPx :
XSI=XXSSI-tX),
Quot)',! X/S(SI)
= [w: rP'XF -+> MjM is a coherent OXsl-module fiat over OSI ] .
The isomorphism between the ws are defined in the same manner as in the
case of the functor Grass. Here we do not assume that :F is Os-fiat.
For simplicity, we always assume that schemes S' (and also S) are Noethe-
rian. Each M E QuotF/X/S(S') has its Hilbert polynomial PM, and obviously
for g : SI! -+ S', g* M has the same Hilbert polynomial. Thus we can split the
functor as
Quot F / X / S = UQuot;/x/s'
P
where Quot;/x/s(S') = [1T: ¢*xF -» ME QuotF/x/s(S')lpM = p]. Here is

a theorem of Grothendieck.
Theorem 6.3 Let the notation be as above. Suppose that S is Noetherian,
that F/ s is coherent, and that XIs is a projective scheme of finite type with
dims X = n. Then Quot;/x/s is representable by a projective scheme QP =
Quotj./x/s of finite type over S. Thus we have, for any S-scheme S',
We give a sketch of a proof of this theorem.

We recall X = Projs(A) for a sheaf A of graded Os-algebras generated by
AI. We cover Quot;/x/s by subfunctors Q j indexed by nonnegative integers
j defined as follows. QJ (S') consists of isomorphism classes of 1T : F/xsi -»
M/xsi satisfying the following three conditions:
(a) R'fsl,*M(n)/xsl = 0 for all i > 0 and n ~ j;
(b) R'fsl,*JC(n)/xsl = 0 for all i > 0 and n ~ j, where K = Ker(1T);
(c) Akfsl,*(K(j)) = fS',*(K(j + k)) for all k > o.
Covering S' by affine schemes Spec(B,) and writing B, as a union of Noethe-
rian rings, we can reduce proofs to Noetherian S'; so, we assume that S' is
Noetherian as we remarked already. By a theorem of Serre ([EGA] III Sec-
tion 2 or [ALG] III.5.2), for any coherent sheaf Q/ XS" we have the vanishing:
R'fsl,*Q(n)/xsl = 0 for n» O. Thus (a) and (b) will be satisfied for a given
1T for j » O. Since F is coherent (and X / s is of finite type), it is finitely pre-
sented; so, K is finitely presented because M is finitely presented and locally
free. Thus K is generated by K(j) for some j, and the last condition will be
fulfilled if j» O. Hence Quot;/x/s(S') is covered by QJ(S') for each S'.
Since the sequence 0 -+ K -+ F -+ M -+ 0 is exact, we have X(F) =
X(K) + x(M). Thus the conditions (a) and (b) tell us that Hilbert's polyno-
mials Pdn) and PF(n) give the exact Os,-rank of fS',*K and fs',*:F. Van-
ishing of RI fSI,*X/xSI = 0 implies that fSI,*¢*XX = ¢* f*X/x ([EGA] III,
7.7.5, 7.7.10, 7.8.4, [ALG] III.12.10 or [GIT] 0.5); so, the conditions (a) and
(b) are stable under base change. The tensor product is a right exact func-
tor; so, the surjectivity of Pk : Ak ® fSI,*JC(j) -+ fSI,*K(j + k) is also kept
under base change. Thus Q is a well-defined contravariant functor, and we
-J
have Quot;/x/s = UJ Qj" By the conditions (a-c), 1T E QJ(S') is determined
by M J = fS',*M(j) as a flat quotient of rank P(j) of FJ = fS',*F(j). Thus

7[" H (7["J : FJ ---» MJ ) induces a functorial injection Q/S') y GrassFJ,p(J) (S').
If 7["' is in the image of Q , then 7["' satisfies:
-J
(i) FJ+k/AkK' for K' = Ker(7["') is locally Os,-free of rank P(j + k) for all
k ::::: o. Here we consider AkK' in A . F;
(ii) Define a graded module K'* by A· K'. We require the associated sheaf
~*
K' = K'/x , on Xs' to satisfy (b) and the quotient M' = F/K' to satisfy
(a) (these (a) and (b) are open conditions: [ABV] Section 5 Corollary or
[ALG] Theorem III.12.8).
For a graded A-module M = EBk M k , putting M(t) = EBk;:::t M k , we have
M(t) ~ M as already remarked. Then the condition (i) ensures that fS',*M'
is locally Os,-free, and the image of Q (S') is characterized by (i) and (ii).
-J
We prove the represent ability of Q , assuming that j

-J
= O. The general
case follows from the same argument replacing 0 by j everywhere. Let 7["univ :
Fo/ c ---» Mo (Fo/c = Fa 00 s Oc) be the universal quotient defined over
G := Grass. Here we change our notation and write Mo for the universal
quotient of Fo/ c (with rank P(O)). Thus for any morphism 7["' : Fa -+ M6 over
S' with M6 locally free of rank P(O), we have a unique morphism ¢ : S' -+ G
such that 7["' = ¢*7["unw. Let K = Ker(7["unW). Write g : G -+ S for the
structure morphism. We consider the subset:
Z _ { Gldimk(s) (AkFO/C/(Ak K )) 0 k(s) = P(k), and }

- s E the stalk (AkFo/ c / (AkK))s is OC,s-free for all k ::::: 0 .
Write Mk = AkFo/c/(AkK) and put M = EBk>o Mk. Then Mk = AkMO.

The Oc-module M is flat on a generic point of .2'0 := Supp(M) = G. Since
flatness is an open condition, we find an open connected subscheme Va C Zo
which is maximal among open subschemes V over which M is flat. Repeating
this process, replacing M and Zo by M 00 zo OZl and Zl = Zo - Va, we
can split Zo = Ui Vi into a finite disjoint union (as a topological space) of
connected subschemes Vi so that M 0zo Ov; is flat over Vi. Then we find a
polynomial Q,(n) such that rankov (Mn 0zo Ov.) = Q,(n) if n ::::: n,. By this
fact, the subscheme '
UN = {s E Zol dimk(s)(Mn 00 zo k(s)) ::::; P(n) 0::::; 'Vn ::::; N}

stabilizes as N grows. Since the function s H dimk(s)(Mn 00 zo k(s)) is up-
per semicontinuous (e.g., [ABV] Section 5 Corollary or [ALG] III.12.8), the
scheme UN is an open subscheme of Zoo We have now verified that on an open
subscheme U = Uoo of Zo, we have dimk(s)(Mn 00 zo k(s)) ::::; P(n) (if n ::::: 0)
for all s E U. We have an exact sequence Oij ~ O~(k) -+ Mk -+ 0, and Z
is given by the closed subscheme of U on which all matrix coefficients of Pk
vanish for all k ::::: O. Thus the image of Qo falls into ~. The condition (ii) can
be checked to be satisfied on an open subscheme of Z. Thus we have
Proposition 6.4 The functor Q is represented by a quasi-projective scheme

-J
Q J of finite type over S.
Here the word "quasi-projective" means that the scheme has an open immer-
sion into a projective scheme. Since Grass is projective, QJ is quasi-projective.
The next step is to show that the increasing sequence of quasi-projective

schemes {QJ}J stabilizes after j ~ No; so, Quot;IXIS is represented by a
quasi-projective scheme. See [FGA] Expose 221, Section 2 for an argument
valid for a general X IS'
We prove this fact only for :F = Ox in a manner slightly different from
[FGA] Expose 221. First assume X is the projective space Pis and:F = Ox.
Writing X = Proj(Os[to, ... ,tn]) and D2 C X for Spec (Os [f,"" ~7])'
then Ri f*O(j) = 0 for all i > 0 if j > n + 1 by a computation of cohomology
groups by Cech cohomology with respect to the covering X = U;=o D J (see
[ALG] III.5). A version of the argument of Grothendieck for X = pn to
prove (a) and (b) for sufficiently large j and all M and JC is as follows. Since
PF = PK+PM with PM = P, PK is determined by P. Choosing homogeneous
generators Xl, ... ,Xr of degree -p of JC, we have a surjection 1To : O(p y --+ JC
taking (al,"" ar) M L~=l aixi. Here rand p are determined by the first
two leading terms of PK and hence those of P. Let JC o = Ker(1To), Then
rPO(p) = P Ko +PK. Let ro = rand p = Po· The polynomial P Ko is determined
by PK. Thus the first two leading terms of PK o are bounded below and above
independent of JC. Repeating this argument, we find an integer Nl » 0 such
that for integers Pi > -Nl (i = 0,1, ... , n) we have the following exact
sequences: 0 --+ JC 2 --+ O(P2Y' --+ JC 2 - l --+ 0 with JC- l = JC. By the associated
long exact sequence, if j ~ Nl + n + 1, Rq+l f*JC i (j) ~ Rq f*JC l - l (j). Since
cohomological dimension of pn is n (as easily checked by Cech cohomology),
for i > 0, 0 = Rn+2+l f*JCn(j) = Rn+2 f*JC n - l (j) = ... = R" f*JC(j). By the
same argument, R" f*M(j) = 0 for all i > 0 and all j ~ Nl + n + 2 = No. SO
QJ stabilizes after j ~ No·
It is customary to identify 1T E Quotoxlxls(S') with a closed immersion
of SpecQ (Im( 1T)) into X; so, Quot~ x I XI s represents the contravariant functor
Hilb~/s(S')
= {closed subschemes of X S ' flat over S' with Hilbert polynomial P} .
This scheme is called the Hilbert scheme of X for the polynomial P. There is
a direct way (as indicated in [GIT] 0.5.iii) of constructing Hilb~/S for general
X c pn assuming its existence for X = Pis as proved above. We give details
of this. Let X C Pis be a projective scheme. We write H = Hilb~nlS and
let 1T : Z --+ H be the universal projective scheme over H with i : Z '-+ PiH'
In other words, for any closed subscheme Z C Pis, for S' E SCH IS with
Hilbert polynomial P, we have a unique S-morphism </> : S' -t H such that

the following diagram commutes.
</>* Z =Z XH S' ~ </>*p/H = p/H XH S'
11 11
Z ----+
c
We define X H = X Xpn/S p/nH and Zx Z Xpn/H X H = X Xpn/S Z as a
subscheme of p/w We define Hilh~/s to be the image of Zx in H, that
is, the topological image with reduced scheme structure. Since Zx -t H is
projective, the image Hilb~/ S is a well-defined closed subscheme of Hilb~n / s·
Then, if Z/SI is a closed sub scheme of X/S', </>* Z = Z c X, and hence
</>*(Zx) = Z with </> : S' -t H factoring through Hilb x / s . Thus we have
Z C XSI = X Xs S' {:=::? </> factors through Hilb~/s.
From this, Zx -t Hilb~/s is the universal closed subscheme in X xsHilb~/s,

and Hilb x / s represents the functor Hilb~/s.
We now finish the proof of the following theorem ([FGA] Expose 221,3.2).
Theorem 6.5 (Grothendieck) For a projective scheme XIs over a Noethe-
rian connected scheme S and a numerical polynomial P(t) E lQ[t], the functor
Hilb~/s is represented by a projective scheme Hilb~/s.
Proof. We only need to prove projectivity of Hilb~/s by the valuative crite-

rion. Let 7r : Ox." -t M/'1 E Quot~x/x/s('T]) for 'T] = Spec(K) of the field
K of fractions of a discrete valuation ring V. Then we define Ker( 7r) / S for
S' = Spec(V) by the largest subsheaf over S' of OXSI inducing Ker(7r), that
is, OXSI n Ker(7r), which is a coherent sheaf with quotient M/sl locally free
over XSI inducing M/ry after tensoring K, because V is a discrete valuation
ring. Thus the point 7r E Quot~x/x/s('T]) extends to Quot~x/x/s(S'). Since
Quot~x/x/s = U~~o Qj is quasi-projective, it is separated; so, it is proper.
Since Quot~x/x/s is quasi-projective, it has to be projective. 0
6.1.5 Morphisms Between Schemes
Let Y/ s -4
XIS be a morphism of S-schemes. Let Secy/x/s SCH/s-t
SETS be contravariant functors given by
Secy/x/s(S') = Homxsl(Xsl, YSI).

Each section s : XSI -t YS' defines a closed subscheme (J(XSI) of YS' iso-
morphic to X s' via f S'. Thus if X is flat over S, (J (X SI) is flat over S'; so,
O"(Xs') E Hilby/s(S'). Write H = Hilb y/ s = Up Hilb~/s. Then we have the

universal closed subscheme Z of Y H = Y Xs H satisfying the commutative
diagram
1 1
H=H
such that for any S-scheme S' and a closed subscheme V y YS' flat over S',
we have a unique morphism <pv : S' -+ Hover S such that the pullback of
the above square by <Pv is identical to
1
S'
1
Sf.
We consider the S-subschemes U c H such~that lH : Z C YH ~ XH for a

given f --=- Y -+ X induces an jsomorphism fu : Zu ~ Xu. If there exists th~
largest U with this property, U represents the functor Secy/x/s. The largest U
is given by fJ = Z - (Supp(Coker(11: 110xH -+ Oz)) USupp(Ker(11))).
Thus fJ is an open subscheme of H; so, each connected component of fJ is
quasi-projective, since Hilb~/s is projective over S.
We consider two S-schemes X and Y. Then Yx = Y Xs X has projection
Px : Yx -+ X. Note that Secyx/x/s(S') = Horns' (Xs', Ys'). Thus if XIs
is flat, we can apply the above argument to Px : Y x -+ X and the functor
Homs(X, Y) : S' H Horns' (Xs' , YS') is representable.
Theorem 6.6 Let XIs and Y/ s be projective schemes over a Noetherian
scheme S. Suppose XIs is fiat. Then the functors Secy/x/s for an S-
morphism Y ~ X and Horns (X, Y) are representable by S-schemes Secy/x/s
and H(X, Y)/s over S, respectively. Each connected component of Secy/x/s
and H(X, Y)/s is quasi-projective over S.
By construction, the scheme representing these functors may not be of fi-
nite type over S, because Hilb x / s could have infinitely many components.
However, each connected component of the scheme is of finite type over S.
If sections s : S Y X and s' : S Y Yare given, we may consider a
subfunctor
Corollary 6.7 Let the notation and the assumption be as in the theorem.
The functor HornS,s' (X, Y)/s is representable by an S-scheme HS's' (X, Y)/s
whose connected components are quasi-projective over S.
Proof. Consider the functorial map Homs'(Xs' , Y s ') :3 ¢ I---t ¢(ss') E Y(S')
which induces a morphism (J : Hom(X, Y)/s -+ Y by Yoneda's lemma
(see [GME] Lemma 1.4.1). Then the functor Homs,s' (X, Y)/s is again rep-
resentable by a scheme HS,s' (X, Y)/s = H(X, Y)/s xY,u,s' S over S. 0
Corollary 6.8 Let the notation and the assumption be as in the theorem. For
a section s : S '---+ X, consider the following functor,
The functor SCH/ s :3 S' I---t E"s,(X s ') E SETS is representable by a scheme
x/
E s over S whose connected components are quasi-projective over S.
6.1.6 Abelian Schemes
An abelian scheme A/ s is a smooth geometrically connected group scheme

proper over a separated locally Noetherian base S. An abelian scheme A/ s
with a polarization is projective (existence of polarization is equivalent to
existence of an ample line bundle; see [ABV] Sections 16-17). Hereafter we
always deal with polarized abelian schemes (so all abelian schemes we study
are projective).
We can drop the "local Noetherian" hypothesis, because a smooth geo-
metrically connected and proper group scheme over any base is a base change
of such a scheme over a locally Noetherian base (cf. [DAV] 1.1.2).
For simplicity, we actually suppose that S is a Noetherian scheme. Since
A is a group, it has the identity section 0 : S -+ A. As in the elliptic curve
case, any S-morphism ¢ : A -+ A' of abelian schemes is a homomorphism
if ¢(OA) = ¢ a 0 = ON (by the Rigidity lemma: [ABV] Section 4, [GIT]
6.4 and [GME] 4.1.5). In particular, if A is an abelian scheme over S, every
scheme endomorphism of A/ s keeping the zero section is a homomorphism of
the group structure. Thus E~/ s is a ring scheme associated with the functor
S' I---t E~, (As') with values in the category ofrings.
Assume that A is an abelian scheme over a Noetherian base S. Take a
connected component E c E~/s' Each connected component of E~/s' in
particular E, is quasi-projective over S. Since S is Noetherian, E is of finite
type over S, because of our construction E~/s = H(A, A)/s XA,u,O S.
Suppose we have a discrete valuation ring V with field of fractions K and a
morphism TJ : Spec(K) -+ E which is over a morphism i : Spec(V) '---+ S. Then
TJ gives rise to a section of E~/s(K). Since homomorphisms of abelian schemes
are kept under specialization (which we call the rigidity of endomorphisms; see
[GME] Subsections 4.1.5-6 and [DAV] 1.2.7), TJ extends to Spec(V) uniquely.
By the valuative criterion of properness, we find that E is projective over S.
If ¢ is an endomorphism of the abelian scheme A / s, Ker( ¢) is again a group
scheme, because it is given by A x A,q"Os S. If dims Ker( ¢) = 0, Ker( ¢) is a
locally free group scheme of finite rank (see [GME] Theorem 4.1.17 (2)); in
this case, we call ¢ an isogeny. Thus ¢ is finite and hence affine (cf. [ALG]
Exercise II.5.17). This implies ker(¢) = Spec(R) for a locally free Os-algebra
R. We define the degree deg(¢) of ¢ by the rankos R. If dims Ker(¢) > 0, we
simply put deg( ¢) = o.
If the connected component E c E~I s contains an isogeny, the degree
is independent of the point of E. To see this, we consider AE = A Xs E.
Then ¢ induces ¢E E End(AEIE)' we have deg(¢) = deg(¢E), and deg(¢E)
is constant on the connected component of E; so, it is constant on E. For an
abelian variety over a field, the number of isogenies with a given positive degree
is finite (positivity of the Rosati involution: see [ABV] Section 21). Thus E is
projective and quasi-finite; so, E is finite over S ([GME] Proposition 1.9.11).
We can generalize this argument to a strongly quasi-projective semi-
abelian scheme. A scheme X -4 S is strongly projective if X Isis finitely
presented and there exists a locally free sheaf Els of a constant finite rank
and we have a closed immersion X '-+ P(E) over S. If XIS has an open
immersion into a strongly projective scheme XIs' we call XIS strongly quasi-
projective. An abelian scheme A ls equipped with a polarization A: A -+ t A is
strongly projective (over any base S), because for a locally ample line bundle
L, Ll = L0 2 is globally ample as remarked in Section 4.1.1 and L~3 is very
ample (e.g., [ABV] Section 17).
As before, we suppose that S is Noetherian and any scheme XIS we con-
sider is of finite type over S. Then automatically XIS is finitely presented. If
XIS is projective, then by definition (cf. [EGA] II.5.5), we have a coherent
sheaf Els and a closed immersion of X into P(E). Covering E by O~+l, we
have a closed immersion P(E) '-+ Pis. Thus XIS is strongly projective under
our finiteness condition. Also if XIS is quasi-projective, it is strongly quasi-
projective under our finiteness assumption. Thus the results stated in [NMD]
Chapter 8 under some strong (quasi) projectivity assumptions hold under the
corresponding projectivity assumptions in our setting.
A semi-abelian scheme 91s is a smooth separated group scheme with ge-
ometrically connected fiber such that each geometric fiber is an extension of
an abelian variety by a torus. The toric rank may depend on the fiber. If the
abelian scheme has polarization, 91s is quasi-projective. We assume that 91s
is a quasi-projective semi-abelian scheme which is an abelian scheme over a
dense open subscheme 0 of S. The semi-abelian scheme over the toroidal com-
pactification of the Mumford moduli extending the universal abelian scheme
(see Theorem 6.20) constructed in [DAV] VI.1 is (strongly) quasi-projective
(because its abelian variety quotient has polarization and the toroidal com-
pactification is Noetherian). Then HilbQls exists as a quasi-projective scheme
(we only proved this when 9/ S is projective; see [NMD] Theorem 8 in Sec-
tion 8.2). Thus the scheme EgiS exists. It is known that any homomorphism of
semi-abelian schemes 90 -+ 90 defined over an open dense sub scheme 0 c S
extends uniquely to 91 s -+ 9j s (endomorphisms are kept under specializa-
tion; a theorem of M. Raynaud, [NMD] 7.4-5 and [DAV] 1.2.7). Thus if 9/s
is an abelian scheme over a dense open subscheme 0 c S, we have a unique
extension of the scheme Eg % over 0 to a scheme End g / s C Eg/ s over S
which represents the functor T I---t End(QT IT) for the group endomorphism
algebra End(QT IT). Applying the valuative criterion using this rigidity of en-
domorphisms, we find that End g / s has connected components each projective
over S. This shows the following fact.
Corollary 6.9 Let 9/s be a quasi-projective semi-abelian scheme over a
Noetherian base S. Suppose that 9 is an abelian scheme over an open dense
subscheme of S. Then the functor T I---t End T (9T/T) is represented by a
scheme End g / s C Eg/ s over S. Each connected component of End g / s is
projective over S. If the connected component contains an isogeny, it is finite
over S. Here EndT(QT/T) denotes endomorphisms of 9 compatible with group
structure on 9.
This result enables us to prove
Corollary 6.10 Let S be a Noetherian scheme, and 9/s be a quasi-projective
semi-abelian scheme which is an abelian scheme over a dense open subscheme
of S. Let D be a finite-dimensional simple algebra over Q and 0 = OD C D
be a subalgebra such that OD is of finite type over Z and D = OD Q9z Q.
Then the functor SIs I---t [(Qs"e E Homalg(OD,End(Qs'/s,»)le(lD) = idg]
is representable by a scheme Eg/ s over S, and each connected component of
Eg/ s is finite over S. If S is a scheme over a discrete valuation ring of residual
characteristic p > 0, each connected component of Eg/ s is unramified over S
in the sense of [SGA] 1.3. The algebra homomorphism e : OD --+ End(Qs' /s')
as above is in fact an embedding.
Proof. We start with a general argument. Taking its connected component, we

may assume that S is connected. Since Endg / s is a disjoint union of projective
schemes over S (by the above corollary), for any given connected projective
scheme Y fiat over S and each connected component E/ s of Eg/ s ' the functor
T I---t HomT (Y x s T, E x s T) is representable by a scheme H (Y, E) / s over
S. Taking a disjoint union of H(Y, E) indexed by connected components of
Eg/ s ' the functor T I---t HomT(Y Xs T, Eg/ s Xs T) is representable by a
scheme H (Y, Eg / s) / s. If Y is a disjoint union of projective schemes Y fiat
over S (with a given S-point y), taking the fibered product of H(Y,Eg/s)/s
over S indexed by connected components Y of Y/s, again the functor T I---t
Homi: 1 (y xsT, Eg/ s xsT) is representable by a scheme H(Y, Eg/s)/s, where
the superscript Hom y ,1 indicates that morphisms are required to take y E
Y(S) to the identity in End(Q/s) (see Corollary 6.7).
Now suppose that Y is a ring scheme fiat over S with the identity sec-
tion 1y. Then the compatibility of ¢ E H(Y, Eg/s)(T) with multiplication
and addition of the two ring schemes Y and Eg/ s is a closed condition. For
example, the condition ¢ 0 +y = + 0 (¢ x ¢) for the addition +y for Y

and + for Egis can be interpreted as the identity +* 0.:1 = +y for the
diagonal map .:1 : H(Y, Egis) -+ H(Y, Egis) Xs H(Y, Egis)' where +* :
H(Y,Eg Is ) xsH(Y,Eg Is ) -+ H(Y Xs y,Eg Is ) is (¢ x ¢') H (+o(¢ X ¢'))
and +y : H(Y, Egis) -+ Z is given by ¢ H (¢ 0 +y). If we have two mor-
phisms of S-schemes a, b : T =t T', the maximal subscheme Ta=b C T on which
a = b is given by .:1::/(T XT' T) for the diagonal map .:1T : T -+ TXT' T,
where the fibered product is taken with respect to a and b. Applying this
to T = H(Y,Eg Is ) and T' = H(Y Xs Y,Eg Is ) and (a,b) = (+* o.:1,+y),
the functor: T -+ {additive morphisms in Horns (y, Eg IS)} is represented by
the closed subscheme Z of H(Y, Egis)' By the same argument applied to Z
and the multiplication of the two ring schemes and taking the intersection
with Hly,idg (y, Egis)' the functor T H HomT-rings(Y Xs T, Egis Xs T) is
representable by a scheme H S-rings (y, Eg I 5)' where ring homomorphisms are
supposed to take ly to the identity map id Q of Q.
Recall the ring 0 = ODin the corollary. Let 0 15 be the constant ring
scheme with fiber 0; so, Q(T) = 07r o (T) for the set 7ro(T) of connected com-
ponents of T. Since each connected component of Q is isomorphic to S, we
can apply the above argument, getting an S-scheme HS-rings(O, Egis)' Since
the endomorphism ring of an abelian scheme over a base is torsion-free (see
Theorem 8.7), any nontrivial ¢ E HomT-rings(QT' Egis xsT) for OT = 0 xsT
is an injection, because OD modulo a nonzero two-sided ideal is a finite ring
(simplicity of D). Thus we have an S-scheme Egis representing the functor:
T H {a : 0,---+ End(QT/T)}, where a is a ring embedding.
We verify projectivity of each connected component E of Egis by the
valuative criterion. Let V be a valuation ring with fraction field K. Suppose a
K-point f of E is over a V-point f E S. As we already remarked, af : OD '---+
End(QK) extends to an algebra homomorphism OD '---+ End(Qv). Then f
extends to a V-point of E. By the valuative criterion, E is proper over S. We
already know that E is quasi-projective and hence is projective over S.
We now verify finiteness of E over S. Take a geometric point S E S.
Since 0 D is generated over Z (as a ring) by finitely many elements Xl, ... ,X m
invertible in D, as : 0 '---+ End(Qsls) is determined by the values at the gener-
ators Xl, .. . , Xm. Suppose that as gives rise to a geometric point of E. Write
[a(x))] (resp. [x)]) for the connected component of EgiS (resp. 0) contain-
ing as(Xj) (resp. x)). We have a morphism E -+ [Xj] Xs [a(Xj)] by sending
a E E(S') to the Sf-point (x),a(x))) of the fibered product. Since a is de-
termined by the values at x)' this functorial map induces an immersion of
E into ([Xl] Xs [a(xdD Xs ... Xs ([Xm] Xs [a(xm)])' Since E is projective,
this immersion is a closed immersion. Since 0 is constant, the above fibered
product is isomorphic to [a(xdJ Xs ... Xs [a(Xm)]' Since a(Xj) is an isogeny
(x) E D X ), [a(x))J is finite over S. This shows the finiteness of EIs, Since we
may assume that a(Xj) is an isogeny of degree prime to p for a given prime
p for all j, [B(x))] is unramified over its image in S if S is a scheme over a

discrete valuation ring of residual characteristic p > O. This shows that E is
unramified over its image in S. 0
We can extend the above result to a semi-simple D by an induction of the

number of simple components of D, which is slightly more technical.
We fix a base Dedekind domain V in a number field. Suppose that for
each maximal ideal P of V and its residue field k(P), OD I8lv k(P) is semi-
simple. We take an OD I8lz V-module V which is V-locally free of finite rank
n. Choose a base Vi,"" Vn of V I8lv k(O) over the field of fractions k(O) of V,
and write p(a) E Mn(V) for the matrix representation of a E OD given by
(avl, ... , avn ) = (Vi, ... , vn)p(a). For each a E OD, we define the character-
istic polynomial P(a; t) = det(t . In - p(a)) E V[t]. Suppose that for a field
which is a V-algebra, we have an OD I8l k-module Vi of dimension n on which
a E OD has characteristic polynomial P(a; t) for all a E OD. Since repre-
sentations of a semi-simple algebra over a field are characterized by the trace
of representations (see [MFG] Theorem 2.6 or [BAL] VIII.6.5, VIII.13.3), we
find that Vi ~ V I8lv k as OD I8lz V-module.
For each semi-abelian scheme Q/s with OD Y End(Q/s) as above, the
tangent space at the origin Lie(Q) is a module over OD I8lz as. Since Q is
smooth over S. Lie(Q) is a locally free Os-module of rank dims Q (d. [NMD]
Proposition 2.2.5). If Sis Spec(k) for a field k and if OD I8lz k is semi-simple,
there are only finitely many isomorphism classes of OD I8lz k-modules of a
given dimension (d. [BAL] VIII.6.5).
Let S be a V-scheme and Q/s be a (generically abelian) semi-abelian
scheme. We consider for an S-scheme T the following condition.
(det) We have B : OD y End(QT/T), and the characteristic polynomial of
B(a) for all a E OD on Lie(QT) over aT is given by the image of
P(a; t) in OT[t].
We consider the functor from S-schemes into SETS given by
T H [(QT, B : OD y End(QT)) I B satisfies (det)] . (6.6)
Since the property (det) is compatible with base change, the above functor is
a well-defined subfunctor of the functor in Corollary 6.10.
Supposing that Q/s is an abelian scheme with a symmetric polarization
A: Q --+ tQ (t A = A), we introduce a sub functor of (6.6): We have the Rosati
involution a H a* := A-iota 0 A of End!Q(Q/s) = End(Q/s) I8lz Q. We
suppose having an involution p of D and consider the following subfunctor
T H [(QT, AT, B) as in (6.6)IB(a P ) = B(a)* for all a E OD] . (6.7)
On each irreducible component I of the scheme representing the functor

of Corollary 6.10, the characteristic polynomial of a E OD on Lie(Qry) for
the generic fiber Qry gives a polynomial PI(a; t) E OI[t]. For all points x E I,
6.2 Quotients by PGL(n) 269
the specialization of PI (a.; t) at x gives the characteristic polynomial of a. on

Lie(Qx), because Lie(QjI) is locally OI-free of rank dims Q. The functor in
(6.6) is therefore represented by the union of irreducible components (charac-
terized by (det)) of the scheme representing the functor in Corollary 6.10. In
particular, the functor in (6.6) is representable by a scheme Mover S whose
connected components are finite over S.
We now study the representability of the functor in (6.7). We have an
involution [p] of the functor of (6.6) induced by (QT, 8) c-+ (QT, * 0 8 0 p). Thus
[p] acts on the S-scheme M representing (6.6) as an involution (still denoted
by [pl). The fixed points by this involution [p] form a closed subscheme Mp =
M XM,lM,[p] M of M. Over M p , 8 satisfies the requirement 8(o. P ) = 8(0.)*; so,
Mp represents the functor (6.7) (although Mp could be empty depending on
the choice of p). Thus we get
Corollary 6.11 Let the notation and the assumption be as. in Corollary 6.10.
Suppose that S is a scheme over a Dedekind domain V in a number field.
Suppose that OD 0v V 1m is semi-simple for every maximal ideal m of V.
Then we have
(1) The functor of (6.6) is represented by a scheme over S;
(2) If QjS is an abelian scheme, the functor (6.7) for an involution p of D is
represented by a scheme over S.
Each connected component of the scheme representing (6.6) and (6.7) is finite
over S, and if V is a valuation ring, it is finite unramified over S.
6.2 Quotients by PGL{n)

A given abelian scheme A with a polarization is projective by the existence of
an ample line bundle I:- / A giving the polarization. Fixing a type of polarization,
we may bound the rank of HO(A, 1:-). Thus we get an embedding A y pm
for a fixed sufficiently large m independent of A (once the dimension of A
and the degree of the polarization are fixed). Hence abelian schemes over S
with polarization give a subset H(S) of Hilb p = (S). By changing the base
of HO(A,I:-), PGL(m + 1) acts on this subset preserving the isomorphism
classes of abelian schemes. Thus to construct the moduli scheme of abelian
varieties, we need to make the quotient space by PGL(m + 1). Although this
is an oversimplified description of the construction of the moduli, it is obvious
from this that we need to study the quotient of quasi-projective schemes
by projective linear transformations. Here we reproduce an ingenious way of
Mumford and Tate (in [GIT] Chapters 3-4), constructing a large open set Ust
in (pm)M, which has a structure of a PGL(m + l)-torsor under the diagonal
action of PGL(m + 1) (see [GME] 1.8.3 for torsors). We later show that the
subset H is actually a scheme and can be embedded into (pm)M for sufficiently
large m and M preserving the action of PG L( m + 1); so, the base of the torsor
H gives the desired moduli space.
6.2.1 Line Bundles on Projective Spaces
We want to determine line bundles on a given projective space pn up to iso-

morphisms. For any given scheme S, we write Pic( S) for the set of isomorphism
classes of invertible sheaves on S. For any given morphism f : X -+ S, we
consider the relative Picard functor PicxIs(T) = Pic(XT )/ fT(Pic(T)), where
h : X T = X xsT -+ T is the base change of f. Let us write P /z for the pro-
jective space Proj(Z[Xo, ... , X n ]). Since pn has a section 0 = (1,0, ... ,0),
we can identify Picpn IZ with Ker( 0*); so, Picpn IZ is local. In other words,
C E Picpn Iz(T) is determined by its restriction CIT, for any faithfully fiat cov-
ering {Tt -+ Th of finite presentation. Suppose that T = Spec(R). We take
a graded R[Xo, ... ,Xn]-module M with the associated sheaf M E Pic(P T).
Then M = EBn>-m Mn and hence fiber-by-fiber, HI of M 0 O(m) vanishes
(see [ALG] IlLS). For 7r : P T -+ T, 7r*(M 0 O(m)) is a locally free sheaf
([ALG] Theorem IlI.12.11). Since M t for a closed point t E T is isomorphic to
O(j) for some j, we may assume that j = 0 (suitably choosing m) and hence
7r*(M 0 O(m)) ~ C for an invertible sheaf CIT. Then 7r*(M 0 0(m)07r* C- 1 )
is canonically isomorphic to OT, and hence, fiber:Qy-fiber trivial, which im-
plies that it is globally trivial. This shows that M ~ 7r*(C) 00(m) for a
suitable C E Pic(T) and m E Z, and
(6.8)
6.2.2 Automorphism Group of a Projective Space
We consider the affine scheme A /z = Spec(Z[Xl"'" Xn]) on which the alge-

braic group GL(n) acts by linear transformation. Recall that GL(n)/z is the
open sub scheme Spec(Z[XtJ' det(~,)) of A/~ = Spec(Z[XtJ])' Similarly, we
define PGL(n) by the open subscheme of pn2_1 given by removing the closed
subscheme defined by Proj(Z[XtJ]/(det(XtJ ))). Then GL(n) and PGL(n) are
group schemes, and the sequence: 1 -+ G m -+ GL(n) -+ PGL(n) -+ 1 is exact
as fppf group sheaves (see Section 8.2.1 for a definition of fppf sheaves). The
action induces automorphisms of p n - 1 ,
Proposition 6.12 If S = Spec(R) is connected, PGLn(R) = Aut(P;R 1).
Proof. By definition, we have PGLn(R) C Aut(P;R 1); so, we need to prove

that a E Aut(P;R 1) is induced by an element of PGL(n). By (6.8), we see that
a*(O(1)) = 0(m)07r*(C). Since 0(1) among invertible sheaves 0(m)07r*(C)
is characterized by the fact that Hq (P;R 1, O( 1)) vanishes if q =I- 0 and
HO(P;R\ 0(1)) ~ R n ,
the same fact for a* (0 (1)) shows that C is trivial and m = 1. Thus a induces
an automorphism of 0(1), which is an element of GLn(R). 0
6.2.3 Quotient of a Product of Projective Spaces
Following an ingenious trick of Mumford and Tate [CIT] Chapter 3, we now

create a large open subset U st of P = (pn)m+1 (for m = n + v with v >
0) stable under the diagonal action of PGL(n + 1) which has a geometric
quotient Z. We write X;') for the homogeneous coordinates of the ith factor
pn in (pn)m+1; thus, i = 0,1, ... , m and j = 0, 1, ... , n. We define .c, by the
pullback of Opn(l) to the ith factor of P = (pn)m+1.
Definition 6.13 For a subsequence a = (ao, a1, ... , an) of (0,1,2, ... , m),
we define Do. = det (x,(o.J)) E r(p, .co.), where .co. = ®;=o .co. J.
O~',J~n
For each a J , GL(n + 1) acts on the column vector t(Xcio. J), xio. J), ... , X~o.J))
by matrix multiplication; thus, Do.b 0 x) = detb)Do.(x) for 'Y E GL(n +1).
Thus Do. is invariant under GL(n + 1) up to units.
Proposition 6.14 Let {Sl, S2, ... ,Sv} be a collection of subsets of the set
{O, 1, ... ,n} such that
(1) SJ and U~:; S, have only one common number p,(j) for all 2 ::; j ::; v,
and we define p,(I) to be the least integer in Sl;
(2) {O,I, ... ,n}=U;=lSJ.
Define an affine open subscheme U c P by
°°
(a) DI i= for 1 = (0,1, ... ,n);
(b) D I (,) i= for all i E Sk and for all k = 1,2, ... , v, where
l(i) = (1- {i}) U {n + k}.

Then we have a PGL(n + 1)-equivariant isomorphism:
US=! PGL(n + 1) XSpec(Z) A nv-n
for the affine space A nv-n of dimension nv - n, where PG L( n + 1) acts on

the right-hand side by multiplication on the left factor PGL(n + 1).
We reproduce the proof given in [CIT] Chapter 3, Section 1.
Proof. Since at least one of the equations DI(i) involves variables of a given
piece of pn, the open sub scheme U is affine, because one could think of P as
the quotient of (A n+1 - {O} )m+l by G;:;::+l by component-wise multiplication
on each piece A n+1 - {O} (a sort of "Proj" of the graded algebra indexed by
multiple degree with values in zm+1).
Identify A nv-n with Spec(R) for
R_
77[ (n+1) , ... , Xn(n+l)., Xo(n+2) , ... , Xn(n+2)..
a. Xo
(n+v)
, ... , Xo
(n+v)]
, ... , Xn
- ((n+k)
x, - 11·Z E S k, k - 12
- " ... )
,v .
To make this legitimate, we count the number of relations: x~n+k) -1 for each
i E UjSj = {0,1,2, ... ,n}. Hi = M(k), the index i produces two equations
for k = 2, ... , v, and the number of equations is (n + 1) + (v - 1) = n + v.
The above spectrum is a linear subspace in (n + 1) v-dimensional affine space
defined by n + v independent linear equations; so, it is isomorphic to A nv-n.
We then define an embedding ~ : Anv-n y U as follows. For a graded
algebra A = EBk Ak and a homogeneous element a E A+ = EBk>oA k , we write
°
D+(a) = Spee(A[~]o) c Proj(A), where the subscript indicates the degree°
component. Writing Pi : p -+ pn for the ith projection, the composite
~, = Pi 0 ~ which has values in D+(Y) for Y E {X6 i ), ... , X~i)} is determined
by an algebra homomorphism ~i : Z[X6') /Y, ... , X~i) /Y] -+ R.
When i ::; n, we take Y given by Y = X?), and ~i is constant taking
X)') / X,(i) to the Kronecker symbol Dij. When i > n, then we have i - n = k for
some k = 1,2, ... , v, and we choose i, : Spee(R) -+ pn sending the coordinate
(Xo(n+l) , ... , Xn(n+l) ; Xo(n+2) , ... , Xn(n+2) ; ... ; Xo(n+v) , ... , Xn(n+v)) 0 f S pee (R) to
the homogeneous coordinate (x6i ), ... , x~)) of pn. This is legitimate because
for j E Sk, we have x~,) = 1 in Spee(R). This defines a closed immersion of
Spee(R) into U, and ~i(Opn(l)) is the structure sheaf R of Spee(R).
Since Do is invariant (up to the scalar) under the action of GL(n+ 1), U is
stable under PGL(n + 1), and we can extend ~ to PGL(n + 1) x Spee(R) 4 U
by ~(g, x) = g(~(x)).
To define the inverse ¢ of ~, for each given x E U, we need to show that
we can find a unique 9 E PGL(n + 1) so that for i ::; n, Pi(g(X)) is the origin
o in D+(X,(')) = An with homogeneous coordinate (0, ... ,0, to, ... , 0). The
the homogeneous coordinates of Pi (X) for °: ;

uniqueness, up to right multiplication by diagonal matrices, is plain, because
i ::; n forms an (n + 1) x (n + 1)
matrix 9 with determinant Dl(X) which is invertible on U.
We need to check that g-lx is in the image of ~ and to be able to choose 9
without ambiguity of diagonal matrices. For that purpose, we introduce some
notation. We have defined Sk n (Sk-l u··· u Sd = {M(k)} for k ~ 2, and M(l)
is the least integer in SI. We write K( i) for each i = 0, i, ... ,n the least integer
with i E S,,(,). By definition, i and M(K(i)) are both in S,,(,). Repeating this,
for e = MO K, eJ(i) E S,,(O-'(')). Writing i >- j if K(i) > K(j), we have
i >- e(i) >- e2(i) >- e3(i) >- ... >- ek(i) = M(l) for some k.
Then we define (Ao(X), Al (X), ... , An (X)), by induction on j,
AJL(1)(X) = 1 if j = 1 and AJ(X) = A£(J) (x) D~(R(J)((~) if j > l.

1(j) X
Over U, AJ(X) is a well-defined section of Lj ® L~(~)"

We finally define ¢ : U -+ PGL(n + 1) x Spee(R) as follows. Writing
two projections of ¢ as ¢p : U -+ PGL(n + 1) C Proj(Z[T'J]oS',Jsn) and
cPR : U -+ Spec(R) , we define cPP by cP'P(Opcd 1)) = .e1-'(1) and cP'P(T'j) =

X,CJ) ® A;l; here Opcd1) is the restriction of 0(1) of Proj(Z[T'J]o:S',J:Sn) to
the affine open subset PGL(n+1). By definition, cP(x) = g·diag[Ao, ... , An]-l
for g = (Po(x), ... ,Pn(x)); so, we have now specified g without any ambiguity.
The point here is that we have now made the choice unique (not just up to
multiple by diagonal matrices). We then define
cP'R(x(n+k))(x) = D 1 (,)(x) . Ai(X) .

, D 1 (I-'(k)) (x) AI-'(k) (x)
It is easy to check that cPR has values in Spec(R). Since cPp(x) and cPR(X)
contain all the information necessary to compute the coordinates of x, it is an
immersion. By a simple combinatorial computation, basically by definition,
we check that cP 0 " is the identity map of PGL(n + 1) x Spec(R); so, we
conclude that" is a surjective isomorphism. 0
Definition 6.15 U st C P is an open subscheme whose geometric points x =

(x(O) , ... ,x(m)) (m = n + v) with x(J) E pn satisfy
number of points XCi) in L dim L +1

--------~---------- < ----~-
m+1 n+1
for every proper linear subspace L C pn(k(x)).
A point x E P is called stable if x E Ust .

We need to verify that Ust is an open subscheme. Let I C {O, 1, ... , m}
and choose an integer no with 0 ::; no ::; n - 1. Suppose that
-III no + 1
->- -.
m+1 - n+1
Then we see P - Ust = U1 , n 0 V(I, no), where
V(I, no) = {x E PI (x(')) 'EI spans a linear subspace of dimension::; no} .
Since the matrix X made of homogeneous coordinates of xC,) for x E V(I, no)
has rank at most no, all the (no + 2) x (no + 2)-minors of X have determinant
O. Thus V(I, no) is a closed subscheme and, hence, Ust is an open subscheme.
Proposition 6.16 A geometric point x E Ust if and only if there are integers
N ~ No > 0 and monomials Po, ... , Pm of Da as in Definition 6.13 such that,
for 0::; i ::; m, P, E r(p,.er! ® ... ® .et;'-No ® ... ® .e!;/.) and P,(x) 7"= o.
We repeat the proof of Mumford ([CIT] Proposition 3.6).
Proof. We associate with the monomials D = IL D~(a) an integral point

d(D) E ]Rm+l = V in the following way. the ith coordinate of d(D) is
L a 3' e(Q). For x E P, define ex

c V to be the simplicial cone spanned
by d(Da) for Da(x) of- 0; that is, ex
is made up of a positive linear com-
bination of d(Da) with Da(x) of- O. Thus d(D) E ~ ex D(x) of- O.
Writing 1 = (1,1, ... ,1) E V, we see that 1 E ~ ex we have
N1 - v ex
E with a sufficiently large N for any given v
E V. Thus tak-
ing v = No(O, ... , 0,1,0, ... ,0) for No in the proposition, we need to show
that 1 E ex~ x E Ust . Then by the lemma following this proof, we need
to prove x E Ust if and only if for any given partition {O, 1, ... , m} = I U J,
there exists Q such that D a (x) of- 0 and
III· IQ n JI < IJI . IQ nIl· (6.9)
Suppose {X(')hEI spans a linear subspace L of pn. Then Da(x) of- 0 implies
the maximum number of Q, in I has to be dim L; so, (6.9) holds (in addition
to Da(x) of- 0) if and only if
III· (n - dimL) :S III·IJ n QI < IJI·IQ n II :S (m + 1 -III)(dimL + 1)

which is equivalent to the inequality in Definition 6.15 of stability of the point
x. 0
We need to prove the following lemma.
Lemma 6.17 Let the notation be as in Proposition 6.16 and its proof. We
have 1 f/- ex
if and only if there exists a partition of sets: {O, 1, ... ,m} = I U J
such that the inequality (6.9) fails to hold for any Q with Da(x) of- O.
Proof. We have 1 f/- ex if and only if there exists an integral linear form
£ : zm+1 -+ Z such that
£(1) = 0 and £(d(Da)) :2': 0 for all Q with Da(x) of- O. (6.10)
Since 1 is fixed by any permutation of coordinates, we may assume that

AO :S A1 :S ... :S Am if we write £(v) = L, A,V, for v = (v,) E jRm+1 = V.
Then the condition (6.10) is verified for £ given by
AO = A1 = ... = A(3 = -(m - (3) and A(3+1 = ... = Am = 13 + 1.
Once this is proven, we just take I = {f3 + 1, ... ,m} and J = {O, 1, ... ,f3},
and the assertion follows.
We prove that the above expression of £ is possible if (6.10) holds. We may
assume that Da(x) of- 0 and £(Da) = L, Aa, is minimal; so, we may assume
that x(O), ... , X ((3) -1) span a linear subspace of dimension j - 1 but X ((3) ) does
not lie in this linear subspace for all j = 1,2, ... ,n. Thus x(O), ... ,x((3n) span
entire pn, and hence D(3(x) of- 0 for 13 = (0,131, ... ,f3n), and
Da(x) of- 0 ~ Qo:2': 0, Q1 :2': 131,'''' Qn :2': f3n.

Hence the condition (6.10) is equivalent to £(d(D(3)) :2': O. Since the set of £s,
C = {(AD, AI,·· ., Am) E zm+1 I AD :s; Al :s; ... :s; Am and LA, = O} ,
is convex, if e( d( D {3)) :::: 0 holds in the interior of C, it has to be true at least

for one face, which is determined by the partition I U J = {O, 1, ... ,m} so
that A, = -(m - (3) if i E J and AJ = (3 + 1 if j E J. 0
Corollary 6.18 Let the notation be as in Propositions 6.14 and 6.16. Suppose
that x E P = (pn)m+l (for m > n) is stable. Then there exists a sequence of
subsets {51, 52, ... , 5 v } for v = m - n satisfying the conditions in Proposi-
tion 6.14 such that x E unust , where U is the open subscheme defined by the
sequence as in the proposition. Moreover, there is an open neighborhood Uo of
x in Un Ust defined by P o:J. 0 for a monomial P in the Das that is a section
of (.co ®.c 1 ® ... ® .cm)M for some integer M > O.
Proof. Choose N > No > 0 and choose Po, ... , Pm as in Proposition 6.16. Let
Q = Q9D (x)#O Do.. Then P = QNo ® p;o ® ... ® P:;,= for a suitable rJ is a
Q
section as in the proposition. Then P o:J. 0 defines an open neighborhood Uo by

Proposition 6.16. We now construct a sequence 5J of subsets of {O, 1, ... ,n}
so that x E U. Since x is stable, there do not exist disjoint proper linear
subspaces L' and L" of pn(k(x)) such that xC,) is either in L' or L" for all
i. Thus all the points xC,) cannot lie in a hyperplane; so, we may assume
that x(O), ... , x(n) cannot be in a hyperplane (b'y reordering coordinates).
Furthermore, we may assume that x;') = o'J by moving around by the action
of PGL(n + 1). Define sets 5~ (k = 1,2, ... , v) by the set of integers i such
that D O,l, ... ,,-l,,+l, ... ,n,n+k(X) o:J. 0; that is, x(n+k) is not in the hyperplane.
We prove that there is no partition {O, 1, ... ,n} = T' U Til such that either
S~ C T' or 5~ C Til for all k. If this is proven, by shrinking 5~ a little if
necessary, we can achieve the condition of Proposition 6.14.
Suppose on the contrary that 5~ C T' or 5~ C Til for all k. Define linear
subs paces L' (resp. L") by X, = 0 for all i E T' (resp. i E Til). Then every
point x(k) is either in L' or L", contradicting the stability of x. 0
Covering Ust by the open sets {U,} as in Corollary 6.18, we have U,

PG L( n + 1) X Z,. Thus on U, n UJ ' Z, xu, (U, n UJ ) is canonically isomorphic
to ZJ x uJ (U, n UJ ); so, the schemes ZJ glue into a scheme Z with projection w :
Ust --+ Z. By our construction, over U = Ui U21 Ust Xz U = PGL(n + 1) x Z.
Thus Ust is a PGL(n + l)-torsor (see [GME] Section 1.8.3 for a definition of
torsor), and hence (Z, w) is the geometric quotient of Ust (see [GIT] Section
0.6 or [GME] Section 1.8.3 for the definition of geometric quotients). The geo-
metric quotient always gives a categorical quotient (see [GIT] Proposition 0.1
or [GME] Proposition 1.8.1).
Thus we get
Theorem 6.19 (Mumford) The pair (Z, w) is the geometric and categorical
quotient of Ust by PGL(n + 1) and is quasi-projective.
6.3 Mumford Moduli
We describe the Mumford construction of the moduli over Z of abelian schemes

of dimension n with a given polarization of degree d2 .
6.3.1 Dual Abelian Scheme and Polarization
Let A/ s be an abelian scheme. We consider the following Picard functor
PicA/s(T) = Pic(AT)/f;'Pic(T) = Ker(OT)
for f : T -+ S. It is known that PiCA/S is represented by a (locally Noetherian)

reduced group scheme (Grothendieck [FGA] Exp. 232, 1961/62, [ABV] Section
13, or [NMD] Chapter 8). Let t A be the identity-connected component PiCA/S
of the group scheme representing PicA/so Then t A/ s is an abelian scheme.
Let .cuniv E tA(tA) = PiCAxstA/tA = Ker(OAxstA) corresponding to the
identity. Then the sheaf .cuniv has the following universal properties:
• 0AxstA.cumv = VtA;
• Let T -+ S be an S-scheme. For any invertible sheaf L on AT algebraically
equivalent to V AT' there exists a unique morphism t L : T -+ t A such that
(idA x td*.c umv = L.
Let L be an invertible sheaf on A. For x E A, we define the translation
Tx(Y) = x + y, which is an automorphism of A. Then (T; L)®L -1 is an element
in t A, and we obtain a morphism A(L) : A -+ t A. This S-homomorphism is
an isogeny (i.e., surjective) if and only if L is ample (i.e., choosing a base of
HO(A, L':i9 n ) for sufficiently large n, one can embed A into PIs over S; see
[ABV] Section 6). The degree of the polarization is defined to be the square
root of the degree of the homomorphism A(L).
There is another construction of A(L). Consider a*(L) ®Pi(L) ®p2(L)-l
as an invertible sheaf on AA = A x s A, where a : A x A -+ A is the addition
of the group scheme A. Then this invertible sheaf induces an A-valued point
of PicA/s(A), which factors through tA, because at the identity, this sheaf
specializes to the trivial invertible sheaf at the origin 0 (so, the image is in
the connected component of PicA/s), We claim this A-valued point of tA is
actually A(L). Indeed, specializing this sheaf to (x : S -+ A) E A(S), we get
T;(L) ® L -1 ® x*(L), which is equivalent in Pic(A)/ f* Pic(S) to T;L ® L -1.
6.3.2 Moduli Problem
Fix positive integers n, d, and N, and define a functor Ad,N : SCH/z[-hl -+

SETS by
Ad,N(S) = [(A, ¢N : (Z/Nz)2n ~ A[NJ, '\)/s] ,
where
6.3 Mumford Moduli 277
(AI) A/ s is an abelian scheme with dims A = n,

(A2) ¢N is an isomorphism of group schemes over 5, and
(A3) .\ is a polarization, locally .\ = A(£) on 5 and deg(.\) = d2 .
It is known (see [GIT] Proposition 6.10) that if .\ is locally of the form
A(L), then 2,\ is globally A(LL1(.\)) for the invertible sheaf LL1(.\) given by
(IA x .\)*(£univ). Here is a theorem of Mumford.
Theorem 6.20 (Mumford) There exists a quasi-projective scheme 9J1d,N
over Z [*"] such that
(1) For any geometric point s = 5pec(k) of 5pec(Z[*"]) , Ad,N(S) ~ 9J1 d,N(S);
(2) If N 2': 3, we have a universal object (X, <PN, A) /9Jl d • N such that for each
triple (A, ¢N,.\) E A d,N(5) there exists a unique morphism [ : 5 -+ 9J1 d,N
with (A, ¢N,.\) ~ (X, <p, A) X9Jl d N 5 = [*(X, <p, A), and the above asso-
ciation (A, ¢ N, .\) / s f--t [ induce·s a functorial isomorphism Ad,N (5) ~
9J1 d,N(5), where 9J1 d,N(5) = Homz[11J(5,9J1d,N);
(3) If N 2': 3, 9J1d,N is smooth over Z[rJd].
When d is prime to N, the pairing A[N] x t A[N] -+ J.LN composed with the po-
larization gives a self-duality eA,N : A[N] x A[N] -+ J.LN. Recall J = (l~ -~n),
fix a primitive Nth root of unity (N, and impose one more compatibility con-
dition in addition to (AI-3) to define a subfunctor A{,N C A1,N.
(A4) We have eA,N(¢N(X), ¢N(Y)) = (';;.Jy for all x, y E (ZjNz?n.

t J'
Since eA,N pulled back by ¢N can be written as (x, y) f--t ( ; . y for a unique
nondegenerate alternating matrix J', 9J1 N = 9J1 1 ,N is a disjoint union of 9J1{N
classifying triples (A, ¢N,.\) satisfying (AI-3) and the condition obtained from
(A4) replacing J by J'. Thus fixing e = (1,0, ... ,0) and e' = (0, ... ,0,1) in
(ZjNz)2n, we have a well-defined morphism A{,N -+!!:.N sending (A, ¢N, .\)/B
to eA,N(¢N(e), ¢N(e')) E J.LN(B), which in turn induces a morphism of
schemes 9J1{,N -+ Z[(N, *"] for the primitive root of unity (N = eX,N(e,e').
Here X/ 9Jl j(, is the universal abelian scheme. This shows that the scheme 9J1/v
is well defined over Z [(N, *"]. Looking at the fiber over C of 9J1~ /Z[ (N, 11 J' by
an argument similar to the Hilbert modular case (see Section 4.1.3 and the
proof of Theorem 6.26), we find 9J1/v(C) ~ r(N)\5)n for the Siegel upper
half-space 5)n = {tz = z E Mn(C) I Im(z) > O} and
r(N) = {a E 5P2n(Z)ia == 12n mod N} .
By the existence of a smooth toroidal compactification ([DAV]; see also Sec-

tion 6.3.6 in the text), 9J1/v is geometrically irreducible over Z[(N, *"] fiber-
by-fiber.
Remark 6.21 If d is prime to N, choosing well a nondegenerate alternating

matrix J' in M 2n (ZjNZ), we could impose
t J'
(A4') We have eA,N(¢N(X), ¢N(Y)) = (;. Y for all x, Y E (7LjN7L)2n.
We then have a smooth geometrically irreducible scheme 9Jt;{N over 7L[(N, d~]
representing the functor classifying triples (A, A, ¢N) satisfying (Al-3) and
(A4'). We apply Corollary 6.11 to the universal abelian scheme XI'JJlJ' and
d,N
OD = OF to verify the represent ability of the functor of (r(N)) in

£r(N)
Section 4.1.2 by a scheme 9Jt(c, r(N)) as follows. First of all, the condition
(det) applied to the integer ring 0 of F is equivalent to (rm4) in the definition
of the AVRMs. Since we can choose the polarization ideal c in its strict ideal
class, we may assume the degree of the polarization and c itself is prime to the
given integer N ;:::: 3. Choosing a 7L-base of 0 EB c* and identifying L = 0 EB c*
with 7L 2n (n = [F : Q]), the polarization pairing L II L ~ c* ...:!':+ Q gives
rise to a pairing (x, y) H t x · J'y for a suitably chosen alternating matrix
J' E M2n(7L) as above. By Corollary 6.11, 9Jt(c, r(N)) is realized as a scheme
finite over 9Jt{N; so, it is affine over 9Jt{N and is quasi-projective. All other
Hilbert-Blum~nthal moduli schemes 9Jt( c: r) there for different level structure
r can be obtained from 9Jt(c, r(N)) by making a quotient by a finite constant
group. So the represent ability claimed in Section 4.1.2 is now verified.
We now give a sketch of Mumford's proof of the above theorem. Let A ls

be an abelian scheme over a ring 8. The key idea is that for a given very ample
invertible sheaf LIA' the embedded image of A under L in pm for m + 1 =
ranks H O(A, L) is determined by the choice of basis b of H O(A, L). In other
words, the image is transported to the other by an element of PGLm+l(8).
Since for an abelian scheme, by the generalized Riemann-Roch theorem (see
[ABV] Section 16), we can compute the Hilbert polynomial P of L, the moduli
functor of (A, b) is a sub functor of Hilb~=. Proving that the image is a quasi-
projective subscheme H of Hilb~=, the (coarse) modulus is constructed as
9Jt 1 = PGL(m+ l)\H. We give more details of this argument in the following
subsections. The proof finishes at the end of Section 6.3.5.
6.3.3 Abelian Scheme with Linear Rigidification
Let (A, ¢N, A)ls E Ad,N(S). We consider the invertible sheaf LIs given by
the direct image of L~(A)3, which is very ample, because .c 3 is very ample
if .c is ample (see [ABV] Section 17) and ample ness of L~(A) follows from
A(L~(A)) = 2A as we already remarked. Let S(L) be the symmetric algebra
of L (as in Section 6.1.1) and write P(L) for the projective bundle Proj(S(L))
which is a projective scheme over S locally isomorphic to Pis' A linear rigidi-
jication is an isomorphism L : P (L) ~ Pis. Thus L is determined by the choice
of a base of L up to scalar multiplication. Since the very ample sheaf L~(A)3
on an abelian scheme satisfies ([ABV] Section 16)
• L = f*(L~(A)3) is locally free of finite rank 6nd, and
• R2f*(L~(A)3) = 0 if i > 0,
the formation of f*(y1()..)3) as above commutes with base change (cf. [ALG]
III.12.11). Thus the association 5 >--+ A~,N(5) = [(A,¢N,)..,L)/5] is a well-
defined contravariant functor. The embedding J : A I 5 '--+ P (L) ~ P I5
deter-
mines the sheaf V:1()..)3 = I*Op(L)(l), which in turn determines).. because
A(L) = 6).. (PicA15 FA is torsion-free; see [ABV] Section 19 Corollary 2). Hav-
ing ¢ N is equivalent to having 2n (linearly independent) sections CTJ = ¢ N (e J )
of A over 5 for the standard base {e1, ... , e2n} of (71,/ N71,j2n. We write CTO = 0
for the identity section 0 of A. We record here what we have seen.
Proposition 6.22 The data (A, ¢N,).., L) are completely determined by the
embedding data (1 : A I 5 '--+ PIS' CTo, CT1, ... , CT2n).
In other words, defining another functor 1-Ld,N(5) : 71,[*"]-5CH --+ SETS by
1-Ld,N(5) = [(1 : A ls '--+ PIS' CTo, CT1,···, CT2n)/5] ,
we have an isomorphism of functors: A~,N ~ 1-Ld,N. Here CT1, ... , CT2n are sup-
posed to give 2n linearly independent elements of A [N]( 5) .
6.3.4 Embedding into the Hilbert Scheme
For simplicity, we just write P for Pillli" We write So for 5pec(71,[*"]). We con-
sider the functor Hilb~ associating with each 5 the set of closed subschemes
of PI 5 flat over 5 with Hilbert polynomial P. As we have already seen, this
functor is represented by a scheme H = Hilb~. Write Z --+ H for the uni-
versal flat family inside PI H with Hilbert polynomial P. For each subscheme
V c PI 5 flat over 5 having Hilbert polynomial P, we have a unique morphism
h: 5 --+ H such that V is given by 5 XH Z C P 15 over 5.
By the generalized Riemann-Roch theorem ([ABV] Section 16), the Hilbert
polynomial of (A, L) (or of J(A)) is given by P(T) = 6n d . Tn. Thus the image
J(A) induces a unique So-morphism h : 5 --+ H such that J(A) = 5 XH Z as
subschemes of P.
k
Let Hk = Hilb~,k = 'z XH Z x:'--.-.-.-x-H-i. By the very definition of the

fibered product, we get
Thus Hk classifies all flat closed subschemes with Hilbert polynomial P having
k sections over 5. The projection p: Z(k) = Z XH Hk --+ Hk (p(zo X z) = z)
gives the universal scheme and has tautological k sections
This shows that 1-Ld,N c Hk for k = 2n + l. For simplicity, write Ho for Hk.
Since "smoothness" is an open condition (because it is local; in other words,
smoothness at a point x of a morphism f follows from formal smoothness of

the local ring at x over the local ring at f(x); see [NMD] Section 2.2), there
is an open sub scheme Hl of Ho over which Z is smooth. Then Hl represents
smooth closed subschemes in P with Hilbert polynomial P and k sections.
Now we use a result of Grothendieck. Abelian varieties have rigidity such
that in a smooth projective family X -+ 8 for connected locally Noetherian 8
with a section 0 : 8 -+ X, if one fiber is an abelian variety with the identity
section induced by 0, X itself is an abelian scheme ([GIT] Theorem 6.14).
This shows that Hl has a closed subscheme H2 over which Z2 = Z(k) XH 1 H2
is an abelian scheme with the identity section 0 induced by 0"0.
Rewrite 7j (j = 1, ... , 2n) for the universal2n sections of Z2 different from
O. We have a maximal closed subscheme H3 C H2 defined by the equations
[N] 07, = 0 for all i, where [N] is the multiplication by the integer N.
The relation E~:l aj7J = 0 for a given a = (a J ) E (ZjNZ)2n - {O} gives
a closed subscheme Ha of H3; so, we define H4 = H3 - Ua Ha. Thus the
abelian scheme Z4 over H4 has 2n linearly independent sections of order N.
Since Z4 is a subscheme of P / H 4 , it has the line bundle L = OZ4 (1) which
is the restriction of O(l)/p. Then we define Hs to be the maximal subscheme
of H4 such that p* L ~ LLl(A)3 for a polarization A : Z4 -+ t Z4, where p :
Zs = Z4 XH4 Hs '---+ Z4 is the inclusion. In other words, for any morphism
¢ : T -+ H 4, if ¢* L ~ LLl(A)3 for a polarization A: t Z4,T -+ Z4,T = Z4 XH4 T,
¢ has an image in Hs (i.e., ¢ factors through Hs). It is proved in [GIT]
Proposition 6.11 that the maximal subscheme H5 with the above property
exists and is closed in H 4 .
6.3.5 Conclusion
By the argument in the previous proposition, the functor Ad,N is represented

over 8 0 by a quasi-projective scheme Hs with the universal abelian scheme Z5
over H 5. The group PGL(m + 1) acts on H5 by t f-t tog (g E PGL(m + 1».
By the following result of Mumford, if N is sufficiently large, we can make the
geometric quotient by Theorem 6.19,
9'Jtd,N = PGL(m + 1)\H5 and Xd,N = PGL(m + 1)\Z5,
which gives the universal abelian scheme over 9'Jtd,N.

Proposition 6.23 (Mumford) Let A/k with projective embedding ¢ : A '---+
Pik be an abelian variety of dimension n over an algebraically closed field k.
Write 9 for the degree of ¢(A) C pm. Let N be a positive integer. Arrange all
points of order N of ¢(A) as fN(A) = (Xl, ... , XN2n) and regard it as a point
fN(A) = (¢(xd, ... , ¢(XN2n» of (pm)N2n (k). Suppose that the image of A in
pm is not contained in any hyperplane. If N > v(m + l)g with char(k) f N,
the point fN(A) E (pm)N2n is stable.
We repeat here the proof of Mumford in [GIT] Proposition 7.7.
Proof. We need to check the inequality of Definition 6.15 for the point fN(A).
Since the total number of the points is M = N 2n , we only need to prove
for any hyperplane H c pm. Write (Y, Z) for the intersection number of
complementary cycles Z and Y in pm ([ALG] 1.7 and Appendix A). Choose
a I-dimensional cycle "( = ¢(A) XP'" L for a linear subspace L c pm of
co dimension n - 1. We suppose that "( contains the origin of ¢(A). Since
multiplication [N] by N acts on divisors by N 2 ([ABV] Section 6, Corollary 3),
it also acts on the I-cycles by N 2 (by the Albanese-Picard duality; [ABV]
Section 23); so, [N]*"( = N 2"( up to numerical equivalence. Here we use the
fact that char(k) f N. Writing h = ¢-1(H) = ¢(A) Xp'" H, we have
N 2([N]*"(, h) = ([N]*"(, [N]*h) = [N]*("(, h)

= deg([N])("(, h) = M("(, h) = M deg(¢(A)) = M· g.
This shows
I{X1, ... ,XM}nHI < - 9 < -1-
.:...:...._-----"----'-
M - N2 m+l'
which finishes the proof. 0
Out of 0, T1, ... , T2n, we create the set of full sections 81, ... ,8M : H5 '-+ Z5
through linear combinations of T J • We embed H5 into (pm)M by sending
x E H5 to (81(X), ... ,8M(X)) E (pm)M. Since the image is contained in the
open set of stable points Ust C (pm)M fiber-by-fiber; so, it is contained in
Ust . Since Ust is a PGL(m + 1)-torsor over its geometric quotient (Z,7r) in
Theorem 6.19, we find the geometric quotient of H5 by PGL(m + 1) as the
image 7r(H5) of H5 in Z. Since we know the Hilbert polynomial of Z5,
we know the degree deg(Z5) (cf. [ALG] 1.7.6). Thus if N is sufficiently large
as in Proposition 6.23, the quotient VJ1d,N = PGL(m + 1)\H5 represents the
functor Ad,N over Z[-b] (see [GIT] Section 3 in Chapter 7 for more details).
Suppose now that N ?: 3. We choose two distinct large primes p and q
outside dN so that Ad,pN and Ad,qN are representable by the above argument,
respectively, over Z[p~] and Z[q~]' The finite (constant) group
acts on (A, A, ¢pN) by ¢pN f--t ¢pN 0 9 for 9 E GL 2n (Z/pNZ). Since we do

not impose (A4) just after Theorem 6.20, the full group GL 2n (Z/pNZ) acts
on the functor (and hence on the moduli) in place of the symplectic group
SP2n(Z/pNZ). Since Aut(A,A,¢N) = {I} if N?: 3 by a result of Serre (see
[ABV] Theorem 5 in Section 21), the action of the finite group GL 2n (Z/pZ) is
free. If we can cover Ad,pN by affine open subschemes stable under the action,
we can again make the geometric quotient
9J1d,N/Z[p\v 1 = GL 2n (Z/pZ)\9J1d,PN/Z[p\v l'
which represents Ad,N over Z[p~] (see [SGA] V.l, [ABV] Section 12, or [GME]
1.8.3). In order to give a sketch of a proof of finding an affine open cover of
9J1 := 9J1d,pN stable under the action, we consider the universal abelian scheme
X ~ 9J1 and ~ = det(7r*.ftx / 9J1 ). The line bundle ~ is ample (cf. [CSM] V.3, or
see the following subsection for another proof using a compactification of the
e
moduli). By the existence of nontrivial sections invariant under GL 2n (Z/pZ)
(given by theta constants) of ~®k for sufficiently large k, 9J1[~] is affine and
stable under GL 2n (Z/pZ). Since abelian schemes are embedded into a projec-
tive space by algebraic theta functions giving rise to theta constants of level
N if N ~ 3 (see [CSM] Appendix II), moving around e, 9J1[~] covers 9J1. Thus
we have the geometric quotient 9J1 d,N/Z[ p\v 1 as above.
By the universality, the two schemes 9J1 d,N/Z[p\vl and 9J1 d,N/Z[q\vl thus ob-
tained glue together uniquely over Z[p:N] giving rise to the scheme 9J1 d,N/Z[-b 1
representing the functor Ad,N over Z[ -tv]. Even if N ::::; 3, the above process
yields the coarse moduli scheme 9J1 d,N over Z[ -tv] (cf. [CSM] Theorem 1.4, or
[GME] 2.3.2 in the elliptic modular case).
Once represent ability of Ad,N is shown, as long as Aut((A,¢N,A)/iB) =
{I} for any prime p f dN, the formal completion of 9J1 d,N at the closed point
x E 9J1 d,N(iFp) carrying (A,¢N,A)/iF p is isomorphic to a power series ring
W[[t 2J ]h:'02:'OJ:,,;n for the ring W of Witt vectors with coefficients in iFp . This
follows from the deformation theory of abelian schemes (see [DAV] I.3 for a
summary, [CBT] Chapter V for details, and also Section Section 8.2 partic-
ularly 8.2.4 in the text). Thus 9J1d,N for N ~ 3 is smooth over Z[d~]. This
finishes the proof of Theorem 6.20.
6.3.6 Smooth Toroidal Compactification
Let us describe briefly without proof the smooth toroidal compactification

MN = M1,N of 9J1~/Z[(N,-bl = 9J1{,N and the universal semi-abelian scheme
Q/MN' following the book by Faltings and Chai [DAV]. We expect the n-
dimensional universal abelian scheme Xn/9J1;(, to degenerate over the gth stra-
tum Sg (0 < 9 ::::; n) of the cuspidal divisor to a semi-abelian scheme Q with
an exact sequence 0 ---t G~ ---t Q ---t X n - g ---t O. Thus we need to create Sg
from the moduli of extensions 0 ---t G~/ S ---t Q ---t A n - g / S ---t 0 for an abelian
scheme A n - g / s of dimension n - g.
In the Hilbert modular case, the situation is simpler, and the universal
AVRM degenerates into G m Q9 a* at the cuspidal divisor over the cusp (n, b).
The character group of G m Q9 a* is canonically isomorphic to a by
via the trace pairing (a, b) >--+ Tr F/Q1 (ab), and F = a Q9z R Indeed we started
(X)
our work in Section 4.1.4 with a simplicial cone decomposition of F:"+ on

which C,') is positive-definite. The term q~ in the q-expansion is identified
with the linear form OIl (ab)*; so, ~ E abo
By an analogy to X(G m Q9 a*), we take X = X(G~rJ and consider XJR =
X Q9z lR (which is the analogue of F(XJ)' For the Siegel modular forms (of
level 1), each term of their Fourier expansion at the infinity cusp is given by
e(Tr(~z)) for z E SJn with symmetric semi-integral matrices ~ in End(X). We
identify the space of such matrices with the symmetric second tensor product
S(X) = Sym 2(X) (S(X) is the degree-2 component of S(X) in Section 6.1.1).
The Z-dual S*(X) is the space of Z-valued symmetric bilinear forms on X.
Writing X R for XQ9 z R, we similarly define S*(XR) and S(XR)' In particular,
S* (XJR) is the space of symmetric lR-bilinear forms on X JR , and S(XJR) =
Sym 2(XJR). We take the cone C(XJR) C S*(XJR) made up of all positive semi-
definite bilinear forms. The interior Co (XJR) is the space of all positive-definite
bilinear forms on X JR .
For the moment, we assume that g = n (with the maximal toric part).
We take a cone decomposition C = {a}" of C(XJR) satisfying the following
conditions similar to (PC1 ~4):
(Pel) a is an open simplicial cone (a = {O} is allowed);
(Pc2) The cones in C are permuted under the action of GL(X), and
CjGL(X) is finite, where I E GL(X) acts on b(x, y) E S*(XJR) by
b(x, y) >--+ b(rx"y);
(Pc3) a is smooth (i.e., generated by a part of a Z-base of S*(X));
(Pc4) {a} is projective, that is, it is sufficiently fine so that the toroidal
compactification is projective (see [DAV] V.5.1 for an exact condition
for projectivity).
We reproduce here some results of the more detailed theorems in [DAV] IV.6.7,
V.2.3 and V.5.S on the smooth toroidal compactification we need later.
Theorem 6.24 (Chai-Faltings) Suppose n > 1 and N ::: 3. There exist
projective normal schemes MN = M~ and M'N over Z[ f:t, (N] containing Wl'/v
as an open dense subscheme satisfying the following five conditions.
(1) The natural action of SP2n(Z) on Wl'/v extends to MN and M'N. The
complement D(XJ = MN - 'lJl'/v is a relative Cartier divisor with normal
crossings. The scheme MN is smooth over Z[ f:t, (N].
(2) For NIN', the natural projection 'lJl'/v, --» 'lJl'/v extends to M N , --+ MN
(resp. M'N, --+ M'Nj, which is equivariant under SP2n(ZjN'Z).
(3) The universal abelian scheme X N over 'lJl'/v extends to a semi-abelian
scheme QN ~ M N , and QN' is the pullback ofQN if NIN'. The Kodaira~
Spencer map induces an isomorphism of Sym 2(7f*(S!9N/zl-h,(NI)) onto
S!}\[N/zl-h,(NI[dlog(D(X))] as locally free sheaves.
(4) Let ~ = 7r*DgNIZ [k,(N]. Then HO(MN,det(~)®J) ~ HO(9Jt)("det(~)®J)

by the restriction map -(Koecher principle), which vanishes if j < O.
(5) Set cBN = EBJ~o HO(MN , det(~)®J). Then cBN is a graded algebra finitely
generated over Z, and we have M'N ~ Proj (cB N ). In particular, MN covers
M'N, and MN is the normalization of a blowing up of M'N along a coherent
sheaf of ideals.
The theorem is proven first constructing MN as an algebraic stack using
only (Pc1-3). Then using (Pc4), an ample line bundle is created on the alge-
braic stack MN, and hence MN is a projective scheme. Even if N :::; 2, the
algebraic stack MN (and the semi-abelian stack 9N over MN) satisfies the
above condition of theorems (except for the smoothness), and (the reduced
scheme of) each stratum of the cuspidal divisor Doo is smooth over Z[*, (N].
The expression M'N = Proj (cB N) is valid for N :::; 2, and M'N is always a
normal scheme. The scheme M'N is called the minimal compactification (with
M'N(C) giving the topological Satake compactification) and MN is called the
smooth projective toroidal compactification.
Here is a very brief description of MN and some indication of its construc-
tion. Since we need the modulus variety 9Jt{ N for g-dimensional (principally
polarized) abelian schemes for g = 1, ... ,n i~ order to describe M N , we write
9Jt~) for the level N modulus 9Jt)(, of principally polarized abelian schemes of
dimension 9 (dropping the superscript J for simplicity). The reader is assumed
to be familiar with the notation of [DAV].
We start with the classification theory of Raynaud extensions (to see what
are the moduli of extensions 0 ~ G;;' ~ G ~ A n - g ~ 0). Since the algebraic
group G admitting its Shimura variety does not appear in this subsection, we
use the symbol G (following [DAV]) to indicate a Raynaud extension (this
convention is in effect only in this subsection; the symbol 9 is used to indicate
the quotient semi-abelian scheme of G by a unit action). Let S = Spec(R)
for an I-adically complete Noetherian integral domain R = ~kRjIk. We
put So = Spec(Rj 1). Start with an extension To "-+ Go -+> Ao over So of an
abelian scheme AOl50 by a torus T 0 150 . Pick a semi-abelian scheme GI5 with
reduction modulo I giving rise to To "-+ Go -+> Ao. Writing Soo = Spf(R) for
the formal completion of S along So, it is known (cf. [NMD] 7.3-4 and [CSM]
Chapter II) that To is uniquely lifted to a formal torus f c a, where a is
the formal completion of G along Go. The quotient A = ajf is a smooth
formal group, which is proper because otherwise Ao has to have an affine
subgroup scheme. Thus A mod J1 for all j > 0 is an abelian scheme over
Spec(RjIJ) = SJ" We call such a formal scheme a formal abelian scheme.
When there exists a subtorus TI5 c G I 5 such that the quotient AI5 = G jT
is an abelian scheme and the extension T "-+ G -+> A gives rise to the formal
extension f "-+ a --» A, we call T "-+ G --» A a Raynaud extension, which
is uniquely determined by f "-+ a --» A (and hence by G I 5) by the formal
existence theorem of Grothendieck ([EGA] III.5.1.4 applied to A). If a is
quasi-projective having an ample invertible sheaf .c on G, then the T-fixed part

of .c 0 [-1] *.c descends to an ample sheaf on A ([DAV] ILl Proposition 1.1);
so, A is projective. Again by the formal existence theorem of Grothendieck
([EGA] III 5.4.5), A is algebraizable. Then Ker(G -* A) has to be a torus T/ s
since its formal completion along To is a formal torus. In our application, R
is a formal monoid ring Z[[q~ll~ (i.e., the q-expansion ring) where ~ runs over
(J'v n its(X) for the dual cone (J'v of (J' E C, where X = X(G~).
Starting from a split torus T/ s and an abelian scheme A/ s , we clas-
sify all Raynaud extensions of A by T in terms of the etale sheaf X(V) =
HOmalg-gp(Tv,Gm/v) for 8-scheme V and Tv = T Xs V. By definition, X
is a constant torsion-free sheaf of Z-rank 9 ({:} dim T = g) under the etale
topology over 8, because T is split. Let A/ s be an abelian scheme of relative
dimension n - g. Since X is constant, we often regard X just as a Z-free mod-
ule of rank 9 and denote it simply by X. Let Y = Homz(X, Z) = X*. Define
a split torus T over 8 by
Then we have
HomS-alg-gp(T, G m) = HomS_alg_gp(Y 0z G m , G m)
~ (X*)* 0 HOmS-alg-gp(Gm, G m ) = X.
For any given extension over an 8-scheme 8',
0---+ T/ s' =T Xs 8' ---+ G/ s' ~ A/ s' =A Xs 8' ---+ 0,
we have the following exact sequence,
HOmS_alg_gp(T,Gm) ~ Ext1_alg_gp(A,Gm) ~ Ext1(G,G m).
Note that HomS_alg_gp(T, G m) = XIs and that the dual abelian scheme tA/ s
is equal to Ext1_alg_gp(A, G m). See [CSM] Remark 2.1.4, group extensions of
A by G m (not just scheme extensions) correspond to Gm-torsors (hence line
bundles) in the identity-connected component of PiCA/So Thus we have an
association G H ¢c E Homz(X, tA).
On the other hand, we take the universal extension
°---+ G m ---+ G ---+ A ---+ 0

over tA C PiCA/So We fix a base 6, ... , ~g of X. Then consider the morphism
T = Homz(X, G m ) ---+ Gm
given by ¢ H Li ¢(~i)' Similarly, we define Homz(X, tA) ---+ tA by using the

same base. Thus we have a commutative square
Homz(X, <G m ) = T --------t <G m
1 1
Homz(X, tA) --------t tAo
Pulling back the above exact sequence 0 ---+ <G m ---+ G ---+ A ---+ 0 over tA by
this square, we get 0 ---+ Homz(X, <G m ) = T ---+ G ---+ A ---+ 0 over Homz(X, tA)
such that for any given X E X, the morphism Homz(X, tA) ---+ tA given by
¢ H ¢(X) gives rise (by push-forward) to an extension
o ---+ <G m = T I Ker(x) ---+ GI Ker(x) ---+ A ---+ 0

which is associated with ¢(X) EtA. Thus the pair (Homz(X, tA ls ), G) repre-
sents the functor S' H [TIs' y G ---» A ls '] on the category of schemes over
S (see [CSM] 2.1~2 for more details).
For a given Raynaud extension T y G ---» A over S given by ¢c E
Hom(X, tA), we can construct its dual tc = GI K(£) (in the same manner
as the construction of tA by Mumford [ABV] Section 13; see [DAV] II.2).
Here K(£) is the stabilizer of £ under the translation action by G. The dual
tc is again a Raynaud extension T y tc ---» tA for the dual torus T =
Hom(Y,<G m ) = X®z<G m (B X(T) = Y = Homz(X,Z)). Thus we have
¢'G E Hom(Y, A) corresponding to this dual Raynaud extension.
We can generalize the above construction with respect to (R,1) to an
integral scheme S (not necessarily affine) and a sheaf of ideals J, because the
universal schemes Hom(X, tA)/o for affine open subsets 0 = Spec(R) c S
glue together by universality.
We now assume X = X(<G~); so, C(XIR ) = UaEC U. We write Xa for
XI nbEa Rad(b) for the radical Rad(b) (i.e., the null space) of b. The space
S(Xa,IR) is a quotient of S(XIR), and by duality we have S*(Xa,IR) C S*(XIR ),
where Xa,IR = Xa ®z lR. We write ~ for the collection of u giving the same
nbEaRad(b) and write Xt, = Xa for u E ~ (which is independent of u E ~).
Then Ct, = {u E Clu C S* (Xt"IR)} is a cone decomposition of Xt"IR satisfying
(Pc1~4). Let g = rankzXa . Then each boundary stratum Sa (correspond-
ing u of dimension g) is constructed from a universal extension ([DAV] IV.3)
over Za = Homz(X a , tXa): Homz(Xa , <G m ) y Ga ---» X a , where Xa = Xn~g
(g = rank Xa) is the (n-g )-dimensional principally polarized universal abelian
scheme. The scheme Sa roughly corresponds to Spec( Ra) in Section 4.1.4 in
the Hilbert modular case but is more complicated because of the contribu-
tion of the abelian variety quotient Xa. The group GL(X) acts on Ua Sa
through Xa (and permuting u E C). Roughly speaking, we can canonically
glue {Sa} aEC into a scheme S = U a Sa as in the Hilbert modular case, and
the quotient of the formal completion along the boundary of S by GL(X)
gives the formal completion of Ml along the cuspidal divisor D(X). Strictly
speaking, first we make the quotient UaEC.; Sa by GL(Xa ), and then we again
glue these schemes by the action of GL(X).
We give an indication of the construction of S~ and the semi-abelian

scheme Q~ over S~ with a symplectic level N-structure for N ?: 3 (when MN
is a scheme). Hereafter N is invertible on the base S, and suppose for the mo-
ment that we have a symplectic level N-structure a: Q[N] ~ (Z/Nz)2n. Since
Q = "G/q(Y)" , we have an exact sequence: 0 -+ G[N]-+ Q[N]-+ YN/Y -+ 0,
where Y N = kyo By moving around by an element of SP2n(Z/NZ) act-
ing on (71.,/Nz)2n, we may assume that a induces an isomorphism ay :
YN /Y -+ (Z/NZ)g and a : G[N] ~ (Z/Nz)2n- g so that the symplec-
tic pairing on the target induces the duality between T[N] ~ P}y and
YN/Y. The stabilizer of a(G[N]) gives rise to a maximal parabolic subgroup
Pn,g(Z/NZ) C SP2n(Z/NZ), and we can extend q : Y -+ tA to qN : Y N -+ tAo
Since A is principally polarized, we identify Y = X and tA = A.
Now we take S = min-g) and X = Xn- g for the unive!:sal abelian scheme
of dimension n - 9 with level N-structure. Let 0 -+ T~ -+ G~ -+ X -+ 0 be the
universal extension over Z~ = Homz(X~, tX). We write E~ = S* (X~) 0z <G m ,
which is the split torus associated with the character group S(X~). Then
S(X~) = Homz(S*(X~), 71.,) for the space S* (X~) of symmetric bilinear forms
b(x, y) = b(y, x). Let a V be the dual cone
a V = {x E S(X)ly(x) ?: 0 't:/y E a}.
Then we make the torus embedding
Let p, v E X~. Then the symmetrization [p0 v] E S(X~) of p0 v E X~ 0 X~

defines a character [p 0 v] : E~ -+ <G m . We consider the abelian scheme
X X 9R(n-g) tx = X X 9R(n-g) X on m(n-g). Here we identify tx with X by the
universal principal polarization. We have a tautological map
sending ¢ : X~ -+ tx to ¢(p). Recall the Poincare line bundle P, that is, the
universal line bundle on X x tx /tx such that for each S-valued point S -'=t tx,
a*P E PicO(X/ s ) is the line bundle represented by a. Let px = P - {O} be
the Gm-torsor associated with P.
We then consider
We have a Gm-torsor X[p0v] = c([p0v])*(p-l)X over Z~ = Homz(X~, tX).

Now we choose a base B = {[Pt 0 vt]li = 1, ... , s} of S(X~). Then 2~ =
ITt X[pi 0 Vi] is an E~-torsor, because E~ = Homz(S(X~), G m) ~ G! by
¢ M ITt ¢([Pt 0 Vi]). We symbolically write 2~ = S(X~) 0z,c* (P-l) x. The
construction of 2~ does not depend on the choice of the base, and the push-
forward of 2~ along [p 0 v] : E~ -+ G m is isomorphic to X[p 0 v]. This
follows from the symmetry of P and the fact that S(Xu) ~ Pic(Zu) given by
n f-7 c( n )*P is a group homomorphism. Since Eu is the product of the pullback
of P itself by c x c, c( n) *P has a tautological section (or trivialization) T over
Eu. This trivializes the biextension associated with P and hence gives rise to
a homomorphism i : Xu ~ Gu lifting the tautological one c = tc : Xu ~
tx = X on Zu over Eu (see [CSM] II, (2.3.3) for the details and [Mu] for
biextensions) .
We form the contraction-product Eu = Eu x Ea Eu = (Eu xZa Eu)/Eu
under the diagonal action. This product has the following universality.
1. We have the following commutative diagram
Tn
Eu xZa Eu ~ ~u
Eu
21
XZa Eu ~
I'
~u,
2. If we have the following commutative diagram
Eu XZa Eu ~ F
'I I'
Eu XZa Eu ~ ~u,
I-'
we have a unique morphism r :Eu ~ F such that r0m = Jl and r 0 I = I.

We have a polyhedral cone decomposition C(X~,lFI:) = UUEC" a. Here we
recall that C(X~,lFI:) C S*(X~,lFI:) for X~,lFI: = X~ @z lR is the cone made up of
positive semi-definite symmetric bilinear forms. Then all the data E u , E u ,
and so on, so far constructed are functorial and glue together over as in
C~ yielding E~ and so on. We write W~ = UUCCO(Xd Z(a), where Z(a) =
Spec(Z[q]qla=O,qES(X a )), The locally closed subscheme E~ XE" W~ of E~ (given
by the contraction product) should be the loci of the quasi-abelian scheme with
maximal split torus of rank;::: g. We write S~ for the formal completion of E~
along E~ xE" W~. Then we have a semi-abelian scheme vG ~ SdGL(Xd
as in [DAV] IV.3.1 with the maximal split torus of rank g on E~ XE" W~ for
g = rankX~.
We now incorporate the level N-structure. Let a : Q[N] ~ (Z/NZ)2n
be a symplectic level structure. This induces qn : Y~,N = ftY~ ~ X (Y~ =
Homz(X~, Z)), which by duality induces Cn : X~,N = ftX~ ~ tx. This gives
rise to an extension 0 ~ T~,N = Homz(X~,N' G m ) ~ G~,N ~ X ~ 0 over
Homz(X~,N' tX). By construction, qN : Y~,N ~ X lifts to qN : Y~,N ~ G~,N so
that the quotient is a principally polarized abelian scheme generically. Then
we have an exact sequence
This semi-abelian scheme Gf"N /qN(Yf"N) has a polarization
AN: Gf"N/qN(Yf"N) -t t(Gf"N/qN(Yf"N))
of degree N 2n. By definition, 9 = gN = (GN/qN(Yf"N))/Tf"N[NJ, which

acquires a principal polarization (because of degree Ng-isogeny image) and
an injection Yf"N/Yf, '--+ 97)[N] with an isotropic image. Thus any level N-
structure on X gives rise to a level N-structure of g. As shown in [DAV] IV.6.5,
this construction gives all possible level N -structures on 9 after translation
by SP2n(X/N X), and the stabilizer of this level N-structure is Pn,g(X/N X).
Thus the number of cusps is the index of Pn,g(X/NX) in SP2n(X/NX).
We write GLn(X)(N) for the kernel of the reduction map of GL(X) into
the group GL(X/NX) modulo N. Since the integral structure is given by
N· S*(X), accordingly we need to take Tf"N = Homz(Xf,..c..N , G m) instead of
Tf" because the parameterization is given by 0 -t Tf"N -t G N -t XN,n-g -t 0
for the universal abelian scheme XN,n-g with level N-structure. We have the
tautological map Cf"N : Xf"N -t tXN,n_g = XN,n-g. The multiplication by N
induces Ef, = Ef"N -t Ef" and we get the torus embedding Ef"N '--+ Ef"N.
The character group of Ef"N is given by -kS(X). The universal semi-abelian
scheme gN with level N-structure is given by
over Ind~::n) Sf"N x Isom(~,N/NYf"N' (Z/NZ)g)/GL(Yf,)' Here the formal

scheme Ind~~::n) Sf"N is the disjoint union of copies of the scheme Sf"N
indexed by the equivalence classes of cusps SP2n(X/N X)/ Pn,g(X/N X) on
which SP2n(X/N X) acts by permutation of the indices.
Since we have already studied via Tate curves the compactification of
9]tJ = 9]t~) when n = 1 (the modulus of elliptic curves), we assume here
n > 1. To make 9]tJ smooth, we assume that N 2: 3. We then define!:;!, =
f*flx/'JJtJ for the universal abelian scheme f : X N -t 9]tJ. This is a locally
free sheaf over VJ1J of rank n. We define det!:;!, = /\ n!:;!,. The vector bundle
!:;!, extends to the toroidal compactification MN by defining!:;!, = f*fl gN / MN
for the semi-abelian scheme 9 extending X to M. Then we define a graded
algebra
EB EB H
00 00
Qj = QjN = HO (MN,det(!:;!,)®m) (;d O (VJ1;("det(!:;!,)®m).

m=O m=O
The last identity (*) follows from Koecher's principle (Theorem 6.24 (4)) if
n > 1. Choosing the toroidal compactification data well, MN is projective
([DAV] V.2.5) so that this graded algebra Qj is finitely generated over Z[-kJ,
and by the first equality, the graded algebra is normal. Thus we may define Qj N
to be the normalization of Q)l in the algebra Rd,N defined below. We define

the minimal compactification by MN = Proj (Q) N ). It is called "minimal"
because any smooth toroidal compactification MN covers M N.
We can define a sheaf of graded algebras over MN by
00
m=-oo
Then MN = SpecM(R) represents the following functor
Then we have MN = G m \MN as a geometric quotient, and MN is a G m -

torsor if N ::::: 3. Here a E G m acts on the functor PN by (A,<PN,>',W) r--t
(A, <PN, >., aw). The relation between the modulus scheme classifying abelian
schemes with level structure and the one classifying with extra information of
nowhere-vanishing n-differentials is exactly the same as in the elliptic modular
case, and it is also interesting that this is proven only after the hard work of
smoothly compactifying the open modulus SJJC/v.
Remark 6.25 Let the notation and the assumption be as in Remark 6.21.
Carrying out the toroidal compactification process for SJJCfN to obtain a
smooth compactification Mi,'rv
over Z[(N, d~], we apply C~rollary 6.11 to
the semi-abelian scheme Q/MJI . Then, automatically, we get a toroidal com-
d,N
pactification of the moduli SJJC( c, r(N)) with respect to the data induced from
the toroidal compactification data of S*(O) (as long as they are compatible).
Here i)-l is embedded into S* (0) regarding ~ E i)-l as a symmetric bilinear
form on 0 x 0 by (x, y) r--t TrF/IQ(~XY). An important compatibility of the
toroidal compactification data is that the polyhedral decomposition of positive
semi-definite C(OIR) C S*(OIR) has to induce that on F::'+ C i)-l Q9z lR = Foo
by taking intersection with the image of i)-l.
6.4 Siegel Modular Variety
In this section, we study the Shimura variety associated with G = GSp(2n) /IQ'
where, for a ring R,
-ci
for I n = (l~ n ) with 0 < n E Z. Consider a homomorphism ho : S/IR :=
ResCjIRG m --t GSp(2n)/IR of algebraic groups given by S(lR) = ex 3 a + bi r--t
(~i/ln ~:i~) E GSP2n(lR). Let X be the collection of all conjugates of ho under
GSP2n (lR), and write X+ for the connected component of X containing h o .
Then we have X ~ SJn U SSn by sending ghOg-l to g( R . In) E SJn, where
6.4 Siegel Modular Variety 291
.YJn= {Z E Mn(CWz = z, J=T(z - z) > O} (the Siegel upper half-space)

and S)n (the Siegel lower half-space) is the complex conjugate of .YJn. The
action of 9 = (~~) E GSP2n(lR) (with n x n matrix a) on .YJ n is given by
z H (az + b) (cz + d) -1 .
For a scheme SIR defined over a finite B-algebra R, when we regard it
as a B-scheme, we write it as SIB' The scheme SIB is not equal to the
Weil restriction of scalars ResRIBSI R discussed in Chapter 4. Then we have
wt~/Z[*J(q ~ G(Q)\(X x G(A(oo)))/F(N), where we put
T(N) = {a E GSP2nCi)la - 1 EN· M2nCZ)}.
Therefore the construction of the Shimura variety for (G, X) follows from the
existence theorem (Theorem 6.20) of the moduli variety over Q of polarized
abelian varieties (proven by Mumford [GIT] and Shimura [Sh3]).
In this section, we give in this section an exposition on its global reciprocity
law and the irreducibility of the Igusa tower over Sh(p)(G, X).
6.4.1 Moduli Functors
We consider the Q-vector space V = Q2n with alternating form (x, y) = txJny.
Thus G(R) = GSP2n(R) acts on VR = V ®Q R by matrix multiplica-
tion preserving the alternating form up to scalar multiplication. For each
(A,A)ls E A 1,1(S), choosing a geometric point s = Spec(k(s)) of each con-
nected component of S, we consider its physical Tate module
Ts(A) = ~NA[N](k(s)),
which is naturally a module over Z. If S is a Q-scheme, Ts(A) is a free Z-

module of rank 2n. If S is a Z(p)-scheme, its prime-to-p part Ts(p) (A) =
Ts(A) ®z Z(p) is a free Z(p)-module of rank 2n. We then define
We consider level structures TJ : VA(oo) ~ Vs(A) and TJ(p) : VA(poc) ~ V}p) (A).
An element 9 E G(A) acts on TJ (and on TJ(p)) by TJ H TJ 0 g. For a closed
subgroup K c G(A (00)), the level K-structure fj = TJ 0 K is defined over S,
if 0' 0 TJ = TJ 0 ks(u) with ks(O') E K for all 0' E 1f1(S,S) for each chosen
s. Changing s to another point s' in the same connected component gives
isomorphisms (called paths) Ls,s': 1f1(S,S) ~ 1f1(S,S') and LL,: Ts(A) ~
Ts, (A) which satisfy the compatibility condition (see Section 4.4 in the text
and [EeH] Section 1.5) LL,(u(x)) = Ls ,8'(U)(LL,(x)); so, if fj is defined over

S, we have
Thus giving fj defined over S at s is equivalent to giving LL, 0 fj again defined

over S at s'; so, we may forget about the choice s.
Since a positive multiple of the polarization is induced by an ample line
bundle globally over S, the Galois action of 11'1 (S, s) preserves the alternating
form (-, ')>.,8 on Ts(A) induced by A and the eN-pairing on A[N] x tA[N]. The
pairing (-, ')>',s is sent to (-, '».,8'
by LL,; so, the following condition
(TJ(x),TJ(Y))>.,s = (x,y) up to the scalar multiple in (A(oo))X (6.11)
is independent of the choice of s. Thus hereafter we drop the subscript s since

it is irrelevant to our purpose. We always impose this condition on TJ in this
section.
We consider the following functor cYP :
Q-SCH -t SETS for each closed
subgroup K of G(A(oo)),
c!J:)(S) = [(A,X,fj = TJOK)/si(A,A) E A 1,1(S), TJ: VA(oo) ~ V(A)] , (6.12)
where X is a polarization class of polarizations proportional to a principal po-

larization on A up to multiplication by Q~, and the alternating form induced
by A on VA(oo) is proportional to (-, .) on VA(oo). Here the isomorphism classes
are taken in the category of abelian schemes up to isogenies; so, we have an
isomorphism (A, X, fj) ...:::'.t (A', X', fj') if we have an isogeny cjJ : A -t A' with
cjJ*X' = X and fj' = cjJ 0 fj up to a nonzero scalar multiple. When K = F(N),
we simply write c!.J) for c!J:). In the definition of the functor c!J:), we do not
need the condition similar to (pol) in Section 4.2.1 because we can choose
an isomorphism V ~ HI (A, Q) for each test object (A, X, fj) compatible with
TJ : VA(oo) ~ V(A) = HI (A, A (00)) (because alternating forms isomorphic over
A(oo) are isomorphic over Q; see Section 7.1.5).
Let L = z2n C V be the self-dual lattice under (-, .). We write L for L0zZ.
Then in the same manner as in Section 4.2.1, we can show an isomorphism of
functors c!J:) ~ CK (for K c G(Z)), where
Here we have an isomorphism (A,A,fj) ~ (A',A',fj') if cjJ : A ...:::'.t A' is an

isomorphism (not an isogeny) of abelian schemes with cjJ* A' = A and fj' = cjJofj.
When K = F(N), we again write CN for CK' Then CN is represented by 9Jt;\r/IQ'
because having TJ mod F(N) with TJ(Z2) = T(A) (so that the Weil pairing
inducing the polarization form (', -)) is equivalent to having cjJN : (Z/Nz)2n =
L/NL ~ A[N] (compatible with the polarization form (-, .)). Thus £~0) is
represented by Sh(G,X)/Q = ~NOO1~/Q'
In order to construct a p-integral model of the Shimura variety, we impose
(6.11) by substituting ",(p) for", and require that X be a polarization class of a
principal polarization up to a prime-to-p scalar multiple. Then we can consider
the following smooth model Sh(p)(G,X)/zCP) = ~pfNOO1~/zCP)' whose quo-
tient Sh~) = Sh(p)(G, X)/ K by a closed subgroup K with K = G(Zp) x K(p)
(K(p) c G(A(p))) represents the following functor £jI) : Z(p)-SCH -+ SETS,
(6.13)
where fj indicates the K(p) orbit ",(p) 0 K(p). Here the isomorphism classes are
taken in the category of abelian schemes up to prime-to-p isogeny; so, we have
an isomorphism (A, X, fj(p)) ~ (A', X' , fj' (p)) if we have an isogeny cp : A -+ A'
with p f I Ker(cp)I = deg(cp), cp*X' = X, and fj'(p) = cp 0 fj(p) up to a prime-to-p
scalar multiple.
6.4.2 Siegel Modular Reciprocity Law
We can let g E G(A(oo)) act on the functor £~0) by (A,X,,,,)/s H (A,X,,,, 0

g)/s, and hence we get a schematic action of G(A(oo)) on Sh(G, X). Since
the isogeny ~ E EndQ(A/s) for ~ E Z(Q) induces an isomorphism (A, X, "') ~
(A, X, '" 0 ~) in the category of abelian schemes up to isogeny, Z(Q) acts
trivially on Sh(G, X), and the action factors through Gr:c~)). Let E be a
-(E) G(ACEoo))
proper subset of rational primes. We define £ (G, X) Z(ZCE)) • When
E = 0, we have
£,(0)(G X) = G(A(oo)) = G(A(oo)) * Gad(Q)

, Z(Q) Z(Q) G(Q)+
in terms of Deligne's formulation described in Section 4.2.2. Note that Z(Q) =

QX is closed in G(A(oo)) (a property special to Q); so, Z(Q) = Z(Q). We put
OO1j~lyc = ~NOO1~/Q[ILNl for N running integers prime to E as a subvariety
of Sh(17) = Sh(G, X)/G(Z17). Recall that Q(17) /Q is the maximal extension
inside QCYc unramified at every pEE. Here is the global reciprocity law.
Theorem 6.26 (Shimura) Let G = GSp(2n)/Q' Then the variety 001(17)
over Q(17) gives a geometrically irreducible component of Sh(17)(G, X)/Q con-
taining the image of 5)n x 1. Writing j{(17) for the function field Q(17)(OO1(17)),
we have the identity Aut(j{(17)) = Aut(001j~)), and we have a canonical iso-
morphism T : £,(17) (G, X) ~ Aut(j{(17)) satisfying the following properties.
(1) On Q(17), g E G(A(17oo)) acts by the Artin reciprocity map in Theorem 2.12
composed with the similitude norm v: G(A(17oo)) ---+ AX;
(2) For f E jt(17) regarded as a function on X+, we have T(a)(f) = f 0 a- 1
for a E G(Z(17))+;
(3) The map T is an isomorphism of topological groups with respect to the
adele topology on l(17)(G,X) and the Krull topology on Aut(jt(17)).
The action of T(a) looks different from the one in Theorem 2.43 and actually,
our T(X) here is the inverse of T(X) in Theorem 2.43 in order to make the
construction of T compatible with our definition of the action on the level
structure TJ H TJ 0 g (see (6.14) below).
Recall the valuation ring Zi~) (with residue field IF) in Q(17) of the p-adic
place ip : Q(17) y Qp. If PEE, we have a modulo p version of this theorem
determining Aut(e~17)) for the function field e~17) of 9J1j;) = 9J1(17) Q9Z(E) IF.
(p)
Since it is similar to Theorem 4.17, we leave the formulation of the result to
the reader (see [H03b]).
We repeat here the argument given for the Hilbert modular version (The-
orem 4.14) based on an argument in [AAF] Section 8 and the theorem of
Faltings and Zarhin (Theorem 4.13). For the original proof of Shimura (and
Miyake), we refer the reader to [Sh5] II 6.5 and [Mt].
Proof. Since the proof is the same for any choice of E, we assume that E = (/)
for simplicity, and we throw all superscripts and subscripts E away from our
notation. By our convention, Z(0) = Q. We first check that T(a- 1 ) (f) = f 0 a
by studying the complex points 9J1j{,(cC) of 9J1j{,.
For a given (A, A) E £1(cC), we choose a base {Wl, ... ,wn } of HO(A, DA/c) ,
and consider the lattice .c A = {U-y Wj) J 11' E Ho (A, Z)} in en. By the Poincare
duality and the principal polarization A, we may assume that
.cA = .c z = t L . In ( 1: )
for Z E -Dn (see [AAF] 4.4), regarding L to be a column vector space. Then
A(cC) ~ en j.c z as complex manifolds, and A is induced by the alternating form
1: )/\
Az : .c z /\ .c z ---+ Z sending tx . I n ( ty . I n ( {n ) to x . I n ty E Z. We write
Az for the abelian variety with Az(cC) ~ enj.c z . Define the main involution ~
of G by a· = J;;l . hJn for a E G(R). Note that (ax, y) = (x, a·y) for the
alternating form (x, y) = tx . Jny on L.
If we have an isogeny ¢ : A(cC) ~ en j.c z ---+ en j.c w ~ Y(cC) preserving
the polarization class, and identifying en with the Lie algebra of the abelian
varieties, we have a e-linear map if> : en ---+ en lifting ¢ with if>(Q.c z ) = Q.c w
preserving the polarization form. In other words, we find a E GL 2n (Q) such
that
if> (e . I n (1: ))
= e· a . I n U':' ) j (a·, z)
for all eEL, where a· = J;;l . hJn = v(a)a- 1 and j(g, z) = cz + dE GLn(cC)
for g = (~ ~). Since a preserves polarization up to positive rational numbers,
v(a) > 0, and hence a E G(Q)+. This shows that w = a(z) for a E G(Q)+.
We define ¢N,z : LIN L ~ Az[N] by ¢N,Ax) = if

I n · ({n) mod .c z' Then
(Az, A, ¢N,z) 2:! (Aw, A, ¢N,w) «=? w = ')'(z) for some,), E r(N) = r(N) n
G(Ql)+. For any given level structure ¢ : LINL ~ Az[N] compatible with
the polarization A, we find')' E SP2n('l!.,) such that ¢ = ¢N,z 0 ')'. Thus we
find 9J1)(,(C) ~ r(N)\SJn, in particular, 9J1)(, over Z[(N, *'],
is geometrically
irreducible and smooth if N ::::: 3. This follows from the same argument in the
Hilbert modular case (given just above Theorem 4.17), using the existence of
the smooth projective toroidal compactification of Faltings-Chai described in
Section 6.3.6, by Zariski's connectedness theorem ([ALG] III.l1.3). This shows
that 9J1 /lQIc y c gives the geometrically irreducible component of Sh( G, X) /lQIc y c
containing X+ x 1 = nn x 1.
By taking the limit ¢z = ~ N¢N,z : i ~ T(A z ) and tensoring A. (00),
we get rJz : VA(oo) ~ V(Az). We let a E G(Ql)+ act on column vectors in
V = L 0z Ql by (left) multiplication. Since the linear map j (a L , z) : <en '3 U H
j(a" z)u E <en for a E G(Ql)+ induces an isogeny
(An-,(Z)' An-'(Z)' ¢N,n-l(z») ----t

(<en It LtaJn ( I':, ) ,j (a Z )*A z , j (a" z )r]n-l (z») ~ (Az, Az , rJz 0 a),
L
, (6.14)
we have T(a(oo»)([z, 1]) = [z, a(oo)] = [a-I(z), 1], writing [z, g] for the complex
point of Sh(G, X)(C) represented by (z x g) E nn x G(A.(oo») as in (4.21).
This shows that a E G(Ql)+ c G(A.(oo») acts on f E ~ by f H f 0 a-I.
By the above expression of the complex point of 9J1, we find that T
brings G I (A.(oo»)/{±I} injectively into Aut(9J1/IQI)' We consider the eN,z-
pairing eN,z : Az[N] x Az[N] ----t /LN· Since eN,z(¢N,z(X), ¢N,Ay)) = (]:,y)
for x, y E LIN L, we find that eN,z(ax, ay) = (~(n)(x,y) for a E G(Z). Then in
the same manner as in the proof of Theorem 4.14 Step 2, we create a section
SA : ZX '3 C H (C~n IOn) E G(A.(oo») of V so that the action of T(SA(C)) on the
q-expansion at 00 factors through the Galois action of the Artin symbol [c, Ql]
on the coefficients of the q-expansion. Thus we get the following commutative
diagram with exact rows (and the section SA of v),
where h is the Artin reciprocity map in Theorem 2.12.

After verifying this, we proceed in exactly the same manner as in Step 3 of
the proof of Theorem 4.14, and we get surjectivity of T out of Theorem 4.13
combined with the results shown in Section 4.4.3. 0
The local reciprocity law at CM points is similar to Corollary 4.20 (see
(rc1-2) in Section 7.2.2 and Theorem 7.7).
6.4.3 Siegel Modular Igusa Tower
We can make the toroidal compactification MN of rrniv (for N prime to p)

choosing the toroidal compactification data carefully so that the group G(Z(p))
acts on the limit M;~;p) = ~pfNMN/Z(p). This is done as follows. Choose
small No ;::: 3 and a sequence N oIN1 IN21·· ·INJINJ+lI··· of positive integers
so that Sh(p)(G,X) = ~jrrnNJ. Fix a simplicial decomposition of C for No
at the cusp 00. We take C sufficiently fine so that MNo is smooth projective
over Z[(No ' JJ Then we take the isomorphic image of the decomposition at
every other cusp. Write this decomposition as C s for each cusp s E MNo.
Using the same decomposition C s for all cusps of MNJ over s, we still get a
smooth projective compactification MNJ which is a finite Galois covering with
Galois group F(No)/f(NJ ) as described in Theorem 6.24. Since the action of
SP2n(ZjNJZ) over rrniv permutes cusps, by our choice, the geometric action
J
of f(No)jf(NJ ) on MNJ extends to the action of SP2n(ZjNj Z) on M NJ .
through coefficients Z[(N, *],

The arithmetic action of GSP2n(ZjNj Z)jSP2n(ZjNJZ) ~ Gal(Q(N,)jQ) is
which extends to MNJ/Z(p) as described in the
proof of the global reciprocity law (Theorem 6.26). Thus GSP2n(ZjNJZ) acts
on MNJ/Z(p). The action of GSP2n(ZjNJ+ 1 Z) is compatible with the projection
map M NJ + 1 -» M NJ , because the polyhedral decomposition for the level Nj+l
is the same as that for the level N J • Taking the limit, we get a good action of
G(Z(p)) on M(P) .
/z(p)
The limit M(p) = ~JMNJ/Z(p) gives a smooth compactification over Z(p)

of Sh(p) keeping the action of G(Z(p)). We then define Mt/)z
(p)
= M(p) j K for
a closed subgroup K(p) c G(Z(p)) with K = G(Zp) x K(p) (i.e., a closed sub-
group maximal at p). In particular, M(p) j F(Nj ) = MNJ/Z(p) for all j. By mini-
mality, on the limit M(p) * = lim pfN M N/Z of the minimal compactifications,
F (p)
the adele group G(A(poo)) acts naturally. In particular, M(p)~ = M(p)* j K is

the minimal compactification of Sh~).
We go through the same process for Sh(G,X)/Q = ~Nrrn:Zr/Q and get
a toroidal compactification M/Q = ~ N M N /Q which has the natural action
of G(Z) extending that on Sh(G, X). We also define MiQ = ~NM;:"/Q on
which G(A (00)) acts. In this case, we have a smooth toroidal compactification
MK/Q = MjK and the minimal compactification MK/Q = M*jK for all
closed subgroups K of G(A(oo)).
We have defined in Section 4.1.6 the Hasse invariant H for any semi-abelian
scheme f : Q -t S over a characteristic p-scheme with a given base of invariant
differentials and verified that H E det~®p-l where det~ = f*(/\n [29/S). In
particular, H vanishes at any point of S where Q has a nonordinary semi-
abelian scheme as its fiber. Here a semi-abelian scheme is p-ordinary if the
connected component Q[PjO of the kernel Q[Pj of multiplication by P is iso-
morphic to fJ~ etale locally. If 9 is an abelian scheme, by the duality, this

is equivalent to having the etale quotient 9[pjet isomorphic to (Zjpz)n etale
locally. Thus we have a well-defined Hasse invariant H E det(~)~;~pl) because
M(p) carries a semi-abelian scheme (depending on the toroidal compactifica-
tion data).
We pick a lift E E HO(M;~;p)' det(~)a(p-I)) (for a suitable 0 < a E Z)
of the Hasse invariant H of the (universal) semi-abelian scheme 9 over the
toroidal compactification M;~:. Then we put Sm = (M(p) ®z(p) Zjpmz) [~],
S:n = (M*(p) ®z(p) Zjpmz)[~], SK,I = SdK, and SK,I = S1/ K all defined
over lFp. Since MK ® IF = Proj(<BK) for <BK = EBo-:;J HO(Sh~jYlp' det~~i),
we have SK,I = Proj(<BK[iTD = Spec(<BK[iTjO) (cf. [GMEj l.3.2), which is
affine. Although MK/Yl p is projective, det~ is not ample on M K ; so, SK,I is
not necessarily affine.
n
We define the self-product 9'K = QK XM K 9K X ... XM K 9K'. Here 9K

is the semi-abelian scheme over M K, which is again given by 9 j K for the
limit 9/ M (p) of semi-abelian scheme 9NJ over MNJ • Then the Cartier dual
Am of the connected component 9'K[pmjo of 9'K[pmj over SK,I is an ad-
ditive Mn(ZjpmZ)-torsor, because we have 9'K[pmjo ~ fJ~: (etale locally)
whose Cartier dual is Mn(Zjpmz). We take its multiplicative GLn(Zjpmz)-
subtorsor TK,m = {(gl,"" gn) E A m igl/\g2/\" ·/\gn -=I- a}. As we show later,
this TK,m is the m-th layer of the Igusa tower over SK,l/Yl p' Thus TK,mjSK,l
is an etale covering on which GLn(Zjpmz) naturally acts. We prove the ge-
ometric irreducibility of TK,m over each connected component of S K,l/IF p for
all open compact subgroups K maximal at p.
Let a be a ring. For two given group schemes (on which a acts linearly)
f : P -+ Sand l' : P' -+ S, we define the following functors Isoms(P, P') :
SCH/ s -+ SETS and Homs(P, P') : SCH/ s -+ SETS,
Homs(P,P')(S') = {¢ E HomS'_SCH(PS"p~,)I¢ is a-linear}
(6.15)
Isoms(P, P')(S') = {¢ E Homs(P, P')(S') I ¢ is an isomorphism} ,
where PS' = P xsS'. Let Pm be the Cartier dual of 9K[pmjo (see Section 3.l.6
for Cartier dual). Having a homomorphism ¢ : (ZjpmZ)/s' -+ Pm/ S' is just
equivalent to knowing the destination of the standard base {e 1, ... , en} in Pm.
By assigning (¢(ed, ... , ¢(e n )) E Am(S') to ¢ E Homs((ZjpmZ)n, Pm)(S') ,
the Mn(ZjpmZ)-torsor Am represents the functor Homs((ZjpmZ)n, Pm). In
the same way, TK,m represents Isoms((ZjpmZ)n,Pm ). By taking the Cartier
dual and taking the inverse of elements in Isoms(Q[pmjO,fJ~=), TK,m also
represents the functor Isoms(fJ~=, 9[Pmjo). Each element g E GLn(Zjpmz)
acts on ¢ E Isoms((Zjpmz)n, Pm)(S') by ¢ M ¢ 0 g.
Let W be a complete discrete valuation ring of the p-adic completion of
the maximal unramified extension of Qp in Qp. Put W = i;l(W) C Q. Thus
the residue field IF of Wand W is an algebraic closure of lFp .
Theorem 6.27 (Chai-Faltings) Let K be an open compact subgroup of

G(AJoo) maximal at p. Then the projection 7fo(TK,d -t 7fo(SK,d is bijective;
so, the pullback in TK,l of an irreducible component of SK,l is irreducible. If
v(K) = ZX, then SK,l/F p and TK,m/F p are geometrically irreducible.
This is, of course, a generalization of a result of Igusa (Theorem 3.3). This
has been proven in [DAV] V.7 showing that the p-adic monodromy group at
the cusp 00 is large. We give a proof via the global reciprocity law. Our proof
is based on the existence of the smooth toroidal compactification of the Siegel
modular variety given by [DAV], but our viewpoint is slightly different from
the treatment in [DAV] V.7, looking into the monodromy of the generic point.
This viewpoint can be generalized to prove irreducibility of the (generalized)
Igusa tower for general unitary and symplectic groups over totally real fields
bigger than Q (see Section 8.4 and [H03a] Lecture 10), which cannot be at-
tained by the computation in [DAV] of cuspidal monodromy (see the remark
after Theorem 4.21).
Proof. We need to prove that IK,m = TK,m XS K ,l SK is geometrically irre-
ducible for each geometrically irreducible component SK C SK,l.
The valuation ring W is the strict Henselization of Z(p) in Q associated
with ip : Q Y Qp. Since all the assertions are geometric, we may extend
the scalar of all the schemes involved from Z(p) to W whose residue field is
iFp, which we have written as F. Since K is maximal at p, it contains F(N)
for some N ~ 3 prime to p, and we may assume that K = F(N). Then
MK is smooth projective over W, and hence by Zariski's connectedness theo-
rem ([ALG] III.11.4), geometrically irreducible components of a special fiber
are in bijection with those of the generic fiber (see the argument just above
Theorem 4.17 applied to the Hilbert modular varieties). Thus a geometri-
cally irreducible component of 9J1 N,l = 9J1N ®w F is given by 9J1'/v (p f N).
Since G(Z/ NZ) acts transitively on all other irreducible components, we may
assume SN := Sf'(N) = Mk,l[~]' Write S/F = ~pfNSN which remains geo-
metrically irreducible (because W contains all Z[/LN] for N prime to p). Let
So C S/F be the subscheme obtained from S removing the cuspidal divisors
and SjF = M*[~] for the minimal compactification of MiF of 9J11F'
We write IK,a for the scheme IK,a xS K S'k. Then write IK,a for the
normalization of S'K/F in IK,a/F' Since S'K is normal at cusps, IK,a contains
a GLn(Zp)-orbit of infinity cusps. We choose a coherent system of the infinity
cusps of IK a (indexed by a) and write it as 00. We have a tower of irreducible
subvarietie~ {Ca C Ta}a such that the image of C a in IK a contains the
infinity cusp. We may identify Gal(Coo/Co) with a closed ~ubgroup H C
GLn(Zp). Thus we need to show that H = GLn(Zp).
We write the ring B[[S2(X)]] for X = zn studied in Section 6.3.6 as a
q-expansion ring B[[q~lL~ for symmetric n x n matrices ~. Here B is a base
ring. Since we have a (rational) toroidal compactification of 9J11iQ[(NJ' for all
f E HO(9J1fv/l3' det~®k) (with a Q[(N]-algebra B), we have its q-expansion
f(q) = 2:~2':O a(~, f)q~ with a(~, f) E 8, where ~ runs over positive semi-
definite n x n matrices with coefficients in Q. The minimal compactification
of >JJi.!v/Q is given by Proj(r8f(N)). Thus any element f in the function field
jt = QCYC(>JJi.) for >JJi. = ~ N>JJi.!v Q[J.!N 1 can be written as J with n, d E
r8k for a suitable integer k > 0, w~ere r8k is the kth graded component of
r8. The minimal compactification is projective and can be embedded into a
projective space using theta series, which has integral q-expansion coefficients
(cf. [CSM] Appendix II and [AAF] Section 6). Thus for a given f E jt or
more generally f in the field of fractions Frac( r8) of r8, we can take nand
d with q-expansion coefficients bounded under the p-adic valuation ordp of
W. Thus, the valuation v(n) = Inf~ ordp(a(~, n)) is a well-defined element in
Q U {oo}. Similarly v(d) E Q is also well-defined. Then we define the valuation
v: Frac(r8) -+ Q U {oo} by v(f) = v(n) - v(d).
We have an isomorphism r : £(G, X) ~ Aut(jt) = Aut (>JJi./Q) of
topological groups (Theorem 6.26). Since the restriction of r(x) to QCYc is
given by the action of the similitude norm v(x) E AX /QxlR.~ composed
with the Artin reciprocity (Theorem 2.12), r induces another isomorphism
G 1 (A(00))/{±1} ~ Aut(>JJi./QCYc) = Aut(jt/QCYC).
We write Vp for the decomposition group in Aut(>JJi./ Q) for this valuation
v. As seen in the proof of Theorem 3.3, Vp is a closed subgroup of r(£(G, X)),
and we prove that Vp/Ip contains GLn(7l,p) for the inertia subgroup Ip.
The scheme S* is geometrically irreducible, that is, S*0wQ and S*0wiFp
are both irreducible. We then construct T/:. = ~ KTJ(,a and T; = ~ KTj(,a'
where K c G(Jr,(oo)) runs over all open compact subgroups maximal at p.
Then T~/ S is an etale covering, and we take the geometrically irreducible
component C':x, of T~ containing the infinity cusp, and put C~ = C':x, n T~.
Since the action of GLn(7l,p) for the diagonal subgroup GL(n) of G 1 commutes
with the action of G 1 (A(poo)) and coincides with the action of GLn(7l,p) on
T':x,/ So, we have a commutative diagram with exact rows,
(CO SO)
A u too; ~
-----, Aut(SO)
tu tu
'-+ (G 1 (A (poo)))/ {±1} x H -# G 1 (A (poo))/ {±1},
where Aut(C~; SO) is the group of automorphisms of C~ that induce auto-

morphisms of So. This shows that the irreducible component C~ is stable
under the action of g E G 1 (A(poo)). In particular, the infinity cusp of C':x, is
sent by g to another cusp of the same irreducible component C':x,. This shows
that r(g) for g E G 1 (A(poo))/{±1} c £(G, X) preserves v.
There is another way to show that G 1 (A(poo))/{±1} preserves the val-
uation v. Let P(Q) = {( g~) E SP2n(Q)la E GLn(Q)}. We see easily that
r (g ~) -1 for (g ~) E P(Q)+ := P(Q) n G(Q)+ sends q-expansion 2:~ a~q~
of a modular form to 2:~a~exp(27l"iTr(~bd-1))qd-l~a. Thus the image of
P(Q)+ in £(G, X) preserves v. For the unipotent radical U of P, the ad-
ditive group U(Q) is dense in U(A.(oo)). Thus U(A.(oo)) C Dp, since Dp is a

closed subgroup. Since sip) is irreducible (if p f N), the action of G1(A.(poo))
on Q(p)(S(p)) = j{G(Zp) preserves v restricted to j{G(Zp). Indeed, the action
of G 1 (A. (poo)) on HO(S)~, det(~)®k) preserves v by the irreducibility of si/~.
By the density theorem Theorem 8.3,
D' = {f E EB HO(S}~, det(~)®k) 0z Qlv(f) 2: O}

k~O
is dense in the space of p-adic module forms of level poo. From this, we con-
clude that v restricted to j{U(Zp) is preserved by G1 (A. (poo))U(Zp)/ {±1 }U(Zp),
which is enough to show the stability of von j{ under G 1(A.(poo))/{±1} (be-
cause T(U(Zp)) C T(U(A.(oo)) C Dp).
The closure of the image of G1(A.(poo))P(Q) in £(G, X) is in Dp. Note
that the projection of G1(A.(poo))P(Q) into G1(Qp) contains P(Q) C G1(Qp),
which is dense in P(Qp). Thus Dp contains the image under T of
and U(Zp) for the unipotent radical U of P.

The action of a E GLn(Zp) is given by (A,.x, 7]) r-+ (A,.x, 7] 0 a), which in-
duces the action ip r-+ ip 0 a for a = (t ao-' ~). This shows that Gal (C~ / C8) =
GLn(Zp), and hence TK,O'. = CO'. is irreducible over ]Fp. 0
Here is the determination of the p-decomposition group Dp in G(~~~;)+

Theorem 6.28 Let Dp C Aut(j{) be the decomposition group of the valuation
v defined in the above proof. The morphism T in Theorem 6.27 induces an
isomorphism {g E G(A.(oo))lgp E P(Qp), v(g) E Q;Q~} /Z(Q) 2:! Dp.
Proof. We have already shown that T(P(Q)) C Dp and T(G1(A.(poo))) C Dp.

6n
Since T (t' l~) with tEA. x acts on the q-expansion coefficients at the infin-
6n
ity cusp through the Artin reciprocity law, we find that T (t' l~) with t E Q;
is in Dp. These elements topologically generate the subgroup of £( G, X) in
the corollary. Conversely, if T(g) E Dp for 9 E G(A.(oo)), since T(g) acts by
v(g) on QCYc, we find that v(g) E Q;Q~. Thus modifying 9 by an element
in P(Q), we may assume that gp E G1(Zp) for G 1 = Sp(2n). Thus we may
assume that F = ]Fp.
Let D = {gp I T(g) E Dp and gp E G1(Zp)}. By the above argument, we
need to show that D = p(1)(Zp) = P(Zp) n GI(Zp). For each a E G(Qp), we
can think of KO'. = aG(Zp)a- 1 n G(Zp). Then Sh K " is a covering of Sh;~ =
ShG(zp) in two ways. One is coming from i : ShK", :3 [x, g] r-+ [x, g] E Sh(p)
and the other coming from a : Sh K", :3 [x, g] r-+ [x, gal E Sh(p). In other
words, 9JtK", can be considered as a correspondence [G I (Zp)apG 1(Zp)] in-
side 9Jt(p) x 9Jt>~ given by the image of i x a. For a p-isogeny (A, X, 7](p)) ~
(A~-, X, r/ p )) with ordinary A over a base S on which p is nilpotent, choos-

ing ordinary level structure (Qp/Zp)n ~ A[poo]et ~ A~ [poo]et and its dual
inverse Jl;oo ~ A[POO]O ~ A~[POO]O under the polarization pairing, we have
two identifications Homs(A~[pOO]et,Qp/Zp) ~ Homs(A~[pOO]et,Qp/Zp) ~ Z;
and Homs(A[pOO]O, Jlpoo) ~ Homs(A~[pOO]et,Jlpoo) ~ Z;, and ~ induces two
n x n matrices a, d E Mn(Zp) (where a sends Homs(A~[pOO]et, Qp/Zp) into
Homs(A~[pOO]et,Qp/Zp)). The elementary divisor of (o~) is independent of
the choice of the ordinary level structures. Then ShK,/s[kl/Sh~)s[k] for
the Hasse invariant H classifies all isogenies (A, X, 1](p)) ~ (A~, X, 1](p)) (for
ordinary abelian schemes A) whose elementary divisor is the same as the ele-
mentary divisor of a. We may restrict ourselves to ordinary locus S h K", liB) k]
to study irreducible components of ShK",/IFp because ShK",/IFp[k] is Zariski
dense in ShK",/IFp (e.g., [DAV] V.4.3). We call the pair of elementary divisors
of a and d the p-isogeny type of ~, which gives a finer invariant of p-isogenies
than the elementary divisor of a.
This interpretation of Sh K ", (as the moduli of isogenies of elementary di-
visor type a) also works w~l over any W-algebra as long as we interpret
it as a formal subscheme Sh K ", of the self-product of the formal comple-
tion Sh~) of Sh~\~] along its modulo p fiber (for a lift E of the Hasse
invariant). The scheme ShK",/w is the algebraization of ShK",/w realized
as a closed subscheme of Sh(p) xw Sh;% (see [DAV] VIl.4) and is finite
over Sh;%. We count the number of geometrically irreducible components of
ShK",/F = ShK",/w ®w IF over fJTCW. We write fJTC a / K (K = Frac(W)) for
the unique geometrically irreducible component of ShK",/'Q over fJTC;~. Since
Gal(fJTC/ij/fJTC)~) = G 1 (Zp), by Hilbert's theory of decomposition of primes
([BCM] V.5.2-3), the irreducible components modulo mw are in bijection
with double cosets K}}\G 1 ('Zp)/ D for K}}) = Ka n G 1 (Zp), since irreducible
components correspond to minimal primes of the integral closure in the func-
tion field of fJTCK",/K of the valuation ring of v in the function field of fJTC;~.
Let am = (10 p.21n ). Thus if we can prove
IK~\Gl(ZP)/DI ~ IK~)\Gl(Zp)/K~)1
for all m, we conclude D = nm

K~) = p(1)(Zp), since p(1)(Zp) cD. We let
G(Zp) act on Lm = (Z/pmz)2n by matrix multiplication. We consider the set
.em of maximally isotropic free Z/pmZ-submodules of Lm; so,
.em = {V c Lm I V ~ (Z/pmzt and tv· Jnv' = 0 for all v, v' E V} .
Then G 1 (Zp) acts on.e m transitively. We write Vm for the element of .em gen-
erated by the first n standard basis of (Z/pmz)2n. Then K~) C K(l) is the
stabilizer of V'rn in .em. Identifying .em with the subgroups of A[P1nj of a generic
p-ordinary member A/ w of the universal abelian scheme A/ 9J1 (p) , 9J1 m = 9J1K=
classifies isogenies cp : A --t A/V whose kernel is free of rank n over 7l/pm71
generically. On the special fiber of characteristic p of A, this kernel V = Ker( cp)
degenerates into a 7l/pm 71-submodule of A[pm](lF) ~ (71/p m71)n. This image is
determined up to isomorphisms by the intersection VnA[pmjo (Qp) for the con-
nected component A[pmjo over W, which also determines the p-isogeny type of
cp. In [Shlj Theorem 5, it is proven (under some assumptions later removed in
[DAV] VIl.4) that each irreducible component of 9J1 m / IF is determined by the
characteristic 0 intersection VnA[pm]o(Qp). We only need a weaker assertion
that the p-isogeny type only depends on components. Indeed, the p-isogeny
type is locally constant on S h K = /F)1I] by the Serre-Tate deformation theory
(see Theorem 8.9 and the argument around (8.25)). Thus IK,\; \G 1 (71 p)/DI 2>
l.em/rvl, where V rv V' if V n (A[pm]o(Q p)) = V' n (A[pm]o(Qp)). It is easy to
show that the number l.em/rvl is equal to IK,\;\G 1 (71 p )/K'\;) I, which shows
the desired assertion (and also the bijection between .e m / rv and irreducible
components of 9J1 K=/IF). We refer the reader to the book of Faltings and
Chai [DAV] VIl.4 for more detailed information on the mod p correspondence
[G 1 (71 p)O:pG 1 (71 p)]. 0
7
Shimura Varieties
In this chapter, we describe a construction of Shimura varieties of PEL type

following [Sh5], [D2], and [Ko]. The construction is given in the first section.
The last theorem (Theorem 7.7) follows basically from the discussion of the
first section if we limit our groups G to those described in the first section;
however, the theorem is formulated for more general groups.
Shimura originally constructed canonical models in the 1950s to the 1960s
as a tower of quasi-projective geometrically connected varieties (over a tower
of canonical abelian extensions of the reflex field) with a specific reciprocity
law at special algebraic points (in the case of Shimura varieties of PEL type,
they are called eM points carrying abelian varieties of eM type; [Sh5]). His
theory includes interesting cases of canonical models of non-PEL type (e.g.,
Shimura curves YOD(~) for totally real fields F i- Q studied in Section 4.3.3),
but in this book, we restrict ourselves to the case where we have a canonical
family of abelian varieties over the canonical model (so, the construction of
the models is easier, as was basically done in [Sh3]).
Deligne reformulated Shimura's tower as a projective limit of (possibly
nonconnected) models over the reflex field (incorporating the theory of motives
in its scope). We follow Deligne's treatment in order to avoid the definition of
the canonical fields of definition of the connected components, although the
global reciprocity law (Theorem 7.7) essentially contains this information.
Kottwitz extended Deligne's definition of Shimura varieties of PEL type
to a projective limit of schemes over a valuation ring of mixed characteristic
(a,p), when the level is prime to p. Since we are interested in the formal
completion at p of the Kottwitz model (and an analogue of the Igusa tower
over the Kottwitz model), what we use most in the following chapter for p-adic
automorphic forms is Kottwitz's formulation.
Under the notation we introduced for GL(2) and GSp(2n), roughly speak-
ing, the canonical model of Deligne is given by Sh(G, X), the model of Kot-
twitz is given by Sh(p) (G, X), and the model of Shimura is the geometrically
irreducible component of Sh(G, X) which is the image of X+ x 1.

304 7 Shimura Varieties
7.1 PEL Moduli Varieties

We construct the moduli of abelian schemes with specific endomorphism al-
gebra, which is a simple algebra. We fix a rational prime p.
7.1.1 Polarization, Endomorphism, and Lattice
Let D be a finite-dimensional simple CQl-algebra with center F. We write 0 =

OF for the integer ring of F. Let S be a set of prime ideals of 0 over p. We
always assume
(unr) We have an isomorphism Dp = D Q91I] CQlp ~ EBpEs Md(Fp) and Fp/CQlp
is unramified for all pES.
Let * be an involution on D that satisfies TrD/II](XX*) > 0 for all 0 # x E
Doo = D Q91I] lR (a positive involution). Here TrD/1I] : D -+ CQl is the reduced
trace Tr of Dover F composed with the field trace TrFjIQ' We call such an
involution a positive involution. We define a subfield Fo of F by the field fixed
by the involution *. Then Fo has to be totally real because TrFo/II](X2) =
TrFo/lI](xx*) > 0 for all x E Fox is true only for totally real fields. Then we
conclude either F = Fo or F is a totally imaginary quadratic extension of
Fo (a eM field). If F # Fo, the involution * coincides on F with "complex
conjugation" which is the generator of Gal(F/Fo). The involution * is called
an involution of the first kind if F = Fo, and otherwise it is called an involution
of the second kind. We fix a maximal order OD of D stable under * (see [BNT]
V.1 and X.1 for orders). It is an easy exercise to show that we can choose a
maximal order OD stable under any given involution of D. We assume that
the isomorphism in (unr) induces
OD,p = OD Q9z Zp ~ EB Md(OF,p).

pES
We fix a left D-module V of finite type and assume that we have a nonde-
generate alternating form ( , ) : V x V -+ CQl such that (bv, w) = (v, b*w) for
all bED. Write Vp = V Q91I] CQlp and V00 = V Q91I] R We also assume having an
OD-submodule LeV of finite type such that
(L1) L Q9z CQl = V;
(L2) (,) induces Homz p (Lp, Zp) ~ L p, where Lp = L Q9z Zp-
Put C = EndD(V), which is a semi-simple CQl-algebra with involution denoted
by i (which is often not positive) given by (cv, w) = (v, c'w). Then we define
algebraic CQl-groups G, GU, and U by
G(R) = {x E C Q91I] Rjxx" E R X }, U(R) = {x E G(R) jxx" = I} ,

and GU(R) = {x E C Q91I] Rjxx" E (Fo Q91I] R)X}. (7.1)
7.1 PEL Moduli Varieties 305
As before, we write G I for the derived group of G (and GU), and assume
(sc) G I is simply connected with noncompact G I (JR.).
By this assumption, we have the strong approximation theorem valid for G I ,
that is, GI(Q) is dense in GI(A(oo)) (see [Knl). This type of group covers all
the cases of Shimura varieties we have studied so far. If we take D to be a
totally real field F with the identity involution *, V = F2, L = 0 EB c*, and
((x,x'), (y,y')) = xy' - yx', we have C = M 2(F) with involution x f---t x< =
det(x)x- 1 and therefore GU = ResF/i(JIGL(2) and G 1 = U = ResF/i(JISL(2);
so, we are in the Hilbert modular case. If we take D = Q with the identity
involution, V = Q2g and L = 7/.,2g with (x, y) = tx · Jgy, we have C = M2g (Q)
with involution x f---t Jgtx· J 9 1 and therefore G = GSp(2g) and G I = U =
Sp(2g); so, we are in the Siegel modular case. If * is of the second kind, U is
not equal to the derived group G I = Guder = Gder, and G I is often written as
SU (a special unitary group). We have the inclusion G I = SU cUe G c GU
and exact sequences of fppf-sheaves (see Section 8.2.1 for fppf-sheaves),
1 -+ U -+ G ~ G m -+ 1, 1 -+ U -+ GU ~ ResFo/i(JIGm -+ 1,
1 -+ SU -+ U ~ ResFo/i(JI Ker(Norm : ResF/FoGm -+ Gm/ Fo ) -+ 1,
det x II If' If'
and 1 -+ SU -+ GU ------+ ResF/i(JIllJTm x ResFo/i(JIllJTm, (7.2)
where 1/( x) = xx<.

In order to bring a complex structure into the real vector space V00
V 0i(J1 JR., we use an JR.-algebra homomorphism h : C '-+ Coo = C 0i(J1 JR. with
h(z) = h(z)<. We call such an algebra homomorphism L-homomorphism. Then
h(i)< = -h(i) for i = A and hence x P = h(i)-1x<h(i) is an involution of
Coo. We suppose
(pos) The symmetric real bilinear form (v, w) f---t (v, h(i)w) on Voo is positive-
definite.
Thus we have 0 < (xv, xv) = (v, (xxP)v) for all 0 -=1= v E Voo and x E Coo, and
hence xx P only has positive eigenvalues. In particular, p is a positive involution
of C (i.e., Trc/i(JI(xx P) > 0 unless x = 0). Such an h : C -+ Coo induces a
homomorphism h : § := ResCjIRGm -+ G /IR of algebraic groups over R We
can easily verify that this homomorphism satisfies the conditions (Dl-2) in
the following section, and to have an algebra homomorphism h : C -+ Coo
as above is equivalent to having a group homomorphism h : § /IR -+ G /IR
satisfying (Dl-2). Any two homomorphisms satisfying (pos) are conjugates
under GI(JR.)+ (see Section 7.1.5). Fix one such h := ho : § -+ G satisfying
(Dl-2), and define X = Xc for the collection of all conjugates of ho under
G(JR.). On X, G(JR.) acts by conjugation (from the left), and the stabilizer
Co c G(JR.) of ho is a maximal compact subgroup of G(JR.) modulo center by
(pos). We can also think of the set Xcu of all conjugates of ho under GU(JR.).
The connected components X(j and X(ju containing ho of these two spaces
are identical.
Since h : C -+ Coo is an lR-algebra homomorphism, we can split VIC =
V ®IQI C into the direct sum of eigenspaces VIC = VI E9 V2 so that h(z) acts on
VI (resp. V2 ) through multiplication by z (resp. z); thereby, we get a complex
vector space structure on Voo by the projection Voo ~ VI. Since h(C) C Coo,
h(z) commutes with the action of D; so, ~ is stable under the action of
DIC = D ®IQI C. We get the representation PI : D '-+ Endc(Vd. We define E
for the subfield of C fixed by {o- E Aut(C)lpl ~ pd. If h'(Z) = g. h(z)g-I for
g E GU(lR), hi induces a similar decomposition VIC = V{ E9 V;, and g induces
a D-linear isomorphism between VI and V{; so, E is independent of the choice
of hi in the GU(lR)-conjugacy class of h. This field E is called the reflex field
of (G, X) (and is a canonical field of definition of our canonical models of
the Shimura variety). By the positivity (pos), the quotient complex torus
Voo/L = VdL has a Riemann form induced by (-, .). The theta functions with
respect to the associated Hermitian form (.,.) in (pos) gives global sections
of an ample line bundle (e.g., [ABV] Chapter I) on Vd L and hence embed
VI / L into a projective space over C. The embedded image is the analytic
space associated with an abelian variety A h / IC • Multiplication by b E OD on
Vd L induces an embedding i : D '-+ EndlQl(Ah/d. Since (bx, y) = (x, b*y),
i(b) H i(b*) is induced by the Rosati involution of the polarization of the
Riemann form (', .). The representation PI is given by the action of D on the
tangent space Lie(Ah) = VI at the origin of A h . Since Ah is projective, the
field of definition of the abelian variety Ah is a field of finite type over Q. By
Corollary 6.11, the couple (Ah'" : D '-+ EndlQl(Ah)) is defined over a field of
finite type over Q. By definition, the field E is the field of rationality of the
representation of D on Lie(Ah); so, the field of definition of (Ah' ,,) always
contains this field E. It would then be natural to expect that the moduli
variety of triples (A, A, ,,) for an abelian variety A with D-linear isomorphism
Lie(A) ~ VI is defined over E.
Since the isomorphism class of PI is determined by Tr(PI) (see [MFG]
Proposition 2.9), E is generated over Q by Tr(pI(b)) for all bED. The field E
is a subfield of the Galois closure pgal of P /Q, and E is a finite extension of Q.
Moreover, under (unr), p is unramified in E/Q. We write OE for the integer
ring of E. Let Z(p) = Zp n Q and put OD,(p) = 0 ®z Z(p), O(p) = 0 ®z Z(p),
and V = OE ®z Z(p). The ring V is a semi-local Dedekind domain with
quotient field E in which p is unramified by (unr).
Let K(p) be an open compact subgroup of G(A,(poo)). We study the clas-
sification problem of the following quadruples (A,A,i,17(P))/s: A is a (pro-
jective) abelian scheme over a base S, t A = Pic~/ s (A) is the dual abelian
scheme of A, A : A -+ t A is an isogeny with degree prime to p (prime-to-p
isogeny) fiber-by-fiber geometrically induced from an ample divisor (polariza-
tion), i : OD,(p) '-+ End~(P) (A) = Ends(A) ®z Z(p) is a Z(p)-algebra embed-
ding (taking 1 to the identity of A) with A 0 i(a*) = 4(a) 0 A for all a E OD,
and 'Pj(p) is the level K(p)-structure. In other words, regarding t A as a left

OD-module by OD 3 b r-+ ti(b*) E EndeA), ,X is D-linear. Hereafter we call
,X D-linear in this sense. The base scheme S is assumed to be a scheme over
Spec(V).
We clarify the meaning of the level K(pLstructure 'Pj(p). Fix a base (geo-
metric) point s E S and write As for the fiber of A at s. We consider the
Tate module T(As) = ~NA[N](k(s)) and V(p)(A) = T(A) 0z A (p<XJ) ,
where N runs over all positive integers ordered by divisibility. The prime-
to-p level structure Tf(p) : V(A(Pexo)) = V 01Q1 A(Pexo) So! V(p)(As) is an OD-
linear isomorphism. The duality pairing eN : A[N] x t A[N] --+ /IN composed
with ,X gives, after taking the limit with respect to N, an alternating form
(-, .) .. : V(p)(A s) x V(p)(As) --+ A(pexo)(l) := Q 0z ~pfN/lN (see (Pl-3) in
Section 8.2.3). Hereafter we require that Tf(p) send the alternating form (.,.)
to (-, .) .. up to the scalar multiple by (A(Pexo))x . This is possible, because
A(pexo)(l) So! A(Pexo) up to the scalar multiple by (A(Pexo))x . Then Tf(p) is re-
quired to be an isomorphism of skew Hermitian D-modules with respect to
the pairing (-,.);.. on T(A).
The algebraic fundamental group HI (S, s) (cf. Section 4.4) acts on V(p) (As)
leaving the skew Hermitian form (-,.) .. stable up to the scalar (because it
keeps the Weil eN-pairing; see [ABV] Section 20). We write 'Pj(p) for the orbit
Tf(p) oK(p) and suppose that u 0 'Pj(p) = 'Pj(p) for all u E Hl(S,S) (this is a way
of describing that the level structure Tf(p) is defined over S). Even if we change
the point s E S, everything will be conjugated by an isomorphism (as seen in
Section 6.4.1); so, the definition does not depend on the choice of s as long as
S is connected. For nonconnected S, we choose one geometric point at each
connected component.
As examples of K(p) and open compact subgroups of G(A(exo)), we could
offer the following subgroups for L = L 0z Z,
r = {x E G(A(exo))lxL = L}, rep) = {x E rlxp = I};
r(N) = {x E rlxt' == t' mod NL W E L for an integer p f N > o}.

An open compact subgroup K = Kp x K(p) of G(A(exo)) is called maximal at
p if Kp = rp for the p-component rp of r as above.
Since a maximal compact subgroup of GLD(V(A(exo))) = AutDA(V(A(exo)))
is the stabilizer of a Z-lattice in V(A(exo)) stable under a maximal order OD,
we find a lattice L with L 0z Z(p) stable under K(p), where Z(p) = I1ih 71..£
for t' running through all primes different from p. Changing L by a sublattice
of p-power index if necessary, we may assume that L satisfies the conditions
(Ll-2). We call a quadruple A/ s = (A,'x, i, 'Pj(p))/s isomorphic to Ais =
(A',A',i','Pj'(p))/s if we have an OD-linear isogeny ¢: A --+ A' defined over S
such that p f deg(¢), ¢* A' = t¢o,X' 0 ¢ = c'x with c E 71..(p)' ¢oi 0 ¢-1 = i', and
'Pj' (p) = ¢ 0 'Pj(p). Thus ¢ brings the prime- to-p polarization class ):.' = {cA'l c E
Z(p)} of )..' to the class:\" of )..: ¢*:\"' = :\". In this case, we write A ~ A'. We
write A ~ A' if the isogeny is an isomorphism of abelian schemes; that is,
deg(¢) = l.
We take the fibered category C = Cv,Y of the quadruples (A,)..,i,'i](P))ls
over the category V-SC H of V-schemes and define
\ . -(p)) IS' (A' ,/\\' ,Z., ,7)

H omc/ s ((A ,/\,z,7) -,(p))
IS )
_ { 'It¢o)..'o¢=c).. with O<CEZ(p),}
- ¢ E Homs(A, A) ®z Z(p) .. _ (p) _(). (7.3)
¢ 0 Z = z' 0 ¢ and 7)' = ¢ 0 7) p
We consider the functor E};) : So-SCH --+ SETS given by
E(p) (S) = E};) (S) = {£lIS = (A, :\", i, 'i](p)) IS I A satisfies (det) below} I ~ .
Here the determinant condition (det) is given as follows. Recall V = 0 E ®z
Z(p) for the integer ring OE of E. As already remarked, V is a semi-local
Dedekind domain with quotient field E. The condition (unr) guarantees that
ov ® V 1m is semi-simple for all maximal ideals m c V. We choose a Z(p)-base
{aJh:S:J:S:r of OV,(p) and consider a homogeneous polynomial
f(t 1 , ... , tr) = det(altl + ... + attrlvJ.

Then f(t) E V[tl,"" t r ], and the coefficients of f(t) generate E over Q. Here
OE is the integer ring of E. Since A ls is a group scheme, its tangent space at
the zero section has a Lie algebra structure over Os. We write Lie(A) for this
Lie algebra. Since A is smooth over S, Lie(A) is a locally free Os-module of
rank dims A (see [NMDJ Proposition 2.2.5). In our case, for a given quadruple
.d = (A,)..,i,'i](P))IS, the Lie algebra Lie(A) of A over Os is an OV,(p) ®zOs-
mod ule via i. Since Lie (A) is locally free of rank dims A over Os, we can think
of g(t 1 , ... , tr ) = det(alh + ... + artrIL,e(A)) E Os[h, ... , trJ. We impose
j(j(tl, ... , tr)) = g(tl, ... , tr) in Os[t], (det)
where j : V --+ Os is the structure homomorphism. Since f(t) contains as

coefficients all data of the characteristic polynomial of Pl (a), this condition
is equivalent to (det) in Section 6.l.6. Over a field of characteristic 0, one
can characterize representations of a semi-simple algebra by their trace, but
over a general base, we need (as explained in Section 6.l.6 around the con-
dition (det) there) the entire characteristic polynomials to determine a given
representation; so, the determinant has to be fixed as above.
In the rest of this section, we give a sketch of a proof of the following
theorem.
Theorem 7.1 Let the notation and the assumptions be as above. If the open
subgroup K of G(fo.,Joo)) maximal at p is sufficiently smail, the functor E};)
is representable by a quasi-projective smooth scheme ShY::) over V. For any

K maximal at p, the coarse moduli scheme ShY::) of []I) exists as a quasi-
projective scheme of finite type over V. If D is a division algebra and V = D,
Sh~~v is projective over V.
Even if D = F and V = F n with n > 1, ShY::) is projective if the Hermitian

pairing (-,.) is anisotropic (i.e., (v, v) = 0 <=? v = 0 for v E V), because the
polarization pairing has to have an isotropic subspace given by the toric part
of the semi-abelian scheme if the universal abelian variety degenerates into a
semi-abelian scheme.
The generic fiber of the variety Sh~jE actually could be a finite dis-
joint union of Shimura varieties S h (p) ( G, X G) associated with different pairs
(G, X G) in the sense of Deligne, depending on the choice of (D, V, (., .)) and
G, because our data only determine the Hermitian form (-,.) over D)!:oo) not
over D; so, if the Hasse principle fails for such Hermitian forms (i.e., the
isomorphism class of (VA(POO) , (-, .)) over D)!:oo) does not determine uniquely
the isomorphism class (V, (-, .)) over D), there could be different groups G IQ
associated with the functor []I). This delicate point is studied later in Sec-
tion 7.1.5.
7.1.2 Construction of the Moduli
If two abelian schemes A and A' over S are isogenous by a prime-to-p isogeny
over S, we write A ~ A', abusing our notation slightly. If K(p) is sufficiently
small so that Auts(A, A, i, ry(p)) = {lA}, the prime-to-p isogeny giving the
isomorphism A ~ A' in the definition of [(p) can be taken (as explained in
Section 4.2.1) to be an isomorphism by changing A' in the isogeny class under
"~" (and insisting 7](p)(Dp)) = T(A) @z Z(p); d. [D1] 4.10). Therefore we
have []I)(S) S:! [K(S) for another functor [K : SCHlv -+ SETS given by
Here Ais is an abbreviated form of (A,A,i,ry(P))ls. Under this setting, we

change the morphism set of CIS from Homc / s to Isomc / s :
' . -(p)) IS, (A' ,A,I ,2·1 ,7]

I somc/ s ((A ,A,2,7] -I(P))
IS )
= { ¢ E Isoms(A,A)
'I ¢ t¢
0
0
i
A' 0 ¢ = CA with c E Z(), }
= i' 0 ¢ and ry'(P) =¢0
P
ry(p)
. (7.4)
Then we claim that the fibered category Cover V-SC H is an algebraic stack
for any given K(p) and is an algebraic space if Autc/s (A) = {lA} for all
objects A of CIS. See [DM] and [DAV] 1.4.6-9 for the definitions of algebraic
310 7 Shirnura Varieties
stacks, but this statement is basically equivalent to the fact that £jf) has a
coarse moduli scheme over V (compatible with base change). Indeed, if K(p)
is sufficiently small, £jf) is representable by a quasi-projective scheme over V
as we show.
Forgetting D-linearity of 7)(p) and restricting i to Q, we have a functor
from CD,v into the fibered category CIQJ,v of (A,.A, i, fj(p)) for D = Q for a
suitable choice of an open compact subgroup GSp(2d)(A(PCXJ)) (d = dim A =
~ dimlQJ V). As we have shown in Theorem 6.20, this fibered category is proven
to be an algebraic stack and is representable by a quasi-projective scheme
M = MIQJ/v by Mumford (see also [CIT], [Sh3], [CSM], and [DAV]).
We now supplement the above outline with details. We show that we
can replace ~ by ~ in the definition of the functor £);), imposing an ad-
ditional condition. Let A/s = (A,.A,i,fj(p))/s. Then 7)(p) induces V(A(PCXJ)) ~
V(PCXJ)(A) and, therefore, we have 7)(p) ([)p)) c V(PCXJ)(A) for [)p) = L0z7!,(p).
If 7)(p)([)p)) c T(p)(A), the cokernel is an etale group subscheme C c A/ s
locally free over S of rank prime to p. Make the quotient abelian scheme
A' = AIC over S (see [ABV] Section 12 and [CME] Proposition l.8.4) , and
then we have a prime-to-p isogeny cp : A -+ A' with Ker(cp) = C. We then
consider A' = (A', .A', i', 7)'(p))/s given by .A = ¢ 0,Ai 0 cp, i'(a) = cp 0 i(a) 0 cp-1,
and 7)'(p) = cp 0 7)(p). Then A' satisfies 7)'(p) (j}p)) = T(p) (A).
If 7)(p) ([)p)) ~ T(p) (A), we can find a prime- to-p isogeny A' -+ A such that
cpo 7)'(p) = 7)(p),.A' = ¢O.A 0 cp, i'(a) = cp-1 0 i(a) 0 cp and 7)'(p) ([)p)) = T(p)(A).
This fact follows from the canonical identification T(A) = 11"1 (A, 0) for the
origin 0 of A (see [ABV] Section 18).
If neither 7)(p) (j}p)) C T(p) (A) nor 7)(p) (j}p)) ~ T(p)(A), we can find two
prime- to-p isogenies A LA" and A' ~ A" for A" = (A",.A", i", 7)" (p)) / s so
that 7)"(P) ([)p)) = 7)(p) (j}p)) n T(p) (A") and 7)'(p) (j}p)) = T(p) (A'). Thus we
can always find in the prime-to-p isogeny class of a given A/ s , a quadruple
d;s with 7)'(p) (j}p)) = T(p)(A').
If cp : A/ s -+ A;s is a prime-to-p isogeny with 7)(p) (j}p)) = T(p)(A) and
7)'(p) (Dp)) = T(p)(A'), then deg(cp) = 1 and A/s ~ A;s by cp. Thus insisting
7)(p)([)p)) = T(p)(A), we can replace ~ by ~ in order to define the functor
£jf). In other words,
The functor defined in this way can be proven to be representable (as now
show) by a V-scheme Sh~jV"
Since r(pCXJ)(N) for N prime to p gives a fundamental system of neigh-

borhoods of the identity in G(A(PCXJ)), we may assume that K = r(N) =
r(pOO) (N) x rp. We only need to show that CD,v is relatively representable
over the Mumford moduli MIQ given by Theorem 6.20. In order to emphasize
the dependence on (D, V, (-, .)), we write £fi for the functor £K with respect
to K and D. Write 2d = dimlQ(V). Then d = dims A for Als E £fi(S),
and d is independent of the choice of AI s by (det). For simplicity, we as-
sume that the polarization pairing ( , ) in (L2) gives the self-duality of L.
We then identify with GSP2d(Z) the similitude group of ( , ) acting on L. In
this way, we let GSp2d(R) act on V(R) = V 01Q R. Write Ko for the open
compact subgroup of GSp2d(A.(oo») maximal among those preserving Land
the level N structures (the group Ko was written as r(N) in Section 6.4.1).
Then Ko n G(A.(oo») = K and K = r(N) c G(A.(oo»). As described in Theo-
rem 6.20, £Ko is representable by a quasi-projective scheme M = MIQ = wtN
defined over Z(p). Let A = (A, A, r;(P»)IM be the universal quadruple over
M with the universal abelian scheme A E A. We consider the functor from
M-SCH into SETS:
TIS H [(AT,AT,i: OD '-+ End T (A T ))li(ID) = idA, Xii 0 ti 0 AT = i 0 *].
This functor is representable by a scheme MDIM (Corollary 6.11). Since the
level structure r;(p) on A gives rise to a level structure fj(p) of (AT, i), we
have a triple A = (AT,i,fj(P»)IT. Thus £fi is a subfunctor of the above
functor. Again by the rigidity of endomorphisms under specialization (de-
scribed in Section 6.1.6), £fi is represented by a closed subscheme Sh~jv
of MD whose connected components are (each) finite over MIQ. We show
that Sh~) is of finite type over MIQ (so it is projective and finite over MIQ).
Take a geometric point x E MIQ; suppose that we have i : OD '-+ EndA
as above, which gives rise to a geometric point y E Sh~). For a given T,
if T is connected, End(AT) 0;z Q is a semi-simple algebra of finite dimen-
sion with positive involution (see [ABV] IV). Thus the number of embedding
D '-+ End(AT) 0;z Q is finite up to inner automorphisms. Moreover, the num-
ber is bounded by a constant only depending on the dimension of A, that is
d, because dim(EndT(A T ) 0;z Q) is bounded by 4d 2 (e.g., [GME] Theorem
4.1.19). If one changes i by an inner automorphism induced by ex E End(AT)
and if we suppose that (AT,exiex-l,).,fj(p») is still an element of £fi(T), it is
easy to see, by the condition that r/ p) (Ltp») = T(p) (A) combined with (LI-2),
that ex has to be an automorphism of AT. Since automorphisms of an abelian
variety preserving a given polarization are only finitely many by the positivity
of the polarization ([ABV] Theorem 5 in Section 21), there are only finitely
many possibilities of ·having i : OD '-+ End(AT) that give rise to an ele-
ment of £fi(T). Thus Sh~jv -t MIQIV is quasi-finite. Then the projectivity of
each connected component of Sh~) over MIQ implies the finiteness of the map
Sh~) -t MIQ. If K is small, the morphism is unramified and finite. Actually,
one can show that the morphism Sh~) -t MIQ is a closed immersion (over Q)
if N is sufficiently large (cf. [D 1] 1.15; see Section 8.4.2 in the text).
The rigidity of endomorphism of abelian schemes under specialization com-

bined with the deformation theory of abelian schemes tells us, for N suffi-
ciently large, Sh~)v is finite over M Q / v and smooth over V (see Theorem 8.8,
the remark after the theorem, and [LaR] 6.2).
Although we assumed that L is self-dual, replacing GSp2d by the simili-
tude group of (-,.) (which is a conjugate in GL(2d) of GSp(2d)) and using
the Mumford moduli MD,N for the positive square root (Pffafian) D of the
discriminant of (', . ), we can easily generalize the above argument to a given
(nonself-dual) polarization of degree prime to p.
If D is a division algebra and V = D, the Shimura variety skC p ) is known to
be projective over V. Indeed, if the image of MD intersects with the cuspidal
divisor of the toroidal compactification M Q described in Section 6.3.6, the
pullback of the semi-abelian scheme 9/ MQ to MD yields a non-abelian semi-
abelian variety 9D/k with D C EndQW/k) as a fiber over a geometric point
Spec(k) E MD. Since 9D is not an abelian scheme, its toric part T C 9D
is nontrivial and stable under all endomorphisms of 9D (rigidity of the toric
part; see [NMD] 9.2). Thus D acts on its character group X(T) ®z Q, whose
dimension over Q is less than dim 9D = ~ dimQ V. This is impossible since
rank D V = 1. Thus MD never intersects with the cuspidal divisor of M Q and
hence is projective (by the valuative criterion of properness). This finishes the
proof of Theorem 7.1. We record here a byproduct of our proof.
Corollary 7.2 Suppose the pairing in (L1-2) induces a self-duality on L.
Then forgetting OD-linear structure Ell --+ E~t induces a finite morphism of
Sh~) into Sh~~ (GSp(2d), X) and induces a finite unramified morphism if K
is sufficiently small, where 2d = dimQ V and X is the symmetric space of
GSp(2d)(JR).
We now replace prime-to-p isogenies in the definition of EX') by isogenies
and prime-to-p structure 7](p) by the full level structure 7] : V(A(oo)) ~ V(A).
We define a functor Ejp :
E-SCH --+ SETS given by
Ei:) (S) = {A/ s = (A, "X, i, iJ) / s I A satisfies (det) and (pol) below} / ~ .
Here ~ is up to isogeny (not up to prime-to-p-isogeny). Let hA : § --+
Autc(H1 (A, JR)) for § = ReslC/IRGm be the homomorphism inducing the Hodge
decomposition of H 1 (A,Q (see just above (pol) in Section 4.2.1). Note that
Vs(A) = H 1 (A,A(00)) = H 1 (A,Q) ®Q A(oo), and recall the condition (pol)
adjusted to our setting:
(pol) There exists an D-linear isomorphism I : V ~ HI (A, Q) such that
1- 10 hA 0 fE Xc, (f ® 1A,(oo») E (7]0 K) and EA (f(x), f(y)) = a(x, y)
up to a E QX,
where EA is the Riemann form on H 1 (A,Q) (see [ABV] Section 1). This
condition really determines the group G as a similitude group of EA under
the identification by f.
In the same way, we can prove the represent ability of the version E<jJ)
of the functor EAf) replacing in the definition of EAf) prime-to-p-isogenies by
isogenies. The new functor E<jJ) therefore gives rise to the Shimura variety
Sh(G, X) defined over the reflex field E. Recall the warning we made after
Theorem 7.1: the condition (pol) is not imposed in the definition of EAf); so,
EAf) restricted to E-algebras may be a disjoint union of E<jJ) for a different
choice of an alternating form on V which is locally equivalent to (-, -). This
subtle point is studied in Section 7.1.5. See Theorem 7.7 for the description
of the variety Sh(G, X).
Depending on the data at the cusps governing toroidal compactifications,
a semi-abelian scheme 9/ MQ is constructed in [DAV] (see Section 6.3.6). Then
even in the quasi-projective (but nonprojective) case, a similar argument using
Corollary 6.11 (applied to 9/ MQ in place of A/ MQ ) gives a scheme projective
over the smooth toroidal compactification M IQl of the Mumford moduli MIQl.
Since the endomorphism algebra of an abelian variety A/k for an algebraically
closed field k (after tensoring Q) is semi-simple, there are only finitely many
possibilities of embedding D into Endk(A) ®z Q up to conjugation. Thus
the morphism M D -+ M IQl has finite geometric fiber everywhere; that is,
the morphism is quasi-finite. Since the scheme M D is proper over M IQl (see
Corollary 6.11), it has to be finite. Thus writing ~ = f*J2 A / MD for f : A -+
MD and defining det(~) by its maximal exterior product, we can define a
graded algebra:
00
n=O
Moreover, as seen in Section 6.3.6, MQ = Proj(IBl§) and hence we have the
minimal compactification of Sh K defined by Sh'K = Proj(IB}§) finite over
M Q.
The smoothness of M D /V around the cusps for a well-chosen cuspidal
datum was shown by Fujiwara for G and GU of types A and C ([F]). If we
choose the cuspidal data for GSp(2d) and G so that they are compatible
(in other words, so that the pullback of the semi-abelian scheme over M IQl is
the semi-abelian scheme over M D associated with the cuspidal data for G),
this guarantees that the q-expansion parameter is well defined over V and
projectivity of M D over V (if the level is prime to p), because it is finite over
projective MIQl.
Even if K is not very small, we always have a coarse moduli scheme ShY::)
representing the functor Ell over V. The above arguments all work well. We
write Sh K for a toroidal compactification of Sh K and Sh'K for the minimal
compactification. Since the natural morphisms
Sh'K -+ MQ and Sh K -+ MIQl
are quasi-finite and projective, they are finite. Let V be the image of Sh'K
in M Q. Then V = Proj(IB'K) for a graded algebra IB'K which is the quotient
of <Bff. Then, assuming the existence of smooth toroidal compactification of

Sh K , we have, for any open compact subgroup K c G(A(oo») maximal at p,
Sh'K = Proj(<Bfj). (Proj)
Here <Bfj is the integral closure of <B'K in the total quotient ring of <B'K. This
follows from the fact that Sh'K is smooth outside cusps, and at the cusps, if K
is sufficiently small, M D is finite over M Q (and normal over V). The graded
algebra <Bfj is the graded algebra of automorphic forms on G if dim Sh K > 1.
7.1.3 Moduli Variety for Similitude Groups
Recall that Fo is the subfield of F fixed by the involution *. We write 0 0

for the integer ring of Fo. We have formulated the moduli problem for the
group G with similitude norm v : G -+ Q. But we can consider polarization >.
modulo totally positive elements in vt for Vo = 0 0 ®z Z(p). Then we obtain
the classification problem for the Fo-similitude group GU/ Q . Note that the
connected component X+ of the symmetric spaces of GU and G are identical
and that GU has a nice property that its center Z S:! ResF/QG m is cohomolog-
ically trivial (i.e., H;t(Gal(M/Fo), Z(M)) = 0 for all Galois extensions M/ Fo;
Hilbert's Theorem 90).
The group ResF/QGL(2) (for F = Fo, D = F, and V = F2) we considered
in Section 4.2 is a group of this type; so, we have a detailed exposition of
the Shimura variety of GU in this special case of Hilbert modular functions.
Although we have written this group as GU, as this example shows, the group
GU may not be a unitary group.
We now give some details of the Shimura variety for general GU. To define
the moduli problem of abelian schemes associated with GU/ Q , we need to
modify slightly the morphisms of the fibered category CD,v, We define a new
fibered category A(p) = Ag)v over V-SCH to be the category of quadruples
A/ s = (A, >., i, Tj(p») for Tj(p) '= 'T}(p) mod K, where K c GU(A(oo») is a closed
subgroup maximal at p. Recall that 0 0 is the integer ring of Fo and that
Vo = 0 0 ®z Z(p). Again Vo is a semi-local Dedekind domain under (unr).
Then we define
Horn A(P) (A/ s , A;s)
{¢ ¢
/S
= E Hom(A, A') ® Z P
()¢
I;;; 0
0
>.' 0 = >. 0 i(a) with a E vt+} , (7.5)
i = i' 0 ¢ and Tj'(p) = ¢ 0 Tj(p)
where vt+ indicates the group of totally positive units in Vo. We then consider
the functor
£~(P) (S) = {AlsiA satisfies (det)} / ~,
where ~ indicates isomorphism classes in Ag~s'
7.1 PEL Moduli Varieties 31G
We may allow all isogenies as morphisms to define the category A (0) (with
full level structure) and define a functor from E-SCH to SETS by
Eft) (S) = {AISiA satisfies (det) and (pol) below} / ~,
where ~ indicates isomorphism classes in A~~ s. Note that Vs (A) is canonically

isomorphic to HI (A, 11(00)) = HI (A, Q) ®iQl 11(00). Recall
(pol) There exists a D-linear isomorphism 1 : V ~ HI (A, Q) such that
(J ® 1",(=)) E (7) 0 K), 1- 10 hA 01 E Xc, and (J(x), I(Y))).. = a(x, Y)o
up to a E Fox,
where TrF'o/iQl((-,·).>.) = EAC,·) for the Riemann form EA on H 1 (A,Q) and
similarly TrFo/iQl( (-, ')0) = (-, .).
In the Hilbert modular case, we constructed the modular variety first rep-
resenting the functor [r( N) with a fixed polarization. Since Z~ = {I}, fixing
polarization is a moduli problem of G(Q) = {g E GL 2 (F)1 det(g) E Q},
because of [;~)N) ~ [r(N) proven in Section 4.2.1. Then we extended the
construction to GU = ResFIiQlGL(2) by forming quotients of a disjoint union
of the G-moduli 9J1, by subgroups of O~ = v(GU(Z)) (like construction of
9J1 r !(lJt) in Section 4.1.2). In the general case of GU, the Hermitian symmetric
domain X+ is the same for G and GU. The group GU is the similitude group
of the Hermitian pairing (-, -)0 : V x V -+ Fo given by TrFo liQl( (-, ')0) = (', .).
As we show in Section 7.1.5, the local equivalence class (the genus class) of the
Hermitian alternating form (-, ')0 under GU is trivial, but modulo G(Q), there
may be several subclasses (giving rise to a finite number of unitary groups
G2/iQl with G 2(11) = G(11); see Remark 7.4). Following the same process (as
in the Hilbert modular case), starting from a disjoint union U2 Sh( G 2 , X) of
Sh(G 2, X) (and making quotients by subgroups of v(G 2(Z)) = v(G(Z))), the
above functors [R(0) and [R(P) can be proved to be representable if K is suffi-
ciently small. We write Sh( GU, X) IE (resp. Sh(p) (GU, X) IV) with X = Xcu
for the scheme representing [R(0) (resp [R(P)), because in this case, we prove
that Sh(p) Xv E ~ Sh(GU, X)jGU(Zp) in Theorem 7.5.
The compactification of the moduli space MK also can be done as de-
scribed above. The only point we need to make explicit is that if the class
:.\ of polarizations modulo multiplication of totally positive elements in Vox
is defined over S, we can always find a representative). defined over S. In
other words, for each geometric point s E S, if the action of 7fl (S, s) pre-
serves :.\, then the question is, "Can we find ). : A -+ t A well defined over
S7" Taking an etale finite Galois extension S' / S over which), is well de-
fined, we have a Galois 1-cocycle c : Gal(S' j S) :3 IJ -+ ).1-0' with values in
the trivial Gal(S'jS)-module Vo\ = Vox nFo~' Since Gal(S'jS) is finite and
c: Gal(S' / S) -+ Vo\ is a homomorphism with values in a torsion-free module,
we find c is triviaL and), is indeed defined over S.
We can also show that 2>' is globally induced by a line bundle on A rela-
tively ample over S as follows: picking one symmetric polarization >., the pull-
back LL1(>.) by 1 x >. of the universal line bundle over A Xs t A/A (the Poincare
bundle) is always ample and is equal to 2>' (see [GIT] Proposition 6.10); so, in
the class "X, we can always find a polarization globally over S associated with
a line bundle on A.
7.1.4 Classification of G
We define an algebraic group GO/ Fo by Go(R) = {x E C0Fo Rlxx" = I} for

Fo-algebras R. Then we have U = ResFo/«:JiGO. The involution * of D either
induces a nontrivial involution on F (a positive involution of the second kind)
or the identity map on F (a positive involution of the first kind). If * is of the
second kind, F is a CM field over the totally real Fo, and * coincides on F
with the unique nontrivial automorphism over Fo (see [Sh] and [ABV] Section
21). Then Go is an inner form of a quasi-split unitary group over Fo. We call
this case "Case A" and call the group G "type A" (following the classification
of Lie algebras; see [BLI]).
When * induces the identity map on F, then F = Fo is totally real, and
the group Go is an inner form of either the symplectic group (Case C and the
group of type C) or an orthogonal group of even variables (Case D and the
group of type D). We have (cf. [Sh] Sections 1 and 2 or [Ko] Section 4)
Mk(CYO and x" = Is,/xIs,t in Case A,

{
Coo ~ M2k(lR)Io and x" = JktxIk1 in Case C, (7.6)
Mk(lliIfo and x" = -itxi in Case D,
where lliI = lR + lRi + lRj + lRk is the algebra of Hamilton quaternions, 10 is the
set of all field embeddings of Fo into lR, Jk = ( 10k -Jk) and Is,t = (~ -~t) are
as before, and x f-t X is either complex conjugation or quaternion conjugation.
There are type-B (orthogonal) groups admitting Shimura varieties, but
they are non-PEL type except for the case when the signature is either (2,2)
or (2,1) (and hence the groups are isogenous to the product of inner forms
of PGL(2)). In Case D, the simple connectedness (sc) fails, but the argument
proving Theorem 7.1 still (almost) works assuming p > 2 (see [Ko]), and we
get Sh)~. When K(p) is sufficiently small, Sh K is smooth over V. This follows
from the fact that the deformation ring of a quadruple (A, >., i, r;(p)) is always
formally smooth (see [LaR] 6.2, also Theorem 8.8 in the text and the remark
after the theorem). If C = EndD(V) is a division algebra ({o} D is a division
algebra and rank D V = 1), as we have already seen, Sh~jv is projective.
Since the stabilizer Co of the fixed ho E X in G(lR) is a maximal compact
subgroup modulo center by (pos), the connected component of X is isomorphic
to the symmetric domain H = G(lR)/Co . An explicit form of H as a classical
bounded matrix domain is given in [Sh] (see also [ACM] Chapter VI for the
domains in Cases A and C), along with an explicit method of constructing

all possible analytic families of abelian varieties over the domain. We have
computed H already in Section 5.2.2 for unitary and symplectic groups (which
are groups of types A and C). The complex analytic space ShK(r.c) is given by
G(Q)\G(A)/ KCo = G(Q)\(X x G(A(oo»))/ K, and its connected component
is given by r\H for r = (gKg-1G(JR;.)+) n G(Q) with a suitable 9 E G(A(oo»).
7.1.5 Generic Fiber of ShY;)
In the definition of the functor [jJ'), the structure of the Tate module
V(p)(A) = H1(A,Q) ®IQI A(poo) is specified modulo K for a test object
(A, X, TJ(p») over a characteristic 0 field inside C However, this information
may not uniquely determine the D-Hermitian structure of Hl (A, Q) which
supplies us with the exact information of the algebraic group giving rise to
the Shimura variety; so, the variety Sh~) representing the functor [jJ') may
be a disjoint union of the reduction modulo p of characteristic 0 Shimura va-
rieties associated with different groups locally isomorphic to each other (but
not globally). We study this point in this section to relate Sh(p) to the char-
acteristic 0 fiber of the Shimura variety.
A (D, *)- Hermitian form (.,.) is a symmetric bilinear form on V with
values in Fo such that (bx, y) = (x, b*y) for all x, y E V. The given alternating
form (-,.) : V x V --+ Q can be uniquely lifted to an Fo-linear alternating
form C·)o: V x V --+ Fo so that (.,.) = TrFo/IQI((-,·)o). Pick (A,X,7J(P»)/k E
[f2(P) (k) for an extension field k/ E inside C. Similarly, the Riemann form
EA on Hl (A(r.c) , Q) with values in Q can be uniquely lifted to an Fo-linear
alternating form (., .)" so that TrFo/lQI( (., .».) = E A (-, .). Then the test object
(A, X, 7J(p») /k supplies us with the following information.
The D-module VA := H1(A(r.c),Q) with Fo-linear Riemann form C·)" :
VA X VA --+ Fo and an isomorphism TJ(p) : V ®IQI A(poo) ~ VA ®IQI A(poo)
sending C .) 0 to (-, .)" up to the constant in 0;' (p)"
The Shimura variety S h( GU, X) / E is defined for the group GU; so, our ques-
tion is if the above data are sufficient to determine GU (i.e., to verify (pol)
using information only from defining data of [f2 (p) ). There are three points to
check:
(1) (VIR, C ·)0) ~ (VA ®IQI JR;., (-, .),,), and for any two i-homomorphisms C Y
Coo are conjugates under G(JR;.);
(2) (Vp, C ·)0) ~ (VA ®IQI Qp, C·),,);
(3) the Hasse principle for the alternating form C ·)0.
First we prove (1) following [Ko], Section 2. Recall that 10 is the set of all real
embeddings of Fo (regarded as the set of all Archimedean places of Fo). For
any semi-simple F-algebra B and a B-module M, we put Ba = B ®Fo,a JR;.
and Ma = M ®Fo,a JR;.; so, BIR = B ®IQI JR;. = ITaEl o Ba and MIR = ITaEl o Ma.
~1R 7 Shimura Varieties
Each Dcr has positive involution induced by *, which we still denote by *. Now
a (D, *)-Hermitian form on Vcr is a symmetric lR-bilinear form with (bv, w) =
(v,b*w) for all v,w E Vcr' We start with
Lemma 7.3 Let (.,.) and (.,.)' be two positive-definite (D, *)-Hermitian
forms on Vcr. Then there exists c E C; such that (cv, cw) = (v, w)' for all
V,WEVcr .
Proof. The Hermitian form (-,.) (resp. (', .)') induces an identification ¢ : Vcr ~
Vcr:= HomlR(Vcr,lR) (resp. ¢': Vcr ~ Vcr). We regard Vcr as a Dcr-module by
f H fob*. Since ¢(bv)(w) = (bv, w) = (v, b*w) = ¢(b*w), we find Cl = ¢-lo¢'
is an element in AutD<7(Vcr) = C;. We find (C1V,W) = ¢(cv)(w) = ¢(¢-l 0
¢'(v))( w) = (v, w)'. Since the two forms are (D, *)-Hermitian, we find Cl = ct.
Consider the set C+ = {c E Ccrld = c and c > O}, where c > 0 {:? (cv,w) is
positive-definite. Any positive-definite symmetric inner products of a finite-
dimensional real vector space are equivalent over lR (Sylvester's theorem),
C+ is therefore isomorphic to a positive subcone of the space of positive-
definite symmetric matrices of EndlR(Vcr). In particular, log: C+ ~ 10g(C+)
and exp : log( C+) ~ C+ are well defined by the standard power series expan-
sions. Therefore, we have c = JC1 = exp(log(cd2)) in C+, which gives the
desired element satisfying (cv,cw) = (c 2 v,w) = (C1V,W) = (v,w)'. 0
If we have two L-homomorphism h, h' : C '--+ Ccr, we have two Hermitian

forms (v, w) = (v, h(i)w)cr and (v, w)' = (v, h'(i)w)cr. We find c E C X with
(v, h'(i)w)cr = (v, w)' = (cv, cw) = (cv, h(i)cw)cr = (v, c'h(i)cw)cr. This shows
that h'(z) = c'h(z)c. Since h(i)h( -i) = h(i)h(i)' = 1, we verify cc' = 1
because cd E C+; so, hand h' are conjugates under Go(Fo,cr).
The Hodge decomposition Hl(A, q = HO(A, [lAIc) EEl HO(A, [lAIc) in-
duces, by Poincare duality, an isomorphism hA : C --t End(CIR ) so that hA(Z)
acts on H(A, [lAIc) by multiplication by z and on H(A, [lAIc) by multipli-
cation by z. By the theory of Lefschetz-Poincare ([ABV] Section 1), the Her-
mitian pairing (x, hA (i)y) >. is positive-definite. Since VA 01Qi Qe ~ V 01Qi Qe
for a prime P, V ~ VA as D-modules by faithfully flat descent for QdQ.
Thus applying the above argument to (v, w) = (v, h(i)w) for hEX and
(v,w)' = (V,hA(i)w)>., we conclude hA is a conjugate of h under G1(lR);
so, h A EX. Now we apply the above argument to (v, w) as above and
(v, w)' = (v,·h(i)w)>.; we conclude (1).
As for (3), we consider a functor 9 : k H Isom((V 01Qi k, (-, .)), (VA, (-, .) >.))
for number fields k. Here the isomorphism preserves the alternating forms up
to elements in (ResFo/IQiGm)(k) = (k 01Qi FoV. The group GU acts on the
functor by ¢ H ¢ 0 g; so, obviously 9 gives a GU-torsor (i.e., a principal
homogeneous space over GU) defined over Q. As is well known (see [GCH]
1.5 and 111.1 and [Sp]) , the GU-torsors are classified by the (non-abelian)
Galois cohomology set H1(Q,GU) = Hl(Gal(Q/Q), GU(Q)). Thus the proof
of (3) boils down to showing injectivity of the (diagonal) restriction map
Hq(Q, GU) -+ Hq(lR., GU) x TIe Hq(Qe, GU) for q = 1 (the Hasse principle
for Hq(Q, GU)).
For groups of types Band D, G I is not simply connected; so, often the
Hasse principle fails (cf. [GCH] III.4.7 and [Kn1]). Thus we assume one of the
following two conditions which guarantees the condition (sc).
(B1) The involution * on D is a positive involution of the second kind
(Case A);
(B2) The involution * is of the first kind, and D~ = D 0Q lR. ~ Md(F~) for
F~ = F 0Q lR. (Case C).
These conditions eliminate groups of types Band D.

We now prove (3) under (B1) or (B2). We consider the exact sequence:
1 -+ G I -+ GU -+ ZZU -+ 1 for co center ZGU of GU. Then by the long exact
sequence associated with the above short exact sequence, we get the follow-
ing exact sequence (see [Sp] Proposition 2.2), HI (Q, Gd -+ HI (Q, GU) -+
HI(Q, ZGU). Since G I is simply connected (of type A or C), it is known
that HI(Q, Gd is trivial (see [Kn1]). Thus we need to show either that
HI(Q, ZGU) = 0 or the Hasse principle holds for HI(Q, ZGU). We write
TK = ResK/QG m for a number field K. When the group is of type C,
ZGU ~ T F , because GU is either ResF/QGSp(2g) for totally real F or a
similitude group of a Hermitian form over a totally indefinite quaternion al-
gebra over F (see [We]). Then HI(Q, ZGU) = 0 for type C groups GU follows
from Hilbert's Theorem 90.
If the group GU is of type A, it is a unitary group; so, taking r with Ccr ~
Mr (!C) , we find G I = {x E CINc(x) = v(x) = I} for the similitude norm
v(x) = xx· and the reduced norm Nc : C -+ F. Since C is of dimension r2
over F, we find that Nc(x)Nc(x)C = v(xt for the generator c of Gal(Fj Fo).
This shows that ZGU = {(x, y) E TF X TFo INF/Fo (x) = yr}. If r = 2m is even,
we have ZGU ~ TI X TFo by sending (x,y) f-c> (xy-m,y) E TI X TFo for TI =
Ker(NF/Fo : TF -+ TFo)' From the exact sequence, 1 -+ TI -+ TF -+ TFo -+ 1,
X 1
we have another 1 -+ Ker(NF/Fo) -+ F X ----"-+ Fo -+ H (Q, Td -+ 1. The
NF/FO
vanishing of Hl(Q, T F ) follows from Hilbert's Theorem 90. This shows that
H1(Q, ZGU) ~ Hl(Q, Td ~ Fox jNF/Fo(FX) for which the Hasse principle
holds (the norm theorem of class field theory for cyclic extensions). Thus the
Hasse principle holds for Hl(Q, GU) as desired.
Now assume that r is odd. Let Zl ~ /-lr be the center of G 1 . Then we
have an exact sequence of groups 1 -+ Zl(Q) -+ Z(Q) -+ ZGu(Q) -+ 1,
because over Q, GU ~ (GL(r) x Gm)Io. Since H2(Q, /-lr) is the kernel of
multiplication by r of the Brauer group of Q (Kummer's theory), the Hasse
principle holds for H 2(Q,/-lr) (class field theory; see [BNT] XIII.3). For
the same reason, Hasse's principle holds for H2(Q, TF)' Since Z = T F , by
Hilbert's Theorem 90, Hl(Q, Z) = O. Then by the short exact sequence,
o = HI(Q, Z) -+ Hl(Q, ZGU) -+ H2(Q, /-lr) -+ H2(Q, Z), we conclude the
Hasse principle for H1(Q, ZGU) and hence for H1(Q, GU). This shows (3) for
GU. The Hasse principle sometimes fails for G of type A (but in a rather
harmless way; see Remark 7.4 and [Ko] Section 7).
Now we prove (2). Let Oc = {x E ClxL = L}. We need to prove that
if L~ is a self-dual OD,p-Iattice of Vp = V @IQI <Qlp under (-, ·)0, then we have
an element g E GU(<Qlp) such that gLp = L~. We can do this at each place
p of Fo above p. By (unr) , OD,p = TI'lJlp Md(0'lJ). This shows that Oc,p =
TI'lJlp Mr(O'lJ). Thus we may assume that d = 1 (replacing L by EL ~ og for
the *-invariant idempotent E = diag[l, 0, ... ,0] E OD,p). Then the existence
of g E U(<Qlp) c GU(<Qlp) as above follows from [EPEl, Lemma 5.9 (note here
any self-dual lattice of Vp is maximal in the sense of [EPEl, Chapter I). There
is another proof in [Kol, Section 7 reducing the result to a theorem of S. Lang
on the vanishing of HI(IFq, H) for a connected algebraic group H over a finite
field IF q. When the group G is of type D, G may not be connected; so, the
proof of (2) for such groups may also fail.
Since the proof in the case of groups of type C is easy, let us give more
details of Case C. In this case, Lp = 0; for an even integer r. Choose hE C so
that h'Lp = L~. Then the (x, y) f-+ (h'x, h'y)o induces a new skew Hermitian
form (-'·h on Lp; so, we need to find h E O~ such that (hx, hyh = (x, y)o.
Take Xl, YI E Lp with (Xl, YI)O a unit u in Op. Changing YI by U-IYI, we may
assume that (Xl, YI)O = 1. Then Lp = (OpXI + 0pYI) EEl (OpXI + OpYI).l. By
induction on rankLp, we find that (-, ·)0 is equivalent to (x, y) f-+ xJr / 2 t y on
0;. Similarly, (-'·h is equivalent to (x, y) f-+ xJr / 2 t y on 0;, and we find the
desired h E Oc with (hx, hYh = (x, y)o.
Remark 7.4 If G is of type A, Kottwitz proved that Shy/:) @ E is a union

of the Shimura varieties Sh( G n X) j Kp for different unitary groups Gi with
Gi(A(oo)) ~ G(A(oo)) indexed by i E Ker(HI(<Ql, G) --+ TIe HI (<Qle, G)). Thus
if Hasse's principle fails for HI (<Ql, G), the functor EJ!) is bigger than E]1) over
E-algebras (see [Kol, Section 8), although Sh(G, X) and Sh(GU, X) share the
same neutral component. Thus to have the identity sMp) @v E = ShjG(Zp),
we need to have the cohomological triviality condition ((CT) in the following
subsection) for the center Z of the group G.
We have proved that Sh(p)(GU,X)@vE = Sh(GU,X)jGU(Zp) for GU of

types A and C, because we recover the polarization condition (pol) from the
A~ - () -
data defining EK . For each (Ao, AO, T/l ) over IFp, by deformation theory of
abelian schemes due to Grothendieck-Messing ([LaR] 6.2, [Zil, and [CBT]; see
also Theorem 8.8 and the remark after that), we can lift the triple (Ao, :Xo, T/~p))
to a triple (A,:X, T/(p)) over W = W(iFp) (the ring of Witt vectors with coeffi-
cients in iFp ). Indeed, there is a canonical lift if Ao is ordinary corresponding
°
to the Serre-Tate coordinate q = under (unr) (see Sections 8.2.4 and 8.2.5).
Then for the quotient field lC of W, we find that (A,:X, T/(p)) E E;J&(71 p ) (lC),
and the universal deformation space of (Ao, :Xo, T/~p)) is a formal spectrum of
7.2 General Shimura Varieties 321
a power series ring over W, and hence Sh(p)(GU,X) is smooth over V. We

have proven
Theorem 7.5 Suppose (unr) and one of the conditions (B1-2). Then the p-
integral model Sh(p) (GU, X) representing £t(P) is smooth over OE,(p), and
we have Sh(p) (GU, X) Q90E,(P) E = Sh(GU, X)IGU(7Lp)/E' where GU(7Lp) =
{x E GU(Qp)lxLp = Lp},
7.2 General Shimura Varieties

We describe canonical models of Shimura varieties of PEL and non-PEL types
as well as the global and local reciprocity laws characterizing the variety in a
succinct way for our later use.
7.2.1 Axioms Defining Shimura Varieties
We first formulate Shimura varieties as in Deligne [D1] and [D2J, and in many
places, our exposition follows Milne's expository article [Mil, Chapter II. We
consider a reductive group G /iIJ. with G(IR)+ ICo isomorphic to a Hermitian
symmetric domain X+, We write the adjoint group of G as Gad = G I Z in
this section (which was also written as PG before). We can identify X+ with
a connected component of the set X of the G(IR)-conjugacy class of a fixed
homomorphism (of real algebraic groups) ho : § /IR y G /IR whose stabilizer
(under conjugation) is Co. Each maximal compact subgroup of G1(IR) can be
given as the centralizer of such a homomorphism. In other words, we have
the origin 0 E X+ and the corresponding homomorphism ho : § y G with
centralizer Co in G(IR). We write x E X when we consider x geometrically
as a point of the symmetric domain of G and write hx : § Y G /IR when we
consider x E X as a homomorphism.
We start with such a pair (G, X) and require (G, X) to satisfy the following
four axioms.
(D1) For each x E X the Hodge structure on Lie(G) defined by the ho-
momorphism hx is of type {(-1,1),(O,O),(1,-1)}. In other words,
Ad(hx(z)) (z E (Cx = §(IR)) has only eigenvalues z-lz, 1, and zZ-l.
(D2) For each x E X, 9 f-t hx(i)ghx(i)-l is a Cartan involution of the adjoint
group Gj~ = PG /IR; so, the centralizer of hx in G 1(IR) is a maximal
compact subgroup. In other words, the involution gives the twisted
complex conjugation (which we referred to in the introduction) making
the centralizer of hx the subgroup of real points of a real algebraic
group.
(D3) The adjoint group Gj& has no factor defined over Q whose real points
form a compact group (this condition excludes a case we treated in
Section 4.3: G = G D = ResF/ilJ.Dx for a definite quaternion algebra D
over a totally real field F).
(D4) The identity component ZO = ZO(G) of the center Z = Z(G) of G

splits over a CM field.
In addition to these conditions, we assume the following two extra conditions
to avoid technicality.
(SC) The derived group G 1 of G is simply connected (this condition excludes
some orthogonal groups of type B and groups of type D but includes
their Spin and Clifford covering groups).
(CT) The center Z of G is a cohomologically trivial torus.
Here (CT) means the vanishing of the Galois cohomology H1(F, Z) =
Hlt(Gal(Ql/F),Z(Q)) for each finite extension F/Ql (see [MFG] Chapter 4
for Galois cohomology). Thus the group GU satisfies the condition (CT) as
we already remarked but the group G in (7.1) may not.
Example 7.6 We quote the following three types of the pairs (G, X) satis-
fying the above six conditions (Dl-4), (SC) and (CT) from [MiS].
A. (Unitary groups: type A groups) Let (D/F' *) be a central simple algebra
over a totally imaginary quadratic extension F over a totally real field Fo
(a CM field over Fo) having a positive involution * inducing nontrivial
automorphism c on F / Fo. We write 10 for the set of real embedding of
the totally real field Fo. The group G IrQ is given by
G (k) = {g E (D ®rQ k) x Ixx' E (Fo ®rQ k) x }
for Ql-algebras k. Then G(JR:) ~ fL'Elo GU(m", n,,), where
with Im,n = (10 -t ). Then X is the G(JR:) conjugacy class of ho : § =

rex y G(JR:) given by Z H (( Z10u Z1~u)) ,,' These are included in the
groups we have defined in (7.1) (this is the case where V = D and hence
C = t D under the notation of (7.1)).
B. (Orthogonal groups: type B groups) Let q be a quadratic form on a vector
space V over a totally real field F such that it has signature
(m", n,,) E {(d, 0), (0, d), (d - 2,2), (2, d - 2)}
on V" = V ®F,,, JR: (d = dimF V). We suppose that dimF V 2' 3. We write
Is for the places of F where q has signature (d - 2,2) or (2, d - 2). We
put Ir = I - Is. Then G = ResF/rQGpin(q) for the special Clifford group
Gpin(q)/F for q. By definition,
G 1 (JR:) ~ II Spin(d-2,2)(JR:) x II Spin(d,O)(JR:).

We follow lSi] for the description of the origin ho : § '---t G(JR) of X.

Decompose ho = flo. ha and ha = 1 if 0' E I r . Write
v: G = ResF/QGpin(q) -7> ResF/Q(Gpin(q)/Spin(q)) = ResF/QG m / F

for the lift of the similitude map by Gpin(q) -7> GO(q), and define the
involution LOG by ggL = v(g). For 0' E Is, take two orthogonal vectors
el,e2 E V 0F,a JR with q(eJ ) = -1 for j = 1,2. Then in the Clifford
algebra, ja = ele2 satisfies J;= -1 and j~ = -ja' Then ha(a + bA) =
a + bja E Gpin(qa), where qa is the quadratic form induced on V 0F,a JR
from q.
C. (Symplectic groups: type C groups) Let D / F be a quaternion algebra (we
allow D = M 2 (F)) for a totally real field F. Let V be a free left D-
module ofrank d. Write Lfor the involution on D given by x+x L = Tr(x)
and xx L = N(x) for the reduced norm and trace N, Tr : D --+ F. Let
ID = {O' E liD 0F,a JR ~ M2(JR)} and ID = 1- I D. We suppose that
ID =I- 0. We consider a nondegenerate symmetric form ( , ) : V x V --+ F
such that (bx,y) = (x,bLy). Then, for Q-algebras k,
G(k) = {g E AutD<8IQk(V 0Q k)l(gx, gy) = v(g)(x, y)},
where v(g) E (F 0Q kY for g E G(k); that is, v : G --+ ResF/QGm is a

morphism of group schemes. We assume that ( , ) induces a positive- or
negative-definite symmetric form on V 0F,a JR if 0' E ID. Then G(JR) ~
I1aEI D GSp2d(JR) x I1aEID GUfrIT(d, 0), and the space X is the G(JR) con-
jugacyclassofho:Cx 3z=a+bAH ((::tf d
~id))
d aEID
x (l)aEID.
Here
GUfrIT(d, O) = {x E GLd(lHI)lxxL = v(x) E JR X }
for the algebra lHI of Hamilton quaternions. If there is no component iso-

morphic to GUfrIT in G(JR), this type of group is included in the groups we
have defined in (7.1).
We fix an origin 0 E X+, which we assume to be the one given in the above
examples when we are in Cases A, B, and C. We first consider the Shimura
variety over C in Deligne's sense,
Mc(G,X)(C) = ~KG(Q)\G(A)/KCo, (7.7)
where Co is the centralizer of ho in G(JR) and K runs through open-compact

subgroups of G(ACool). This projective limit can be identified set-theoretically
with
(7.8)
where the action is given by ,,/(x,g)z = bx,,,/xz) bE G(Q) and z E Z(Q))

and Z(Q) is the closure of Z(Q) in G(ACool) (see [D2] Proposition 2.1.10 and
[Mi] pages 323-324). We write [x, g] for the complex point of the Shimura
variety represented by the pair x E X and g E G(A.(oo)). When G is given by
GU of type A or C in (7.1), GU(Q)/GU(QJ+ acts on 7ro(X) transitively; so,
we can rewrite (7.8) as
We recall the amalgamated product G(A.(oo)) *G(IQI)+ Gad(Q)+ defined in

(4.30). We let Gad(Q) act on G(A.(oo)) by conjugation, and we make a semi-
direct product G(A.(oo)) ~ Gad(Q)+, where Gad(Q)+ = Gad(Q) n Gad(JR)+
for the identity-connected component Gad(JR)+ of Gad(JR). We write G(Q)+
for the subgroup of G(Q) whose image in Gad(Q) falls in the identity-
connected component of Gad(JR). We embed G(Q) into the semi-direct product
G(A.(oo)) ~ Gad(Q) by "( H b- 1 ~ adb)). Then the image of G(Q) is a nor-
mal subgroup of G(A.(oo)) ~ Gad(Q)+. We then define G(A.(oo)) * Gad(Q)+ by
G(A.(oo)) ~ Gad(Q)+ /G(Q). This group acts on Mc(G, X) on the right by
[x,g](g' ~ "() = b- 1 (x), (gg')'Y],
where gt = C 1 gt. The action is well defined because we have
b- 1 a(x), (agg')'Y] = [a,,,(-I(x), a' (gg')'Y] = b- 1 (x), (gg')'Y] = [x, g](g' ~ "()
and [x,g]b- 1 ~ adb)) = b- 1 (x),(n- 1 ),] = b- 1 (x),"(-lg]
for a, "( E G(Q).
7.2.2 Reciprocity Law at Special Points
Since §/c ~ G m x G m by Z H (z, z), for any morphism of algebraic groups

p: G -t GL(V) defined over Q for a Q-vector space V, the composition pohx
(x E X) gives a double grading on V ®IQI C (a Hodge structure). Deligne's
viewpoint is to regard the Shimura variety over C to be a modulus of such
Hodge structures with additional restrictions ([D2] Section 1). More boldly,
one might expect that a given Shimura variety Sh(G, X) is a fine modulus of
a specific family of motives, since each motive has a natural Hodge structure
of its Hodge realization (see [D2] Introduction, [Mi] II.3, and [Mi2]). In the
case of a Shimura variety of PEL type, by construction, it is a fine modulus
of specific abelian schemes up to isogeny; so, it is a modulus of the motives
Hl(A) associated with abelian varieties A. Via this heuristic, we expect that
Mc(G, X) has a canonical model defined over a specific number field, the
reflex field we now introduce.
We can pull back hx to G m by putting /Lx(z) = hx(z, 1), getting a homo-
morphism /Lx : G m '-t G defined over C. Let Mx (C) be the conjugacy class
of the homomorphism /Lo : G m '-t G. Then Mx(C) is the set of complex
points of an algebraic subvariety of G. Let E = E(G, X) be the smallest field
of definition of Mx, that is, the fixed field of
{O" E Aut(C) 100(Mx (C)) = Mx(C)}.

Thus E is a finite extension of Q, which is known to be a totally real field or
a CM field. This follows from the classification of the pairs (G, X) satisfying
(Dl-3) done in [D2] 1.3.8, showing that E(G, X) is a subfield of the fixed field
K of the subgroup of Gal(Q/Q) fixing the Dynkin diagram D associated with
the pair (G, X) ([D2] 2.3.5). In Aut(D), complex conjugation is in its center,
and hence K is either a totally real field or a CM field ([IAT] 5.11). The field
E is called the reflex field of (G, X) (which used to be called the dual field).
A point x E X is called a special point if there exists a maximal torus
Tx c G defined over Q such that hx has values in Tx(IR). A point [x, g] E
Me(G, X) is called special if x E X is special. By definition, we then have
/-Lx : G m --+ Tx· Let Ex be the minimal field of definition of /-Lx, and write E~b
for the maximal abelian extension of Ex. The field Ex is a CM field containing
E given by the fixed field of {O" E Aut(C)10" 0 /-Lx = /-Lx 0 O"}. This induces the
following homomorphism of algebraic groups,
We then define the reciprocity map rx : Gal(E~b / Ex) --+ Tx(A(oo))/Tx(Q)

as follows. For each 7 E Gal(E~b / Ex), take an idele t E (Ex Q91Q1 AV giving
rise to 7- 1 by the Artin reciprocity map. Then define rx(7) = (Nx(t(oo))
mod Tx (Q) ). It is easy to check that the image only depends on 7.
A scheme Sh(G, X) over E = E(G, X) is called a canonical model or the
Shimura variety of Me( G, X) if the following two conditions are met:
(rel) Sh(G, X) Q9E C ~ Me(G, X);
(rc2) If [x,g] E Me(G, X) is special, then [x,g] E Sh(G,X)(E~b) under the
above isomorphism and 7([X,g]) = [x,rx(7)g] = [x,g· rx(7)9] for all
7 E Gal(E~b/Ex), where rx(7)9 = g-lrx(7)g.
The condition. (rc2) for G = ReSFjIQlGL(2) (for a totally real field F = Fa)
is the local reciprocity law in Section 4.2.3, but in this general case, it is a
requirement for Sh(G, X)jE to be the Shimura variety of (G, X). When (G, X)
comes from the algebra DjF and D-module V as in Section 7.1, special points
are those points on Sh(G, X) carrying abelian varieties of CM-type, and (rc2)
follows from the main theorem of complex multiplication (Theorem 4.19) as
we have proven (rc2) (i.e., Corollary 4.20) for the Hilbert modular variety.
By (7.8), we find [x, ')'g] = [,,(-lX, g] hE G(Q)). For,), E Tx(Q), ')'(x) = x
if x is special with respect to Tx. Thus [x,')'rx(7)a] = ["(-1(x),rx(7)a] =
[x,rx(7)a] is well defined without ambiguity. Since the set of special points is
Zariski dense (actually, it is dense also in complex topology; see [D1] 5.1). In
all three cases (A, B, C), the existence of the canonical models is known (see
[Sh5], lSi], [D1] 4.21,6.4, and [D2] 2.7.20). More generally, if the pair (G, X)
has a symplectic embedding into (GSp(2N), nN U SJN) as in [D2] 2.3.2-3,
the complex proanalytic space Me(G, X) admits a unique canonical model
Sh(G, X)/E(G,X) (see [D2] 2.7.10). This type of Shimura variety is called a
Shimura variety of abelian type. See [Mi] II for a discussion covering a more
general type of Shimura variety.
7.2.3 Shimura's Reciprocity Law
In this subsection, we reformulate Shimura's tower of geometrically connected

canonical models in Deligne's language. Each member satisfies an intricate
relation under the action of the group G(A(oo)) as described in [Sh5] II.5 and
[MiS] Theorem 4.6, but we only describe the tower; so, we do not discuss this
point here.
We now define, following [D2], p : if -+ HI for the universal covering if
of the derived group HI of each reductive group H/Q. Then we put
7r(H) = H(A)/ H(Q)p(if(A))
and write 7r07r(H) for the group of connected components of 7r(H). We put
7f07r(H)= 7r07r(H)/7ro(H(lR)+), where H(lR)+ is the inverse image of the iden-
tity component Had(lR)+ of Had(lR). We consider the following exact sequence
(7.9)
for a torus ZG (called the cocenter of G) defined over Q. Recall /Lx = hx IlG m xl:
(Gm/IC -+ G /IC. For each x EX, we look at the composite /Lx = X 0 /Lx : (Gm -+
ZG. Since {/Lx}xEX belongs to one conjugacy class X, /Lx is independent of
x and is defined over E = E(G, X) by the definition of the reflex field E. We
G Res(!J.x)
can thus define A: ResE/Q(Gm -+ Z by the composite of ResE/Q(Gm )
ResE/QZG and ResE/QZG ~ ZG. We define a homomorphism
R : 7r07r(ResE/Q(Gm) = 7ro(Er / EX) ~ Gal(Eab / E) -+ 7f07r(G)
following [MiS], page 929. Here we identify 7ro(Er / EX) with Gal(Eab / E)
sending a prime element w E F x c Fpx at a prime ideal p to the geo-
metric Frobenius at p (thus compatible with the Artin reciprocity law; see
Theorem 2.12). This homomorphism was defined by Deligne in [D2] (2.6.2.1)
as 7roNE /QqM (using an intricate but explicit argument in terms of tor-
sors without assuming (SC), [D2] 2.4-6). By (SC), the strong approxima-
tion theorem affirms (see [D1] Theorem 2.4) that 7r07r(G) ~ 7r07r(ZG) =
7ro(ZG(A)/ZG(Q)) via X. The.homomorphism A gives rise to a homomor-
phism of 7r07r(ResE/Q(Gm) = 7ro(Er / EX) into 7r07r(ZG) = 7ro(ZG (A)/ZG (Q)).
Thus we have
Composing A with the projection: 7r07r(ZG) = 7ro7r(G) -» 7f07r(G), we obtain

£ : 7r07r(ResE/QGm) = 7ro(Er j EX) ~ 7r07r(ZG) = 7r0 (~~i~D -+> 7f07r(G).
Let t be the subfield of E ab fixed by Ker(£). We then define a topological

subgroup Q+ = Q(G, X) c G(A)+ by the commutativity of the following
diagram with two exact rows,
1 ----i- G(Q)+G(IR)+ ----* G(A)+ ----* 7f07r(G) ----i- 1
"I I Uli (7.10)

1 ----i- G(Q)+G(IR)+ ----* Q( G, X) ----* Gal(tj E) ----i- 1.
a
Here G(Q)+ is the closure in G(A) of G(Q)+ = G(Q) n (G(A(exo))G(IR)+),

and £ restricted to Gal(QcyC E j E) coincides with (the inverse of) the Artin
reciprocity map (cf. Theorem 2.12). In Shimura's paper [Sh5], II, the notation
Q+ is used for the group Q(G, X), which is the notation used in [MiS], 4.13.
As is shown in [MiS], 4.12-13, Q(G, X) gives the stabilizer in G(A)+ of the
geometrically irreducible component of Sh( G, X) containing the image of the
identity of G(A).
Let Z be the set of subgroups K of Q(G, X) satisfying the following two
conditions,
1. Z(Q)G(IR+) c K;
2. Kj Z(Q)G(IR)+ is open in Q(G, X).
We recall that we fixed an origin 0 E X+ eX. We then identify X with
G(IR)jCo for the centralizer Co of ho. The origin is taken to be ho as exhibited
in (A-C) in Example 7.6. We take the geometrically connected component
V/£ of Sh(G, X) containing the image of the identity of G(A) and define
VK / kK = VjK as a geometrically irreducible variety defined over kK, where
kK is the fixed field of cr(K) in t Then this family of geometrically irreducible
varieties {VK } KEZ satisfies many fine properties as described in [Sh5] II.5 (see
also [MiS] Theorem 4.6), which characterize the tower {VK } KEZ of canonical
models.
Define.ft= UKEZkK(VK) for the function field kK(VK) of Sh K over k K .
Then we have the following global and local reciprocity law (we have proved
this at least for Hilbert modular varieties in Theorem 4.14 and Siegel modular
varieties in Theorem 6.26).
Theorem 7.7 (Shimura, Deligne, Milne-Shih) There exists a canonical
group homomorphism T : Q(G, X) ---+ Aut(.ftj E) characterized by the following
five properties.
(1) The kernel of T is Z(Q)G(IR)+;
(2) For a E Q(G, X), T(a)lt = £T(a);
(3) For any a E G(Q)+ c Q(G, X) and z E Sh(G, X), T(a)(f(z))
f(a- 1 (z)) for all f E .ft;
(4) For each special point z E X+ and f E it, f(z) E E~b if f is defined at
z, where X+ is the connected component of X containing o. Moreover,
if f is defined at special z E X+, then T(rz(p))(f) is defined at z and
p(f(z)) = T(rz(p))(f(z));
(5) The map T defines a topological isomorphism of Q(G,X)/Z(Q)G(ffi.)+
onto an open subgroup of Aut (it/ E). For every K E Z, we have T(K) =
Aut (it/kK (Sh K )). Here Aut(it/ E) is equipped with the Krull topology as
described in Section 2.3.
This theorem has been proven by Shimura in [Sh5], II in Case C. After some
work by students of Shimura, the general case in the theorem is covered by
Milne-Shih [MiS], Theorem 4.13. Our formulation here follows [MiS] and is dif-
ferent from Shimura's formulation in which he uses T(X- 1 ) for T in the present
notation to make Galois action as a right action. We followed Shimura's for-
mulation in Theorem 2.43, and this point is obvious from T(a) (f) = f 0 a in
Theorem 2.43 which is the inverse of T(a) in (3) above. Note here that the
group T(9( G, X)) is not just an open subgroup of the field automorphisms of
it but also the subgroup of scheme automorphisms of Sh( G, X) over E.
The cokernel of T also can be described explicitly as a subgroup of
Aut(F/Q) in Cases A, B, and C, and in these cases, one can extend the
group Q( G, X) to describe explicitly the full automorphism group Aut (it/ E)
(see [Sh5] II, [MiS] 4.7 and 4.14, and [H03b]). In the proof of [MiS], one of
the key ingredients of the description of the full group is the determination
of Aut(Cit/CC) by Miyake [Mt], and in [H03b], the theorem of Faltings and
Zarhin (Theorem 4.13) is used in place of Miyake's result.
8
Ordinary p-Adic Automorphic Forms
We first describe a formal theory of false automorphic forms and find a set
of conditions for nearly ordinary p-adic cusp forms to have vertical control.
After this, we describe deformation theory by Serre-Tate of abelian varieties
in order to prove the divisibility by the character value fJ(O in (5.40) of the
Hecke operator associated with an element ~ of the expanding semi-group
D in Section 5.1.2. By this, we have well-defined p-integral Hecke operators
Tp(O on coherent cohomology groups and therefore well-defined nearly p-
ordinary automorphic forms. Verifying the set of conditions on sections of
automorphic vector bundles of a given (projective) Shimura variety of PEL
type, we prove the vertical control theorem for automorphic forms on the
Shimura varieties. We also state a similar result for quasi-projective Shimura
varieties of quasi-split unitary groups without a detailed proof. At the end, we
prove the irreducibility of the Igusa tower over the ordinary locus of symplectic
and unitary Shimura varieties, reducing the problem to the Siegel modular
case. By means of the Serre-Tate deformation coordinates, even when the
Shimura variety does not have a cusp, we are able to define an analogue of
the p-adic valuation v (in (4.36)) of its arithmetic automorphic function field;
so, the irreducibility can be proven even for projective symplectic and unitary
Shimura varieties of PEL type.
8.1 True and False Automorphic Forms

We describe the theory of false automorphic forms for a formal analogue of
the Igusa tower. In our application, we remove nonordinary locus from the
modulus Mjwm of abelian schemes of PEL type and write SmjWm for the
ordinary locus. The scheme for 0 < m < 00 is obtained by inverting the Hasse
invariant. For m = 00, we define a formal scheme SoojW by Soo = ~mSm. If
we choose a lift E E HO (Mjw, f!d.a) of the Hasse invariant and define S jW =
M[1;ljw, Soo is the formal completion of S along its special fiber modulo p. In
this setting, sections in HO(Soo, f!d.K) are called false automorphic forms. On the

330 8 Ordinary p-Adic Automorphic Forms
other hand, sections in HO(M, 1,Jd."') are called true or "classical" automorphic
forms. The theory is a formal generalization of the work of Deligne-Katz in
the elliptic modular case (see [Kl] Appendix III).
The main purpose of this section is threefold:
1. Approximate p-adic automorphic forms by finite sums of classical forms;
2. Define the p-ordinary projectors e and eo;
3. Find a set of (axiomatic) conditions that guarantees the VCT.
8.1.1 An Axiomatic Igusa Tower
Recall the mixed characteristic complete discrete valuation ring W with

residue characteristic p. Let w be a uniformizing parameter of W. Set Wm =
W /pmw. Let S be a flat W-scheme. We put Sm = S Xw W m . Then Sm is a se-
quence of flat Wm-schemes, given with isomorphisms: Sm+10W""+1 Wm ~ Sm.
We write Soo for the formal completion ~ mSm of S. Let P be a rank 9 p-
adic etale sheaf on the Sms; thus, p/ S "" + 1 induces p/s"", P = ~nP/pnp,
and Pn = P /pn P is an etale twist of the constant sheaf (Z/pnz)g.
We can slightly generalize our setting and could suppose that there exists
a finite extension F/G. with integer ring 0 = OF and a homomorphism 0 '-+
Ends",,(P) such that Pn ~ (O/pnO)g for all n locally under etale topology.
Since we can transfer any of our results to this slightly general situation,
just replacing Zp by Op = 0 0z Zp, we state our result only for P with
Pn ~ (Z/pnz)g. This simplification also allows us to save some symbols.
We write 7fm ,n : Tm,n = Isoms"" (Pn , (Z/pnz)g) -t Sm for the finite etale
Sm-scheme representing the following functor, SCH IS"" -t SETS,
The representability follows from the theory of Hilbert schemes as we have

seen (see Section 6.1.5). Indeed, by 7/J f-t 7/J-l, we have
Since 7/J : (Z/pnz)g ~ P n is determined by (7/J( el), ... ,7/J( en)) for the standard
base (el, ... ,en ) of (Z/pnz)g, we find that Isoms",,((Z/pnZ)g,Pn ) is repre-
sented by the open subs cherne of Pi( given by
Because Pi(/Sm is etale, Tm,n/s"" is etale. Since each geometric fiber of Tm,n
over Sm is isomorphic to GLg(Z/prZ) everywhere, it is faithfully flat and
finite. Therefore Tm,n is affine over Sm. We define Vm,n = HO(Tm,n, OT"",n)'
The group GLg(Z/pnZ) acts on Tm,n by 7/J f-t x7/J for x E GLg(Z/pnZ)
without fixed points, and we have T m,n/ rn,n' ~ T m,n' for all n' :S n, where
8.1 True and False Automorphic Forms 331
Then we have a tower Vm,o C Vm,l C ... C Vm,n with Vm,o = HO(Sm, Os,,J.
We put Vm,oo = Un Vm,n and Tm,oo = ~nTm,n. Similar to (4.13), we put
(8.2)
On these modules, GLg(Zp) acts naturally, and for a closed subgroup H C

GLg(Zp), we have for the submodules fixed by H:
(8.3)
Indeed, T 00,001 H = ~mTm,ml Hm for the finite image Hm C GLg(Zlpmz) of

H, and Tm,ml Hm is the geometric quotient of Tm,m by a finite constant group
Hm (cf. [GME] Proposition l.8.4 (b) or [SGA] V.1), since Too,oo = ~mTm,oo
is a formal scheme which is an etale covering of Soo with Too,oo/GLg(Zp) =
Soo. This shows the second identity. The first identity is plain.
8.1.2 Rational Representation and Vector Bundles
For a given ring A or a sheaf of rings A over a scheme, we look at the flag
variety F/A = FlagB/A of all maximal flags in Ag defined in Section 6.l.3. We
write BeG L(g) for the upper triangular Borel subgroup. Let N be its unipo-
tent radical, and put T = BIN for the torus. Note that T ~ BIN ~ t Bit N
for the Borel subgroup t B opposite to B and its unipotent radical tN. Then
F ~ GL(g)IB. We define H/A = FlagN/A = GL(g)IN. Write 1 for the origin
of H represented by the coset N. Each schematic representation of T is com-
pletely reducible (see the following description of schematic representations
and [RAG] I.2 and [GME] 1.6.5), and we have HO(H, OH.) = ffil<EX(T) RA[Ii:]
for the space R A [Ii:] of weight Ii:. Here
RA[Ii:] = {J: GL(g)IN 4 Al E T(GL(g)IN,OCL(g)/N)if(ht) = Ii:(t)f(h)}
for t E T and Al = Spec(A[T]) (the one-dimensional affine space).

The pullback action of GL(g) on RA[Ii:], f(x) >--+ p(h)f(x) = f(h-Ix), gives
a representation p = PI< such that RA[Ii:]N ~ A on which T acts by ~Woli:. Here
Wo is the longest element of the Weyl group of T represented by anti diagonal
matrix Wo = (6~,g+1-J)lJ modulo T. The dual RA[Ii:]* = HomA(RA[Ii:],A)
is the universal representation of highest weight Ii: (cf. [RAG] II.2.13). The
weight Ii: is positive (or dominant) with respect to B if RA [Ii:] f 0, and the set
of all positive weights Ii: E X(T) forms a cone X+(T) of positive weights.
To describe the notion of the universal representation, we recall schematic
representations (cf. [GME] 1.6.5). We regard an A-projective module M as a
covariant functor from A-ALG into the category of abelian groups by M(AI) =
M Q9A AI. For a group scheme G IA and an A-projective module of finite type
M, M is called a schematic representation if we have a morphism of covariant

functors G x M ---+ M such that the induced morphism G(A') x M(A') ---+
M(A') is an action of the group G(A') on the A'-module M(A') = M@AA'.
We take G = GL(g)/A and the schematic representation A(I1:) of B/A
given by a character 11: E X (T) regarded as a character of B by composition:
B - # T ~ G m . Then RA[I1:]* satisfies the following universal property for
schematic representations M of G L(g) / A,
HOmCL(g) (RA[I1:]* ,M) ~ HomB (A(I1:), M) canonically.
A proof and a construction of the above isomorphism is given in [RAG], II.2.13,

whose outline is as follows. For a schematic GL(g)/A-module M,
HOmCL(g) (RA[I1:]*, M) ~ HOmCL(g) (M*, RA[I1:])

~ HomB(M*, A(I1:- 1 )) ~ HomB(A(I1:), M).
The middle equality is given by the Frobenius reciprocity (e.g., [MFG] 2.1.6).
The GL(g)-module RA[I1:]* is generated by a B-stable line of weight 11:, and
any schematic GL(g)-module generated by a B-stable line of weight 11: is a
surjective homomorphic image of RA[I1:]*.
By this universality, taking M = A(O), we know that the coinvariant space
RA[I1:]N := Ho(N, RA[I1:]) is a free A-module of rank 1 on which T acts by -11:,
and there is a unique N-invariant linear form ecan : RA[I1:] ---+ A (up to an
A-unit multiple), which generates (RA[I1:]*)N. We can normalize ecan so that
it is the evaluation of ¢ E RA[I1:] at the origin 1 E GL(g)jN. Then we have a
tautological embedding RA[I1:] '---+ r(GL(g)jN, OCL(g)/N) given by
If h- 1 E M gXg / A for the g x g matrix algebra M gXg / A (regarded as a ring

scheme over A) as a multiplicative semi-group scheme, the action of p(h)
(h E Mgxg(A')) is well-defined on RA'[I1:] for any A-algebra A'.
In [RAG], a slightly different module is considered:
HO(I1:) = Ind?;(g) A(-I1:) = {J: GL(g);tN ---+ A1If(xt) = r"'f(x) 'Vt E T}.
The action of h E GL(g) is given by hf(x) = f(h-1x). In this context, -11:

is a positive weight with respect to tB, and the HO(I1:)N contains the highest
weight vector. Using conjugation by Wo, we can remove the use of the lower
triangular Borel subgroup t B, but we need to modify the results of [RAG]
accordingly when we quote them (this is done without further warning).
Suppose that A is a p-adic ring. Let f E R A/ p>nA[4 By definition, f
induces a function on GLg(Zp) by f(h) = p",(h-1)f(1), and h r--+ ecan 0 f(h)
is an element in R A/ p>nA[I1:] by tautology. This shows the following fact.
which is injective if A is flat over Zp. Here C(GLg(Zp)/N(Zp),A) is the space

of (p-adic) continuous functions with values in A on GLg(Zp)/N(Zp), and [K:]
indicates the K:-eigenspace under the right multiplication 9 H gt by t E T(Zp)
for 9 E GLg(Zp)/N(Zp). The cokernel of the map (c) is large, because
1. the image is made up of polynomial functions of the coordinates of G L(g);
2. C(GLg(Zp)/N(Zp), A) contains all continuous functions of the quotient
p-adic space GLg(Zp)/N(Zp) into A.
We let hE GLg(Zp) act on C(GLg(Zp)/N(Zp), A) by h¢(x) = ¢(h-1x), and
this representation of G Lg (Zp) is called the continuous induction of the trivial
representation of N(Zp) to GLg(Zp) with coefficients in the p-adic ring A.
When A is a finite ring, the space C(GLg(Zp)/N(Zp), A) is equal to the space
of locally constant functions .cC( GLg(Zp)/ N(Zp), A), and we use the symbol
.cC instead of C when A is finite.
8.1.3 Weight of Automorphic Forms and Representations
Let ~m be the locally free sheaf P Q9zp Os", = Pm Q9zp Os", on Sm of rank g.
We define a coherent sheaf ~m,n on Tm,n by (7r:n,nPm) Q9z OT""n (~m = ~m,O
for Tm,o = Sm). On Tm,n with n ::::: m, we have the universal isomorphism
Wean = lean Q9 id : ~m J
n ~ O~m,n
is an isomorphism. Recall HjA := FlagNjA = GL(g)/NjA . Then we can

identify H/T""n with PY : Y = Ym,n = GL(~m,n)/Ncan -+ Tm,n on Vm,n
(n ::::: m), where N ean is the pullback of N under Wean. Thus Wean induces
an isomorphism w~an : PY,*(Oy[K:]) ~ R v""J4 We write ~~,n for the sheaf
PY,*(Oy[K:]) on Tm,n- By definition, GLg(Z/pnZ) acts on Y on the left. The
group GLg(Z/pnZ) acts on ~~,n via the rational structure given by ~m' and
we then descend the sheaf to ~~ on Sm. In other words, for an OT""n -algebra
A, f E HO(SpecT""n (A),~~,n) is a functorial rule assigning f(X, 'Ij;) E RA[K:]
to X/A and 'Ij; : Pn/ X ~ (Z/pnZ)jx' We let h E GLg(Z/pnZ) act on f by
f H {(X, 'Ij;) H p(h)-l f(X, h'lj;)}. Then for any OT""o-algebra A,
A H HO(GLg(Z/pnz), HO(Spec(A) xSm Tm,n,~~,n))
defines a coherent sheaf on Sm (by the Hochschild-Serre spectral sequence:

[EeH] III.2.20), which we write ~~. We have
HO(Sm'~':r') =
{J E HO(Tm,m, R v""", [K:])lf(X, h'lj;) = p(h)f(X, 'Ij;) for Vh E GLg(Zp)} .
There is another description of ~/Sm' Since Pm ~ (Zjpmz)g on Tm,m, the

action of GLg(Zjpnz) on Pm extends to an action of the Galois group on
RZ/pmZ[K:], which determines an etale torsion sheaf P:;' over Sm. Then we have
f±!.~ = OSm 0z P:;'. We have detk(f±!.m) = (/\g f±!.)0k and Symk(f±!.m) = f±!.'::;" for
the first standard dominant weight W1'
By our definition, for each f E HO(Sm, f±!.~),
which can be regarded as a functorial rule assigning each test object
(X/ Sm ,7jJ: Pm/ X ~ (ZjpmZ)/x)
a value 'P(f)(X, 7jJ) E HO(X, Rox [K:]) such that 'P(f)(X, h'lj;) = p(h)'P(f) (X, 7jJ)
for all h E GLg(Zjpmz) and 'P(f)(Y, ¢*7jJ) = ¢*('P(f)(X,7jJ)) for a mor-
phism Y ~ X of Sm-schemes. Similarly, 'P E Vm,n is a functorial rule as-
signing (X, 7jJ) a value 'P(X,7jJn : Pn ~ (Zjpnz)g) E HO(X, Ox) such that
'P(Y,¢*7jJn) = ¢*('P(X,7jJn)) for a morphism Y ~ X of Sm-schemes. Thus we
have a natural map of HO(Tm,m, RTm,m [K:]) into Vm,m associated with each
linear form £ E RVm,m [K:]*. The map associates f E HO(Tm,m, RTm,m [K:]) with
a rule: (X,7jJ) H £('P(f) (X, 7jJ)), which is a matrix coefficient of 'P(f)(X, 7jJ).
We let h E GLg(Zp) act on test objects by (X,7jJ) H (X, h7jJ). In this
way, we identify GLg(Zp) with a subgroup of Aut(Tm,oo/s,J. For the Borel
subgroup B c GL(g), we put T!,n for the quotient Tm,nj B(Zjpnz). Thus
V! n = HO(T! n' OTB ) is made of a functorial rule (X,7jJn) H 'P(X, 'lj;n) E
HO'(X, Ox) such that~(X,b7jJ) = 'P(X,'lj;) for all7jJ and bE B(Zp). We define
V:'n similarly and T;;;,n for the unipotent subgroup NcB. Thus we have
and for these modules, we have the identity (8.3).

J
,.-"--..
Let e1, ... ,eg be the standard base eJ = t (0, ... ,0,1,0, ... ,0) of (Zjpnz)g,
and we consider the standard filtration
given by L J = 2:;=1 (ZjpnZ)e,. Then 7jJ~ln gives a (full) filtration fil = fil"'n
of Pn , and all full filtrations Pn = P~g) :) p~g-l) :) ... :) pta) = {O} of
Pn are given in this way. Since the stabilizer of In is B(Zjpnz), we may re-
gard 'P E V!,n as a functorial rule assigning a value 'P(X, filn) E HO(X, Ox)
to a test object (X, fil n ). To describe V:'n in this way, we need to bring
in an isomorphism of graded modules ¢n : gr(fil n ) ~ EB;=l (Zjpnz) induc-
ing ¢~) : p(J) j p(J-1) ~ (Zjpnz). In other words, T;;;,n classifies triples
(X, fUn, ¢n). Since we pulled back the filtration In by 'l/Jn, h E GLg(71p)
acts on (X, fil,pn' ¢n) by fil n H 'I/J-1h-1'I/Jfil n (h'I/Jn)*ln and ¢n H
'I/J-1h-1'I/J¢n'
We consider Rv""", [Ii] inside V:'m[li], which is the homomorphic image of
HO(8m,!!:!..':;,) under f H eean 0 <p(f). Thus we have a natural map
(8.4)
where V:'m[fi;] is the fi;-eigenspace of the right action of T. The above map is
injective if m = 00. Define
R'm = (8.5)
I<EX+ (T),I<»O
Here» implies "sufficiently regular" as defined in Section 5.1.3.
8.1.4 Density Theorems
We assume that the formal sheaf !!:!..= is algebraizable; so, we have a locally
free sheaf !!:!..;s of finite rank such that!!:!.. 0w Wm = !!:!..m for all m. The
sheaf!!:!.. is determined uniquely if it exists. From !!:!.., we can create !!:!..;s as
7r*(')FlagN(~)[fi;] for 7r: FlagN --t 8 (see Section 6.1.3). The global HO(8,!!:!..) in-
jects into ~mHO(8m'!!:!..m)' We define a graded module R' = EBI< HO(8,!!:!..I<)
which is embedded into R'oo = ~mR~. We call an element of HO(8,!!:!..;s)
a false automorphic form of weight fi;, and a true (or classical) automorphic
form is a section in HO (M,!!:!..;M) for a compactification M ::) 8 of 8 we specify
later.
We suppose now that for all fi; » 0, the short exact sequence
gives rise to an exact sequence:
(Hpl)
This condition is satisfied when 8 m is affine (see [ALG] 111.3.5). As we show
in Theorem 8.3, (Hpl) implies
(Hp2)
From (Hpl), we have
R'/pmR' ~ R'rr, and HO(8,!!:!..I<)/pmHO(8,!!:!..I<) ~ HO(8m,!!:!..':n). (8.6)
We now define a homomorphism (3(m) : R~ --t V;; m in the following way.

We have a canonical isomorphism Wean = lean 0 id : ~m , m ~ (')~Tn,rn . Then
J3(m)(L f,J
1<»0
= {(X/TTn'Tn''lj;) r-+ Llean(W~an(JI«X''lj;)))}
I<
for fl< E HO(Sm'~':n). Here, the image of J3(m) actually falls in V:'n because
lean 0 PI«u) = lean for all u E N(Zp), and w~an(Jl<) E Rk,Tn[K]. By construc-
tion, J3(n) mod pm = J3(m) for all n > m. Taking the projective limit, we
have 13(00) : R'oc, -t VN = \i!!!m
V:'oo'
Since Sm is flat over Wm = Wjpmw and T!/',n = Tm,njN(Zjpnz) is etale
over Sm, V:'oo is flat over W m . Therefore, VN is a W-flat GLg(Zp)-module.
Remark 8.1 Flatness of HO(M,~I<) is subtle. Let S = 9J1(c, r(N)) as [iJ
defined in (r(N)) of Section 4.1.2 for a lift E of the Hasse invariant. Assume
that p is prime to Nand unramified in F jr[}.. We take Pn to be the maximal
etale quotient of X[pn] for the universal AVRM f : X -t Sm. By Corollary 4.2,
we have that ~ ~ f*[2x/s. From this, the locally free sheaf ~ extends to the
locally free sheaf ~/M = f*[2.X/M on the smooth toroidal compactification M
canonically because the universal abelian scheme extends to a semi-abelian
scheme X over the toroidal compactification as described in Sections 4.1.4
and 6.3.6. As we have seen in Remark 4.8, HO(M/wTn'~7M) is not Wm-free
for 1 < m < 00 if k is trivial on T(lFp), but nontrivial on T(Zjpmz). However,
flatness of ~7M implies W-freeness of HO(M/w'~7M)'
Since B normalizes N, we can think of the action of T = B j N on V Nand
the K-eigenspace VN[K] of VN. By definition, 13 = 13(00) induces
13 = 131< : HO(S,~I<) '---+ ~mHO(Sm,~~J -t VN[Kj.
Proposition 8.2 Suppose (Hpl) for S. The above map 131< is an injection.
Proof. Since Tm,n is faithfully flat and etale over Sm, we may make a
base change T::',n/S Tn to T::',n/S Tn xSTn Tm,m, and hence we may suppose
that P is constant. Then V: 00 is made of locally constant functions on
GLg(Zp)jN(Zp) with values i~ Vm,o. By taking the limit, V N is the space
C(GLg(Zp)jN(Zp), Voo,o) of continuous functions on GLg(Zp)jN(Zp) with
values in V00,0 = ~ m Vm,O. Then H O(S, ~I<) is inside the limit of global sec-
tions of \i!!!mRVTn,o[K], which injects into C(GLg(Zp)jN(Zp), Voo,O) [K]. This
shows the assertion. D
We now put, for 13 = 13(00),
Theorem 8.3 Suppose (Hpl) for S. The inclusion 13 = 13(00) : D' -t VN

induces an isomorphism D' jpm D' ~ V N jpmvN for all m. In other words,
D' is p-adically dense in V.
This theorem shows that (Hpl) implies (Hp2).

Proof. The injectivity of D' jpm D' -+ VN jpmvN follows from the definition
(see [Kl] Appendix III), or we can argue as follows. After the faithfully flat
extension to T,XJ,oo = ~mTm,m over Voo,o = ~mVm,O' P becomes constant,
the base So is replaced by Too,oo = So xSo Too, 00 , and D" = D'0voo .o V is the
space of polynomial functions on FlagN = (GL(g)jN)jQp with values in V[~]
whose restriction to GLg(71p)jN(71p) has values in V. In this situation, we
have V/roo.oo = C(GLg(71p)jN(71p), V). Then it is obvious that D"jpmD" =
(D' jpm D') 0voo ,0 V injects into
V/L.oo 0w Wm = C(GLg(71p)jN(71p), Vm,oo) = (VN jpmvN) ®voo,o V.
Then by the faithfully flat descent, the original map is injective.
We prove that D' jpm D' -+ VN jpmvN is surjective. Nakayama's lemma
(e.g., [BCM] 11.3.2) applied to Wm and its nilpotent ideal pWm tells us that
the surjectivity for m > 1 follows from the surjectivity for m = 1. Thus we
need to prove that D'jpD' -+ V N jpV N is surjective. Since Tm,njSm is etale
finite, replacing Sm by Tm,oo, we may again assume that P is constant (see
[Kl] Appendix III pages 364-5), because we can recover the global sections of
i!dl< over Sm as Galois invariants of that over Tm,n. Then
0TN
1.n
= OS1 0w W[(GL(g)jN)(71jp n71)] = OS1 [(GL(g)jN) (71jp n71)].
This shows vtoo = .cC((GL(g)jN)(71p), VI,o), where .cC((GL(g)jN) (71 p), VI,o)
is the space of locally constant functions on the p-adic space (GL(g)jN)(71p)
with values in HO(SI, OS1) = VI,o. Writing V!,o as a union of W-free modules
X of finite rank, we have
.cC((GL(g)jN)(71p), V:,o) = U.cC((GL(g)jN)(71p), X).

x
Thus we need to prove that .cC((GL(g)jN)(71p),XjpX) = TJxjpTJ x , where
'Dx is the space of polynomial functions of homogeneous degree» 0 (with
= W [~] on the flag manifold FlagN) which has values in
coefficients in K
X over GLg(71p). This last fact follows from Mahler's theorem (e.g., [LFE]
Section 3.2) of the density of binomial polynomials in the space of continuous
functions on tl p with values in 7l p .
Let us explain the use of Mahler's theorem in more detail. Let 6 be the
subgroup of GLg(71p) made of permutation matrices. Then the Bruhat-Tits
decomposition (e.g., [RAG] 1.1.9) gives us the following disjoint unions
wEe;
GLg(lFp) = U N(lFp)T(lFp)w, N(lFp),

wEe;
which can be checked by a direct matrix computation. By the above decom-

position of GLg(lFp), we have a polynomial function on GL(g) whose reduc-
tion modulo p gives rise to a characteristic function of a given double coset
N(71.p)T(71.p)w·N(71.p). In other words, each coset except for the one containing
the antidiagonal Wo (the longest element in 6) is Zariski closed, we can find a
polynomial P vanishing outside the coset N(lFp)T(lFp)wo . N(lF p) but nonva-
nishing on N(lFp)T(lFp)wo . N(lF p). Then PWQ = ppN -1 mod p (for a suitable
positive integer N) gives a characteristic function of N(lFp)T(lFp)wo . N(lFp).
Then 1 - PWQ is supperted by the union of N(lFp)T(lFp)w . N(lFp) with W
length less than that of Wo0 By induction on the length of w, we can create a
polynomial Pw with coefficients in the field K of fractions of W with values in
Won GLg(71.p)jN(71.p) whose reduction modulo p is a characteristic function
of N(lFp)T(lFp)w . N(lFp). Thus for each ¢ E .cC(GLg(71.p)jN(71.p), XjpX) , we
only need to write down ¢ restricted to N(71.p)T(71.p)w . N(71.p)jN(71.p) as a
polynomial function in 'D x j p'D x. Again it is easy to see that
for a suitable integer r (basically projecting down to the coordinate of

N(71.p)T(71.p) modulo W· N(71.p)w- 1 n TN(71.p)). Write this coordinate function
as (x, t) = (Xl"'" XT) h, ... , tg), where tj are the diagonal entries of T. Then
on 71.p, any locally constant function ¢ : 71.p ---+ XjpX has a unique polynomial
expansion 71.p :3 z f--t I:n>O Cn (~) with Cn E XjpX (for the nth binomial poly-
nomial (~)) by Mahler's theorem. In particular, on 71.;, the monomial zn(p-1)
is the constant 1 modulo p. Thus we can write any locally constant function
¢ : N(71.p)T(71.p)w . N(71.p)jN(71.p) ---+ XjpX as a polynomial of (x, t), and af-
ter multiplying by a high-power det(t)n(p-1) = (I1 t,t(P-1), the polynomial
function is in the sum of the K-eigenspaces for the sufficiently regular KS. This
finishes the proof.
We now assume that there exists a proper flat scheme M jw such that
ScM and that M - S is a proper closed subscheme of co dimension ~ 1. We
further assume that ~jS extends to M. Then ~~ extends automatically to
Mm by the theory of flag varieties given in Section 6.1.3. By the properness of
M, H1(M,~"') is a W-module of finite type. Taking the projective limit with
respect to m of the exact sequences
we get ~mHO(Mm'~~) = HO(M,~~). Let Rm = EB",»oHO(Mm,~~) and

R = EB",»oHO(M'~7M)' Then R is p-adically dense in Roo = ~mRm.
By definition, R C R'. Note that det(~)p-1 is trivial on 8 1. Let a E
HO(8 1, det(~)p-1) be the section corresponding to 1 E det(~1)P-1 ~ OS,.
We assume that a extends to M1 so that it vanishes outside 8 1, Suppose that
(E) we have a section E E HO(M,det(~)t(p-1)) such that E mod 'W = at.
By further raising the power, that is, replacing E by EP=, we may assume
that E mod p = at. Then by definition,
We now show that ,B(R[i]) n VN is dense in VN. Pick LI< fl< E pmvN
for fl< E HO(8,~I<). We need to approximate f = fl< modulo pm+lv N by an
element in HO(M,~I< 0 det(~)k). This section f E HO(8,~I<) can be written
as f == gdE£ mod pm+l for g£ E HO(M,~I< 0 det(~;d). Then for k > e,
we have f == geEk-£ j Ek mod pm+!. We may assume that k = pmt(p - 1).
Then as a function of (X, 'I/J), Ek mod pm+l is a constant. Thus f == geEk-£
mod pm+l Roo. This shows the density of ,B(R [i]) V N in VN.n
Corollary 8.4 Suppose the following conditions for 8 in addition to (Hp1).
(1) We have ScM for a proper fiat scheme M/ w such that 8 m C Mm =
M 0w W m is Zariski dense for all m;
(2) ~ extends to a locally free vector bundle on M of rank g;
(3) There exist an integer t > 0 and a section E E HO(M, det(~)t(p-l)) such
that E mod w is the constant section 1 generating det(~l)t(p-l) ~ OSI;
(4) M - 8 is the zero locus of the section E.
Put D = ,B (EB 1<»0 H O(M, ~I<)) [i] nV N. Then D is p-adically dense in V N .
8.1.5 p-Ordinary Automorphic Forms
In addition to (Hp1) (and hence (Hp2)), we now suppose having a projector

e (so e 2 = e) acting (continuously) on V N , which projects down VN[II:] onto a
W-free module of finite rank (for all II: »0). We put VN = ~n VN jpnvN =
~nV:'oo. We have VN[II:] = ~mV:'m[4 Since VN is W-flat, VN is p-
divisible, and its direct summand eVN is p-divisible. A p-divisible W-module
is of finite co rank if its Pontryagin dual module is of finite rank over W.
In practice, the projector e will be constructed so that it brings VN [11:]
down onto eHO(M/w,~1< 01I'p) for 11:» 0, where lI'p = QpjZp. This implies
eVN[II:] = eHO(S/w,~1< 0z p lI'p) = eHO(M/w,~1< 0z p lI'p) (8.7)
if II: » O. By (Hp1), HO(S/w,~K, 0z p lI'p) is p-divisible. By assuming (8.7),

eVN[II:] is p-divisible. Since M is proper over W, HO(M/w,~I<) is a W-module
of finite type (see [EGA] III.3.2.1), and eVN[II:] is a p-divisible module of finite
corank. In any case, we simply assume that eV[II:] is p-divisible and of finite
corank for II: » o.
Let VN be the Pontryagin dual module of VN. Since VN is a discrete
T(Zp)-module, VN is a compact W[[T(Zp)]]-module. Let T(Zp)p = rT be the
p-profinite part of T(Zp); in other words, we have T(Zp) = T x Ll for a r
finite group Ll of order prime to p. The continuous group algebra W[[rTll =

~nW[rT/rr]] is isomorphic to a formal power series ring W[[xl, ... ,xgll
sending a set of minimal generators {1'1, ... ,1'g} of r T to {I + Xl, ... , 1 + Xg}
in W[[xl, ... ,xgll. Thus m = (W,Xl, ... ,Xg) is a unique maximal ideal of
W[[rT ]], and W[[rTll is P-adically complete for any closed ideal Pc m under
the m-adic topology. For an algebraic character r;, E X(T), the character
r;, : rT -t W X extends by continuity to a W-algebra homomorphism r;, :
W[[rTll -t W. Then PK = Ker(r;, : W[[rTll -t W) is a closed ideal of W[[rTll
(again by continuity).
We fix a character X : Ll -t IF x for IF = W/wW and write X : Ll -t W X
for the Teichmiiller lift of X. We write Xx c X+(T) for the collection of all
algebraic characters r;, : T -t <G m satisfying the following two conditions:
• r;, == X mod w;
• r;, is sufficiently regular so that the above equation (8.7) holds.
Then Xx is Zariski dense in Spec(W[[rT]])(W). We write V;rd for eVN. Let
us decompose V;rd = EBXE.:1 V;rd[X] into the direct sum of the x-eigenspaces
V;rd[X] under the action of Ll. The summand V;rd[X] is a compact module
over W[[rTll. Then by (8.7), V;rd[X]0W[[IT]],K W is W-free of finite rank s(X)
for r;, E Xx. This implies (V;rd[X]/ P;'V;rd[X]) 0w[[IT]],K W ~ ws(x) for any
m ~ 1. We apply Nakayama's lemma (e.g., [BCM] II.3.2) to the local ring
W[[rTll/ P;' and its nilpotent ideal PK and find that (V;rd[X]/ P;'V;rd[X]) is
a W[[rT]]-module of finite type with minimum number s(X) of generators.
Taking the projective limit with respect to m, we conclude that V;rd[X] is a
W[[rTll-module of finite type with minimum number s(X) of generators. Since
Xx is Zariski dense in Spec(W[[rT]]), we see that V;rd[X] is W[[rTll-free of
rank s = s(X). Indeed, take s generators VI, ... , Vs and consider the surjection
7f: W[[rTW -t V;rd[X] given by 7f(al, ... , as) = Lj aJvJ. After taking modulo
PK, 7f reduces to an isomorphism; so, Ker(7f) C PKV;rd[X]. Since Xx is Zariski
dense, we find Ker(7f) C nKEX-
PKV;rd[X] = {O}, getting injectivity of 7f.
)(
We conclude, assuming (8.7) for the middle equality,
rankW[[ITJ] V;rd[X] = rankw V;rd[X] 0W[[ITJ],K W

= rankw(eHo(M, ~;t) 0z p lI'p)* = rankw eHo(M, ~K) (8.8)
for all r;, E Xx. Therefore we get
Theorem 8.5 Suppose (Hpl), the existence of the idempotent e : VN -t VN
as above and the assumptions of Corollary 8.4. Then V;rd is a well-controlled
W[[T(Zp)ll-projective module of finite type. If we assume (8.7), this means
V;rd 0W[[T(Zp)]],K W ~ Homw(eHO(M/w,~7w)' W)
canonically if r;, is sufficiently regular. For each x-component, V;rd[X] is free

of finite rank over w[[rTll for the maximal p-profinite subgroup r T ofT(Zp).
8.1.6 Construction of the Projector eGL
We construct an approximation eCL of the projector e. In the paper [H02],

Section 2.6, it is written: "eCL is constructed using solely local data of the
Galois group Gal(Tm,ooj8m ) = GLnCZp) , while the projector e will be con-
structed as e = eCeCL for a global projector ec depending on the group G."
This statement is misleading. We indeed need a global input to define eCL. To
explain this point, let us recall from Section 5.1.2 the expanding semi-group
of GLn(!Qp). We introduce two subgroups IB,n and IN,n of GLg(7Lp) by
IB,n = {x E GLg(7Lp) Ix mod pn E B(7Ljpn7L)} ,

IN,n = {x E IB,nIX mod pn E N(7Ljpn7L)}.
Let diag[X l , ... , X J ] denote the diagonal matrix with diagonal blocks Xl to
X J from the top. We recall the subsemi-group of GL(g) given by
D = DCL(g) = {diag[pe 1 , . . . ,peg]lel :-:; e2 :-:; ... :-:; eg}.
Then Ll~ = h,nDh,n Ll~ = N(7Lp)DN(7Lp) and Ll! = B(7Lp)DB(7Lp) are

semi-groups, and we call them expanding semi-groups (see Section 5.1.2). If
confusion is unlikely, we simply write Ll for one of these semi-groups.
The global input we need comes from the fact that Tm,n in our appli-
cation classifies not just trivializations of Pn but abelian varieties X with
X[pn]et S:! Pn . In other words, each 9 E GLn('Qp) acts on 8 m by an appropri-
ate isogeny of classified abelian varieties, and it acts not only on the etale quo-
tient of the p-divisible group of the abelian variety but also on the connected
component of the p-divisible group. The action changes the isomorphism class
of the abelian varieties, and hence it acts on 8 m through endomorphisms (not
necessarily through automorphisms). Since at this point we do not assume
that 8 m classifies abelian varieties, we instead assume having such an ac-
tion of the expanding semi-group (as defined below) on 8 m which is at worst
radiciel mod Pi so, it does not affect the etale trivialization Pn . This action
f
6: 8 m -+ 8 m sends an 8 m -scheme X --'-+ 8 m to 6· X = X XSm.,o 8m.
We consider the following triples
where fil n : Pn / X = P~g) :::) p~g-l) :::) ... :::) p~O) = {O} with 7Ljpn 7L 't;,
p~J) j p~J-l) for j = 1, ... , g. If P is constant, the space classifying the
above test objects over 8 m is given by T!/",n = Tm,njN(7Ljp n7L). Simi-
larly, the classifying space of couples (X, filn) over 8 m is given by T! n =
Tm,nj B(7Ljp n7L). On test objects over Tm,n, we have a natural actio'n of
h E GLn(7Lp) given by (X, 'lj;) H (X, h'lj;). Writing fil n = 'lj;-lln, we then
see that 'lj;-lh-1l n = 'lj;-lln = h· fil n . Thus the Galois action on filtra-
tions is given by h· fil = ('lj;-lh-l'lj;)P~J) and h· ¢ = ('lj;-lh-1'lj;)¢, where
'lj; : Pn ~ (Zjpnz)g such that 'lj;* (In, id) = (filn' ¢n) for the standard identi-
fication id : gr(ln) ~ (Zjpnz)g. Thus these test objects are always invariant
under N(Zp). The new test objects (X, filn, ¢n) are useful in defining an
isogeny action of IS E L1 and in constructing the idempotent eGL. We assume
(d1) IS induces an isomorphism IS* filnjlj.x ~ fil n/ x compatible with the
action of the semi-group on fil n (this holds if IS mod p is radiciel),
where the action of IS E L1 on fil n is the multiplication by IS up to
scalars (as we specify later);
(d2) h· X = X if hE GLn(Zp).
Let diag[xl, ... , x]] denote the diagonal matrix whose diagonal blocks are
given by square matrices Xl to X] from the top. We put a] = diag[l g _],p1)].
We consider the double coset ring R(h,n, L1~) (n = 1,2, ... ,00 and? = B, N)
made of Z-linear combinations of double cosets h,nlSh,n for IS E L1~.
As seen in Lemma 5.2 or its proof, once a decomposition N(Zp)a]N(Zp) =
U) N(Zp)lS) is chosen, h,na)h,n for any nand? is decomposed in the same
way, h,na)h,n = U) h,nlS). In other words, we have for a = I1;:i a),
By this fact, R(h,n, L1~) ~ Z[h, ... , tg] (a polynomial ring). See Lemma 5.2
for a proof, though the semi-group D here is a proper subsemi-group of
the expanding semi-group there, strictly speaking. For IS E L1f, the action
p(lS- l ) is well-defined on RA[/'t:] for any p-adic ring A, because p(lS- l )¢(y) =
¢(lSy) for y E GL(g)jN. Decompose h,nlSh,n = U) h,nlS) and regarding
f E HO(T:;',n,f;d.K) as a function of test objects X/ T ", we define
,00 ,
fl[h,nlSh,n](X) = I:>K(IS;l)f(IS)X), (8,10)

)
where ISX = (IS· X, IS(IS* filn)). Strictly speaking, the sum above is actually
heuristic, because if the action of IS on S is wildly ramified, that is, radiciel
(i.e" purely inseparable in characteristic p), we need to replace the sum by
the trace as already described (4.18) in Section 4.1.10 (for Hilbert modular
forms). We clarify this point in Section 8.3 more carefully; so, for the moment,
we content ourselves with this heuristic description.
Although we have not yet specified the action of IS E L1~ on IS* filn' if it
is done, then the operator is well defined independently of the choice of IS)
because for U E h,n we have
Furthermore, by (8.9),
for f E HO(T!,n,f;d.K), flt(p)n-l E HO(T!,l,f;d.K), (8.11)

where t(p) = I1;:i

tJ . When Pn comes from a universal abelian scheme, we
have a natural isogeny action on test objects, and in this way, we can define
Hecke operators on HO (Sm, Idl<).
Since filn is an element of the flag variety of (7L/pn7L)g, to study the action
of L1 on filtrations, we study general flag varieties. For each commutative ring
A, we consider the free module L = L(A) = Ag and the flag space
y(Ag) = {(Li) IL = Lg, L, ::J L,-l, Ld L,-l ~ A for i = 1,··· , g}, (8.12)
Y(Ag) = {(L" CPi) I(L,) E y(A), CPt : A ~ L,/ L,-l for i = 1, ... , g}. (8.13)
We can extend the above definition to Pn over Sm. Define for T/Sm ,
Y(F)
n jT -
- {(P(i) )
njT
IFn -_ peg)
n ,
pet) p(,-l) p~,) rv 7L (
n ::J n
. )}
, p~'-l) = pn7L 1::; z ::; 9 ,
Y(Pn)jT = {(P~;T' cp,)I(p~')) E y(Pn)jT, CPt : P~7L ~ ;'~)1) (1::; i::; g)}.
After a finite etale extension to T / Sm, the spaces y(Pn ) and Y(Pn ) get isomor-
phic to y((7L/pn7L)g) xSrn T and Y((7L/pn7L)g) xSrn T, respectively. Writing the
standard base of Las e1, .. " eg, we define I = 0:=;=1
Ae" CPt = id) E Y(Ag),
which we call the origin. We may let GLg(A) act on Y(Ag) and y(Ag) by
x((L,), ¢,) = (xL" x 0 CPt). Then GLg(A)/N(A) ~ Y(Ag) by xN(A) H xl.
Now we assume A to be a p-adic ring. We then define
Similarly, we define YnjT(Pnf) for n ::; n' ::; 00. We note that Yn(7Lg) ~
IN,n/N(7Lp) via xl +-+ x and similarly Yn(7Lg) ~ IB,n/N(7Lp). So we have the
conjugate action of L1 on these spaces introduced in Section 5.1.3.
We now write down explicitly the conjugate action of the semi-group L1
on Yn(Ag). Since y(7Lg) = y(Qg) (because y = FlagB is projective), the group
GLg(Qp) acts naturally on y(7Lg). This action is described as follows. Take
x E GLg(Qp). Then x(L,) = (xL, Q9 Qp n L(7Lp)) E y(7Lg). We write x(L,) =
(x· L,), that is, X· Li = xL, Q9 Qp n L(7Lp). We claim
Lemma 8.6 Let d = diag[pe1, ... ,pe g ] E DCL(g). For (L"cp,) E Yn(7Lg),
[d]i : L,/ L,-l -+ b· Ldb· L,-l given by (e mod L,-l) H (p-e'de mod L,-d
for eEL, is a surjective isomorphism.
Proof. The 7Lp-module L, is generated by L,-l and a vector e whose first i - 1
entries vanish. Then de is divisible by pe, because its bottom 9 - i entries are
divisible by pe'+l because of eg 2: ... 2: e,+l 2: e,. From this, the linear
map (d), : (QpL,/QpL,_l) -+ d(QpL,)/d(QpL,-d induced by d is divisible
by pe, on L,/ L,-l C (QpL,/QpL,-d. By definition, we have d· L, = L" and
p-e'(d), : LdL'-l -+ L,/L,_l = d· L,/d· L,-l is a surjective isomorphism.
o
We now define an action of the semi-group L1;; on Yn(;Z~). Since u E IE,n

acts naturally on flags, multiplication by u gives rise to an isomorphism fUji :
L,/ L,-1 ~ u . L, . L i - 1. Defining a linear map [8], : Ld L i - 1 ~ 8· L,/8 . L i - 1
by [u], 0 [d]i 0 [U']i' the above action of DCL(g) on Yn(;Z~) extends to an action
of the semi-group L1;;. In particular, we have 8· (Li' cp,) = (8· L i , [8], 0 CPi) for
(Li' cp,) E Yn(;Z~).
For a given X = (X, fil) = (X,1/J-l1), 1/J brings fil to 1, and hence
the action of L1;; defined on the neighborhood of 1 (after conjugation by 1/J)
is enough to get an association X H {8J X}. By this, after extending base
scheme Sm (e.g., to T = Tm,oo) to trivialize Y(P), we have an action of L1;;
on Y(P). However, this is sufficient to define the Hecke operators [h,n8h,n]
acting on HO(Tm,n/s m , ~K) for the following reason. After extending the scalar,
define fl[h,n8h,n] by (8.10). The formation of fl[h,n8h,n] commutes with the
base change; in other words, it commutes with the Galois action of the base
GLg(Zp) c Gal(T/Sm ); so, fl[h,n8h,n] is actually defined over the original
base scheme T'!n,n/S m ' This justifies the contraction property (8.ll).
Write N for N(Zp) simply. Let t J = NaJN in R(N, L1~) with a J =
diag[l J ,p1g- J], and define t(p) = TI]=1 t J . Let a = TI]:{ aj. By computation,
we see easily that an contracts Yn(Z/pn+lZ) to the origin 1 n+1'
Identifying Yl (Z/pnz) with I E,1/ IN,n, if the filtration fil n corresponds to
x E IE,n, then the filtration is given by L~=1 (Z/pnZ)x J for the jth column
vector Xj of x. Choose a representative set N(Zp)aN(Zp) = UUER N(Zp)a u .
Then we have IE,n+laIN,n = U UER IE,n+lau, and a u x- 1 = x;;:la u' for some
u' E R. This X-I H X;;:1 coincides with the action of au on the flag vari-
ety Yl ((Z/pnz)g) if one identifies elements in I E,1 with a flag. Here we need
to use X-I instead of x, because the action of h E G Lg(Zp) on filtrations
is given by fil H h· fil = (1/J-lh- 11/J)fil as already remarked. The ele-
ment Xu gives rise to a couple (au' X, fi1n+l,u = au(a~filn+d), which is
uniquely determined independently of the choice of au. We then define for
f E V~,n+l flt(p)(X, fil n ) = LUER f(a u . X, fil n+l,u)' Similarly, if we start
from f E V:'n' by the same process, we get Xu E IN,n/N(Zp) corresponding
to (X, filu,n, CPu), because we still have IN,naIN,n = UUER IN,nau. Define
flt(p)(X, filn, CPn) = L f(a u . X, fi1u,n, CPu) (8.14)

uER
and define the idempotent eCL by eCL = lim n --+ oo t(p)n! whenever it is well-
defined. As we have seen (5.10), K:(a)t(p) = t(p) on HO(S,~K), because on ~K
we used the action of L1 coming from schematic induction.
8.1.7 Axiomatic Control Result
In this subsection, we describe a simple prerequisite to have the control the-

orem relating false automorphic forms (sections over S) to true automorphic
8.2 Deformation Theory of Serre and Tate 345
forms (sections over M). Later we will verify the requirement for automorphic
vector bundles on Shimura varieties of PEL type.
Since in this general situation Sm is not supposed to classify anything, we
cannot define Hecke operators acting on HO(Sm, ~"') in this generality. Any-
way, we suppose a Hecke operator T(p) acting on HO(Sm,~"') and HO(Mm , ~"')
such that T(p) == t(p) on HO(SI'~"') if K , » O. We define eo = limn-too T(p)n!.
Further suppose that there exist two projectors ee acting on V:'oo and ee on
HO(Mm'~"') for K, » 0 (depending on the reductive group G) such that
eeeeL = ee, eeeo = ee, and ee == ee mod p; that is, they are equal to each
other on ecHO(SI'~"')' In addition to the above conditions, writing K for the
field of fractions of W, we suppose the following two conditions,
(C) ec(E· f) = E· (ecf) for I E HO(SI'~"') for E in (E) of Section 8.1.3,
(F) dimK eeHO(M/K'~'" Q9 det(~)'K) is bounded independently of k.
Let 71,72' ... be a sequence of linearly independent elements in ecHO(SI, ~"')
over WI' Since HO(S/w,~'lw) Q9w WI = HO(SI'~"') (Hpl), we can lift 7t
to Ii E HO (S /w, ~'lw) so that 7t = (It mod p). Then for any given in-
teger L > 0, we can find a sufficiently large integer m such that Em It E
HO (M/w, ~'lw ). Since multiplication by E is an isomorphism on SI (by def-
inition of SI = Ml [i]), by (C) and ec == ee mod p, the mod p sections
{(ee(Em Ii) mod P)}.=I, ... ,L are linearly independent over WI by Nakayama's
lemma; so, {ee(EmIt)}.=I, ... ,L are linearly independent over W. This implies
dimK eeHO(M/K'~'" Q9 det(~)~<;;-I)m) ?: L.
If rankwl ecHO(SI'~"') = 00, we can take L to be arbitrarily large, which

contradicts the boundedness of the dimension (F). Thus rankwl ecHO(SI'~"')
has to be finite, and rank w1 ecHO(Sll~"'Q9det(~)k(p-l)) is independent of k.
Thus the existence of the desired projector follows from (F), (C) and (8.7).
The condition (F) can be proven in our application, as in the proof of
Theorem 4.9, via group cohomology using the (generalized) Eichler-Shimura
isomorphism combined with the p-adic density of Dcusp in D~usp (see [H95] for
such boundedness for inner and outer forms of GL(n), [TiU] for inner forms
of GSp(2n) , and Section 4.3 in the text for GL(2)).
The condition (C) can be proven either by q-expansion or the fact that
the Hasse invariant does not change after division of an abelian variety by an
etale subgroup.
Hereafter we write H2rd(X, ?) for the image ee (Hq(X, ?)) if X = M and
for the image ec (Hq(X, ?)) if X = Sm and Tm,n.
8.2 Deformation Theory of Serre and Tate

Recall W which is a complete discrete valuation ring of mixed characteristic
with residue field IF of characteristic p. We suppose that IF is algebraic over
lFp = ZjpZ. After Section 8.2.3, we suppose that IF is an algebraic closure of

lFp. In this section, we describe deformation theory of abelian schemes over
local Wm-algebras for Wm = Wjpmw. This theory is useful for proving the
p-integrality of the Hecke operator T(~) after the division by J-l(~) as in (5.40).
We principally follow Katz's exposition [K2].
8.2.1 A Theorem of Drinfeld
In this and the following subsection, we take a local Wm-algebra B as a base

ring. We consider the category B-LR = LR/B made up of p-adically com-
plete local B-algebras R = lj!!!: nRj pn Rand morphisms are supposed to be
p-adically continuous. For simplicity, we always assume that rings we consider
are Noetherian.
Let G : B-LR -t AB be a covariant functor into the category AB of
abelian groups. Thus if we regard G as a functor from the category of affine
B-schemes (or formal B-schemes), it is contravariant. Suppose that, for any
faithfully flat extension of finite type R y G of B-algebras,
1. The group G(R) injects into G(G), that is, G(R) y G(G);
2. Let G' = G ®R G and Gil = G ®R G ®R G. Write L, : G y G' (i = 1,2)
two inclusions (i.e., L1(r) = r ® 1) and L'l : G' Y Gil three inclusions
(i.e., L12(r ® s) = r ® s ® 1). If x E G(G) satisfies G(Ld(x) = G(L2)(X)
and G(L12)(Y) = G(L23)(Y) = G(L13)(Y) for Y := G(Ld(x), then x is in the
image of G(R).
Such a G is called an abelian sheaf On B-LR under the fppf-topology (or
simply an abelian fppf-sheafJ. If A/B is an abelian scheme, then
G(R) = A(R) = Homs(Spec(R) , A)

(S = Spec(B) or Spf(B)) is an fppf-sheaf. We call an fppf-sheaf G/ B a p-
divisible group if for any x E G(R), there exists a finite faithfully flat extension
G of R and a point Y E G(G) such that x = py. If Gis p-divisible of p-torsion
and G[P]/B = Ker(p : G -t G) is a group scheme locally B-free of finite
rank, it is called a Barsotti-Tate group (cf. [CBT] I). If G comes from an
abelian scheme A, it is p-divisible. This also shows that A[POO] = Un A[pn] for
A[pn] = Ker(pn : A -t A) is a Barsotti-Tate group (e.g., [GME] 4.1.18).
Let B be a local W m-algebra and I be an ideal of B such that I"+1 = 0
and N I = 0 for a power N of p. We define a new functor Gland 8 by
G1(R) = Ker(G(R) -t G(RjI)) and 8(R) = Ker(G(R) -t G(RjmR)),
where mR is the maximal ideal of R. When 8(R) = HomB-LRCR, R)(= G(R))

for R = B[lTl, ... , Tnll (i.e., 8/ 13 = Spf(R)/B) and the identity element 0 cor-
responding to the ideal (T1, ... , Tn), we call 8 a formal group. If 8 is formal,
G1(R) = {(t1,"" t n ) E In} by HomB-LR(R, R) 3 ¢ t--'t (¢(Td, ... , ¢(Tn)).
Suppose that G/13 is formal. Then multiplication by [N] induces a contin-

uous algebra homomorphism [N] : R -+ R. Then on the tangent space at the
origin tG = (TI , ... , Tn)/(TI , ... , Tn)2, the multiplication by p on G (which
is written as [P] : G -+ G) induces multiplication by p, because the addition
induced by the group law of G coincides with the addition of the tangent vec-
tors (cf. [ABV] Section 11) by the uniqueness of the B-linear structure on tG.
Thus [N](T,) == NT, mod (TI , ... , Tn)2, and [N](GJ(R» = Gp(R) because
NI = O. Similarly, we have [N](GJu(R» = GJu+l(R) inductively. This implies
that [NV]GJ = Go = {O}. We get
G J C Ker([NV] : G -+ G) if G is formal. (8.15)
Theorem 8.7 (Drinfeld) Let G and H be abelian fppf -sheaves over B-LR
and I be as above, and suppose that JV+l = O. Let Go and Ho be the restriction
of G and H to B / I -LR. Suppose
(i) G is p-divisible;
(ii) H is formal;
(iii) H(R) -+ H(R/J) is surjective for any nilpotent idealJ (i.e., H is formally
smooth).
Then we have the following four assertions:
(1) Hom13_Gp(G, H) and Hom13/J_Gp(G o, Ho) is p-torsion-free, where the sym-
bol ''Homx_Gp'' stands for the homomorphisms of abelian fppf-sheaves
over X-LR;
(2) The natural map, the so-called "reduction mod I" from Hom13_Gp( G, H)
into Hom13/J_Gp(G o,Ho) is injective;
(3) For fo E Hom13/J_Gp(G o, H o), we have a unique P E Hom13_Gp(G, H) such
that P mod I = NV fo. We write jJv f for P in Hom13_Gp(G, H) ®z Q;
(4) In order that f = t;J E Hom13_Gp (G, H), it is necessary and sufficient
that jJv f kills G[NV].
Proof. The first assertion follows from p-divisibility, because if pf(x) = 0 for
all x, taking y with py = x, we find f(x) = pf(y) = 0 and hence f = O.
We have an exact sequence 0 -+ HJ -+ H -+ Ho -+ 0; so, we have another
exact sequence
0-+ Hom(G,HJ) -+ Hom(G, H) mod I) Hom(G,Ho) = Hom(Go,Ho),
which tells us the injectivity since HI is killed by NV and Hom(G, H) is p-

torsion-free.
To show (3), take fo E Hom(G o, Ho). By the surjectivity of H(R) -+
Ho(R/ I), we can lift fo(x mod I) to y E H(R). The class y mod Ker(H -+
Ho) is uniquely determined. Since Ker(H -+ Ho) is killed by NV, NVy for
any x E G(R) is uniquely determined; so, x f-t y induces the functorial map
P = jJv f : G(R) -+ H(R). This shows (3).
The assertion (4) is then clear from p-divisibility of G. The uniqueness of

f follows from the p-torsion-freeness of Hom(G, H). 0
8.2.2 A Theorem of Serre-Tate
Let A/B be the category of abelian schemes defined over B. We consider a

category Def(B, B/ I) of triples (Ao, g, f), where Ao is an abelian scheme over
Bo = B / I, a Barsotti-Tate group 9 over Band f : go ~ Ao [po,"]. We have a
natural functor A/B --+ Def(B, Bo) given by
A f--t (Ao = A mod I, A[poo], id).
Theorem 8.8 (Serre-Tate) The above functor A/B --+ Def(B, Bo) is an
equivalence of categories.
Proof. By Drinfeld's theorem applied to A[POO] and A (both abelian fppf-
sheaf), the functor is fully faithful.
It is known that we can lift Ao to an abelian scheme Y over B. This follows
from the deformation theory of Grothendieck (e.g., [CBT] 2.8.1). Here is a
heuristic argument to find Y/ B under the assumption that Bo is a finite field.
In this case, by a theorem of Tate (see [T1], [D], and [ABV] Section 22), Ao
has complex multiplication by a semi-simple commutative algebra M (with
2 . dim Ao = dimlQ M) whose simple components are all CM fields with its
complex conjugation induced by a Rosati involution. By the theory of abelian
varieties with complex multiplication, Ao can be lifted to an abelian scheme
Y over B with complex multiplication (this is a special case of Theorem 7.1
taking D = V = M). Thus we have an isomorphism a6P ) : Yo[POO] --+ Ao[poo],
and we have a unique lifting (by Drinfeld's theorem) f = NVa(p) : Y[POO] --+ 9
of NVa6P). Since the special fiber is an isogeny having inverse (afr))-l, f is
an isogeny, whose (quasi) inverse is the lift of N V (a6P))-1. Thus Ker(f) is a
finite flat group subscheme of Y. The geometric quotient of Y by a finite flat
group subscheme exists (see [ABV] Section 12) and is an abelian scheme over
B. Then dividing Y by Ker(f), we get the desired A/B E A/B. 0
Suppose Bo is a field of characteristic p. As described in [CBT], IV.2.5 and
V.lo4, we can attach a crystal (]]])B(90) ::) Fl) to the data (Ao, g, c:) formed by
a B-free module ]]])B(90) and a submodule Fl c ]]])B(90) with B-free quotient
]]])B(90)/F 1 such that ]]])B(90) @B Bo = ]]])Bo(90) is the Dieudonne module of
go and Fl @B Bo gives Lie(9o) c ]]])Bo(90)' Then we have
End(A/B) = {a E EndB(Ao/Bo)la*(Fl) C Fl},
regarding End(A/B) C End(Ao/Bo) by Drinfeld's theorem. By this fact com-
bined with a version of the above theorem ([CBT] V.lo1O), one can verify
(using (det) and D-linearity of the polarization Ao) that (A o,Ao,iO,1](p)) E
[(p)(Bo) for the functor [(p) in Theorem 7.1 can be lifted to (A, A, i, 1](p)) E
[(p)(B) (as shown in [LaR] 6.2 and [ZiJ).
8.2.3 Deformation of an Ordinary Abelian Variety
Hereafter we suppose that the residue field IF of W is an algebraic closure of

lF p . We assume the base ring 13 to be a pro-Artinian local ring with residue
field IF. Write CL IB for the category of complete local 13-algebras with residue
field IF (under the ffiB-adic topology). We fix an ordinary abelian variety A o/lF .
Recall that tAIB denotes the dual abelian scheme of an abelian scheme AlB
and that AlB denotes the formal group associated with A. We write T A[poo]et
for the Tate module ~ the~maximal etale quotient of A[poo]. We consider the
deformation functor P = PAo : CLIB --+ SETS given by
PAo(R) = [(AIR, ~A)I AIR is an abelian scheme and ~A : A ®R IF ~ Ao] .
Here f: (A,LA)IR ~ (A',LA')IR if f: A --+ A' is an isomorphism of abelian

schemes with the following commutative diagram,
Theorem 8.9 (Serre-Tate) Let the notation and assumption be as above.

Then we have
(1) A canonical isomorphism peR) ~ Hom(T Ao[poolet x TAo [poo]et, Gm(R))
taking (AIR, ~A) to qAIR(-, .).
(2) The functor P is represented by the formal scheme
Hom Zp (TA 0 [poo]et x rrtA 0 [poolet ,G

m ) -c::,o Ggm·
2
I
(3) qAIR(x, y) = qtAIR(Y, x) under the canonical identification: t(tA) = A.

(4) Let fo : AO/lF --+ Y O/lF be a homomorphism of two ordinary abelian vari-
eties with the dual map tfo : tyo --+ tAo. Then fo is induced by a homo-
morphism f: AIR --+ YIR for A E PAo(R) and Y E Pyo(R) if and only if
qAIR(x, tfo(y)) = qYIR(fo(x), y).
Proof. We give a sketch of the construction of qAIR. First we prepare some

facts. Let f : AIR --+ Y I R be an isogeny; so, Ker(f) is a finite flat group scheme
over S. Pick x E Ker(f), and let .c E Ker(if) c Y be a line bundle on Y with
0*[ = Os (S = Spec( R)). Recall that R is an Artinian 13-algebra. Thus
f*[ = OA· Cover Y by affine subsets Ui = Spec(Ri) so that .cl u, = ¢;10 U ,.
Since Oy.c = Os, we may assume that (¢t/¢j) 0 Oy = 1. Since f : A --+ Y
is finite, it is affine. Write ~ = f- 1(Ut ) = Spec(Ri). Then 1* .cl v, = 'P;10 V ,
with 'Pt = ¢i 0 f, and we have, regarding x : S --+ Ker(f) ,
'Pi 0 X ¢i 0 fox ¢i 0 Oy
-~= = =l.
'PJ 0 x ¢j 0 fox ¢J oOy
The morphisms <p, 0 x glue into a morphism e f(x, £) : S -+ G m , and we

get a pairing e f : Ker(f) x Ker(lf) -+ G m . Since A is a Ker(f)-torsor over
Y, we have A Xy A ~ Ker(f) Xs Y. For a homomorphism ( : Ker(f) -+
G m , we can find a morphism ¢ : Ker(f) Xs Y -+ pI of schemes such that
¢(y+t) = ((t)¢(y) for t E Ker(f). This function ¢ gives rise to a divisor D on
YA = Y Xs A. By definition fj,£(D) = VA A , and ef(x, £(D)) = ((x). Thus,
over A, e f IA : Ker(f) IA X Ker(tf) IA -+ G m is a perfect pairing. Since A -+ S
is faithfully flat, we find that the original e f is perfect.
We apply the above argument to f = [pnj : A -+ A, write the pairing as
en, and verify the following points.
(PI) en(a(x), y) = en(x, h(y)) for a E End(AIR);
(P2) Write Ao[pnjD = /L~n C Ao[pnj. Then en induces an isomorphism of
group schemes Ao[pnjD ~ Hom(tAo[pnjet,/Lpn);
(P3) Taking the limit of the above isomorphisms with respect to n, as formal
groups, we find A ~ Hom(TA[pOOjet, Gm ) ~ Hom(TAo[pOOjet, Gm ).
The induced pairing we denote by EA : A x TAo[poojet -+ Gm . In
particular A = G~.
We now describe q. Since R E GLIB is a projective limit of local B-algebras
with nilpotent maximal ideal, we may assume that R is a local B-algebra with
nilpotent maximal ideal mR. Then A(R) is killed by pno for a sufficiently large
no (applying Drinfeld's theorem to I = mR)' Thus if n ;::: no, the structure of
the p-divisible group A[pOOj is uniquely determined by the extension class of
(8.16)
for n ;::: no, which can be rewritten as the following exact sequence:
0-+ Hom(TAo[pOOjet,Gm)[pnj-+ A[pnj ~ Ao[pnjet -+ O. (8.17)
Take x = ~nXn E TAo[poojet for Xn E A[pnjet. Lift Xn to Vn E A[pnj(R)

so that 7r ( vn ) = x n . Then, for pn : A [pn j -+ A (lifting pn times the inclusion
Ao[pnj Y Ao), qn(x) = pnvn E A(R). Take the limit of qn to get q(x) E
Hom(T Ao[poojet, Gm(R)). This q(x) determines the extension class of (8.16)
and thus the deformation AI R because it is determined by (Ao, A[POO]) by the
Serre-Tate theorem in the previous subsection. Then we define qAI R (x, y) =
q(x)(y).
It is known that for any given q(x, y) as above an extension (8.16) exists by
the theory of Barsotti-Tate groups studied by Messing (see [CBTj Appendix).
This shows assertions (1) and (2). All other assertions follow easily from (Pl-
3). 0
8.2.4 Symplectic Case
We now fix a polarization AO : Ao -+ tAo of degree prime to p. We consider

the functor P = PAO,AO : GRIW -+ SETS given by
Here we call f : (A, AA, LA) --+ (Y, Ay, Ly) an isomorphism if f : (A, LA) ~
(Y, Ly) and tf 0 Ay 0 f = AA. Note that by the Drinfeld theorem, End(A/ R)
is torsion-free and, hence, End(A/R) Y EndlQl(A/R) = End(A/R) ®z 1Ql. We
write 00* = AOI 0 h 0 Ao for a E End(Ao) ®z 1Ql. Since End(A/R) C End(Ao)
again by Drinfeld's theorem, the involution keeps EndlQl(A/R) stable (because
on EndlQl(A/R)' it is given by 00* = A-Iota 0 A). The involution a --+ 00* is
known to be positive (see [ABV] Section 21). The polarization Ao induces an
isomorphism Ao : A [p=] et ~ tA [p=] et. We identify TAo [p=] et and TAo [p=] et
by Ao. Then the involution a f-t h is replaced by the Rosati involution *.
Then it is clear from the previous theorem that
PAo,Ao (R) ~ Homz p(Sym 2 T Ao[p=]et, iG m (R)),

where Sym 2 T Ao[p=]et is the self tensor product T Ao[poo]et ®zp T Ao[p=]et
modulo the subspace generated by x ® y - y ® x for all x, yET Ao[p=]et.
8.2.5 Unitary Case
We fix a division algebra D with positive involution *. Assume (unr) for D

in Section 7.1.1. The center of D is either a CM field F (* inducing complex
conjugation on F) or a totally real field on which * is trivial. We fix a D-
module V with *-Hermitian alternating form ( , ) satisfying conditions (Ll-2)
in Chapter 7. Since D is a division algebra, V ~ Dr; so, we identify V with
the row vector space Dr by fixing a base of V over D. Under this expression,
C = EndD(V) is the matrix algebra Mr(tD) for the opposite algebra tD of
D (i.e., tD = D as sets but the product xy E tD of x, y E tD is defined to
be the product yx E D). The left action of b E Mr(tD) on Dr is given by
bv := v . b with matrix right multiplication v· b. By this expression, (x, y) =
TrD/IQI(xs1/) for an invertible matrix s E Mr(D) with ts* = -s, where TrD/IQi
is the reduced trace D --+ F composed with the field trace TrF/1QI : F --+ 1Ql.
On C, the involution L is given by c" = stc* S-I. Recall the groups G(R) =
{x E C ®IQI Rlxx' E RX} and U(R) = {x E G(R)lxx' = I}.
For simplicity, we suppose either F = Fo = IQl or that F is an imag-
inary quadratic field IQl[R] with discriminant ° > -0 E Z. Let c =
diag[l,O, ... ,O] E OD,p under the identification OD,p = Md(Op) in (unr) of
Section 7.1.1. For the involution * of Md(Fp), x f-t tx* = (tx)* is an Fo-algebra
automorphism of Md(Fp); so, it is inner up to (J E Gal(F/Fo) (by the theo-
rem of Skolem-Noether; [BAL] VIII.10.1). Thus by choosing the identification
OD,p = Md(Op) well, we may assume that c* = c.
We first treat the case where F is imaginary. As before, we write 0 for
the integer ring of F. Since DR = D ®lQIlR can be identified with Mq(C) with
b* = tb for complex conjugation b f-t b, S = A s E Mr(tD R) = Mqr(C) is a
Hermitian matrix. Thus U(lR) is the unitary group of S. Writing the signature
of S as (m,n), we find U(IR) ~ Um,n in Section 5.2.2. By the condition (det),

the representation of F on c;(Lie(A)) for (A,>.,i,1]) E P(R) (R E W-CL) is
mid + nc for id : 0 '--t Wand nontrivial automorphism c of F. We fix an
OD-Iattice L of V such that ( ) induces a self-duality of Lp = L @z Zp-
Lemma 8.10 Suppose (unr) and (Ll-2) in Section 7.1.1 and that D is a divi-
sion algebra with positive involution of the second kind. If we have an ordinary
triple (Ao,>:o,1]~p)) sitting over a characteristic p point of Sh(p)(G,X), then
we have either m = n or p = pp with two distinct prime ideals p and p in F.
Proof. By (unr) , we identify OD,p = Md(Op) for the integer ring 0 of F.
We write c; = diag[l, 0, ... , OJ E OD,p' Suppose we have an ordinary triple
(Ao, >:0, 1]~p)). By Hilbert's zero theorem (Nullstellensatz; [ALGJ I.1.3A), the
ordinary locus of Sh(p)(G,X) has IF-points (IF = iFp); so, we may assume that
Ao is defined over a finite field lFQ for Q = pT. Let (A, >., 1](p)) be the abelian
variety over W (W = i:p 1 (W)) corresponding to the Serre-Tate coordinate
qA = 0 (the canonical lift). Then A has complex multiplication in addition
to the multiplication by OD. We may assume that (A, >.) is defined over a
valuation ring Wo of a number field. Put Wo = ~J Wo/pJWo ~ W(lFQ) C W
for the finite field IF Q. Write K for the field of fractions of Wo. By the self-
duality via >., the locally free group scheme A[pJJ is isomorphic to the Cartier
dual of A[pJ], and Hom(A[pjJi\,vo' J.LPJ) ~ A[pJJ;wo' Taking the limit, we have
(8.18)
whose right-hand side is written as Ti'(A). From (det), in the same manner
as in the proof of Proposition 4.1, we see
c;. Lie(A/wo) ~ c;. J;,°(A) @zp Wo as W o @z O-modules, (8.19)
where J;,°(A) = ~jA[p2JO(ij). By the definition of the signature (m,n), on

c; .Lie(A/ K), the field F acts through a semi-simple representation whose
character is given by m . id + n . c.
Suppose p = p for the moment (so, p = (p) under (unr)). Then J;,°(A) gives
a maximal isotropic subspace of Tp(A) = ~ JA[pJJ (QI) under the Weil pairing.
Since Op is a valuation ring quadratic over Zp, if Op acts on a Zp-module M
of finite rank, its action on M @zp Wo has trace j(id + c) for a positive integer
j. Thus id and complex conjugation c have the same multiplicity on the Zp-
module c;. Lie(A/ K ), and we have m = n.
Now we assume that p ¥- P with p giving rise to the p-adic place of F
corresponding to ip 0 id. Then by the same reasoning as above, we have
J;,°(A) @zp K = (Tp°(A) EB Ti'(A)) @zp K ~ Lie(A/ K ), .
m = rankop c;'Tp°(A) and n = rank o l' c;·Ti' (A), where Tp°(A) = ~JA[pJJO(QI).
The existence of p-ordinary A does not impose any restriction on the signature
(m, n). 0
Remark 8.11 The converse of Lemma 8.10 should be true. Here are some
examples of how to create an ordinary couple (A,).)/w on Sh(p)(G,X) for a
fairly general class of data (D, V, (.,.)) as in Section 7.1.1.
1. Suppose that (p) = pp with P i= P in F, V = F r (row vector space) and
(x, y) = TrF/'Q(xoHty) for a diagonal matrix H E GLr(Q) and 0 E F
with J = -0. We further suppose that 0 has a positive imaginary part
under the identity embedding id of F into C. We take an ordinary ellip-
tic curve E with complex multiplication by 0 over a discrete valuation
ring W with finite residue field IF. We may assume that L = L + EEl L-
so that H is positive definite on L + and negative definite on L -. Define
A = (L+ Q90,id E) x (L- Q9o,c E), where writing E(C) = Cjioo(O), we
regard E as an O-module by sending a E 0 to multiplication by ioo(a)
on Cjioo(O), and L- Q90,c E is the tensor product with L- twisting the
action of 0 on E by c (i.e., a E 0 is sent to multiplication by ioo (c( a)) on
E(C) = Cjioo(O)). Then Tp°(A) = (TO(E) Q90p L+) EEl CP(E) Q90p L-).
Write H+ (resp. H-) for the restriction of H to L+ and L -. We have
H = diag[H+,H-J. Put IHI = diag[H+, -H-J (a positive majorant of H
in the sense of Siegel). If we identify L = (L+ Q90,id 0) EEl (L- Q90,c 0),
(£ Q9 a,l' Q9 a') := TrF/'Q((oaa'C) . (£IHI· tr)) (a, a' E 0 and £,£' E L)
coincides with (-,.) on the original L. We take a polarization on E giv-
ing an alternating form (a, a') = TrF/'Q(oaa'C). Then the tensor prod-
uct of the polarization on E and IHI on L gives the desired polariza-
tion ). which induces the given alternating form (-,.) = TrF/'Q(oH) on
T(A) = (z+Q90,id0) EEl (Z-Q90,cO) = Z. The triple (A,).,7](p))/w,
choosing a level structure 7](p) extending W suitably, gives a desired ordi-
nary couple on Sh(p) (GU(m, n), X)(W) for the signature (m, n) of H. We
can generalize this construction to the case where D is a division algebra
and V = Dr in an obvious way.
2. Suppose that D is a division algebra, V = D m EEl Dm (row vectors), and
((XI,X2), (YI,Y2)) = TrD/'Q(XltY2 - X2tyi), and L = L' EEl L' for L' = 0D
under (-, .). Define another pairing ( , ) : L' x L' ---+ Q by (x, y) =
TrD/'Q(xty*). We take an elliptic curve E;w with complex multiplica-
tion by the integer ring 0' of another imaginary quadratic field F' where
p splits (F' can be F if P splits in F). We define A = E' Q9z L'. Take the
self-dual polarization on E' and identify HI (E', Z) with Z2 so that the in-
tersection product is given by ((Xl, X2), (YI, Y2)) = XIY2-X2YI on Z2. Then
the tensor product (, )' Q9 (, ) on L = L' EEl L' = Z2 Q9z L' = HI(A,Z)
gives the original pairing (., .). We take the corresponding polarization
). on A = E' Q9z L'. Then (A,).) satisfies the desired requirements for
U(m,m).
Hereafter we assume that
(ord) sMp)(G, X) (lF p) has a point carrying an ordinary abelian scheme.
Take a closed point x E sMp)(G, X)(IF) carrying an ordinary abelian

scheme A o/lF . We determine the formal completion along x of sMp) (G, X)
as a formal subscheme of the Serre-Tate deformation space of (Ao, Ao) IIF. For
an Artinian local W-algebra R with residue field IF = Wjmw = iFp , we take
a deformation (A, /"A, A) E 13Ao,AO (R) with a level structure 'r/ : V(A (poo)) ~
V(p)(A) and i : OD <--+ End(AIR) giving rise to a point in Sh(p)(G, X)(R).
We suppose T/(p)(i/p)) = r(p) (A). Via /"A, we recover the point x given by
(AD, AD, io, 'r/6P )) = (A, A, i, 'r/(p))0 R IF of Sh(p)(G, X)(IF). Write Ag := c(A[Poo]).
Since A[pooj ~ A~ as Barsotti-Tate p-divisible groups, the abelian scheme A
as a deformation of A o = A 0R IF is determined by Ag.
First we assume that (p) = PP with p -I Pi so, we have 0 D,p = 0 D 0z Zp ~
Md(Op) EEl Md(Op). Identify V = Dr. Supposing that AIIF is ordinary, we
have Lie(A IIF ) ~ Lie(A[pjjIF)' where A[PjO is the connected component of
A[Pj. Thus we may assume that r A[poojet ~ Mdxn(OpY and r A[poojet ~
Mdxm(OpY. We write the Op-component of TpA~t = TpA3;g as 7PA~t. Then
the symmetric pairing qAIR( , ) : r Ao[poojet x r Ao[poojet --+ Gm(R) induces
a homomorphism qAIR : 7PA~t 0z p TpA~t --+ Gm(R), because we have chosen
c with c* = c, and the pairing is c-Hermitian (i.e., the involution * induces
complex conjugation c). The etale level N-structure outside p always lifts
uniquely to deformations (because A[Njl R is etale over R), and the lift i :
OD <--+ End(AIR) of io is unique if it exists, because /., : OD <--+ End(AIR)
is determined by /., : OD <--+ End(A[.eooj/R) for a prime .e -I p and hence by
T/6P ). Thus we can ignore the level structure while we study deformations of
(AD, AD, i o, T/6P )) IIF. So we consider the functor
Then the above argument combined with Theorem 8.9 shows
(8.20)
because the c-Hermitian form on (7PA~t x TpA~t) 0 (7PA~t X TpA~t) is deter-

mined by its restriction on (7PA~t x {O}) x ({O} X TpA~t).
We now assume that p is inert in Fi so, m = n. For simplicity, we assume
that D = F, V = F 2m , and ((Xl,X2), (Yl,Y2)) = Tr FIIQI (Xl t112 - x2 tIh). We
take Lp = 0;;" EEl 0;;" and write the first factor 0;;" as L1,p. Then C = M 2m (F)
and x" = JmtxJ;;"I for x E M2m(F). Then by the same computation as above
(8.21 )
and TpA~t 00 p TpA~t ~ Mm(Op), identifying L1,p with TpA~t.

The case where F = Q (so the involution is of the first kind) is similar
to the above inert case. We assume that G is of type C. Since G(Qp) ~
8.3 Vertical Control Theorem 355
GSP2n(!!J.p) (i.e., we may take s = I n and * = id by the argument above

Remark 7.4 and (7.6)), similarly to the symplectic case, we have
(8.22)
The results as above (8.20) through (8.22) show the liftability of an or-
dinary point of Sh(p)(G, X)(lF) to characteristic 0 (as used in the proof of
Theorem 7.1).
8.3 Vertical Control Theorem
We prove the VeT for projective Shimura varieties of PEL type. So we re-
turn to the notation introduced in Section 7.1. For simplicity, we give a full
exposition under the following four assumptions.
(u1) F = (Q[R] is an imaginary quadratic field with discriminant -5;
(u2) D / F i=- F is a division algebra with positive involution of the second
kind;
(u3) V = D; so, GU((Q) c ex for the opposite algebra e = tD of D;
(u4) The prime P splits into two distinct primes p and jJ in OF.
We write 0 = OF in this section.
Under (u1), we have G = GU; so, we have an advantage that any state-
ments already given exclusively for G or GU are both valid for our G = GU.
Since F /(Q is a Galois extension, the reflex field E = E( G, X) is a sub field of
F; so, OE,p = Zp by (u4). By (u3), Sh(p)(G, X) is projective over V. To have
the ordinary locus, the condition (ord) in the previous section is always in
force (which can be easily verified as in Remark 8.11 in many cases). It is not
difficult to generalize our result to more general projective Shimura varieties
of PEL type, although the endeavor certainly complicates the notation (see
[H02] for general treatment including quasi-projective Shimura varieties). At
the end of this section, we touch briefly upon the case where P is inert. By
our simplifying assumption, the reflex field for (GU, X) is either (Q or F.
Recall some of the notation introduced in Section Section 7.1 in our special
case. As in 7.1.1, let G = GU be a unitary similitude group defined with
respect to a central division algebra Dover F with a positive involution *.
Fix a prime p, and take V = D and a left OD-ideal L with L071Zp = OD071Zp.
Pick an element with s E D X with s* = -s, and define an alternating form
(x, y) = TrD/Q(xsy*) for x, Y E V, where TrD : D -+ (Q is the reduced trace
composed with TrF/Q : F -+ (Q. Then we see (bx, y) = TrD/Q(bxsy*) =
TrD/Q(xsy*b) = TrD/Q(xs(b*y)*) = (x, bOy) for x, y E V and bED.
Suppose that L satisfies (Ll-2) in Section 7.1.1 for this alternating form;
so, s E O;;,p. As before, we writeL: e -+ e for the involution on e given by
(ex, y) = (x, e"y); then, d = se* S-l. Recall
GU(Q) = G(Q) = bE cXIT!'" EFo = Q},

U(Q) = bE GU(Q)I'y!," = 1} and G1(Q) = bE U(Q)IN(!,) = 1},
where N : C -+ F is the reduced norm map. Suppose (unr) for D ((sc) for G 1
is automatic).
8.3.1 Heeke Operators on Deformation Space
In order to state the control theorem, we need to give a precise definition

of the projector ee, and we devote this subsection to proving the integrality
of suitable p-Hecke operators (which yield ee after taking a p-adic limit).
Let Oc be the order of L in C: Oc = {x E ClxL c L}. We write u(i)
for O~ n U(A'(oo)), where Oc = Oc Q9z Z. We fix an open compact subgroup
K c U(Z) such that K = Kp x K(p) with Kp 3":! GLm+n(Zp) via the projection
to the p-factor.
We consider p-ordinary test objects A = (A,A,i,Tj(p))/R over a local Ar-
tinian W-algebra R. Since the pairing qA/R E S(R) is actually determined by
its restriction to qA/ R : 7pA~t x TpA~t, we look into this restriction. We study
the Op-linear endomorphism algebra End BT (c· A[pooJ / R) of the Barsotti-Tate
group c . A[pooJ/ R. Write each endomorphism Q: as (ao" ~:) with
an E EndBT(c· A[pooJO), bn E HomBT(c· A[poort, c· A[pOOn and

dn E EndBT(c· -r;,et(A)).
If R is an algebraically closed field of characteristic p, the etale-connected

exact sequence c . Ao[pooJo Y c . Ao[pooJ - # C . Ao[pooJet is (noncanonically)
split. In any case, Q: acts on TA,p = 7pA~t E9 TpA~t diagonally via an and dn .
We regard T = G:;' as a maximal split torus of GLm(Op) x GLn(Op), which
is the automorphism group of c· A[pooJet xc· A[(pc)ooJet.
Let X/ Sh <;) be the universal abelian scheme for an open compact sub-
group K c G(A,(oo)) maximal at p. We write, for integers k and k', Pk =
c . (X/s[pkJet), P oo = Uk P k , p~, = c . (X/s[pk'Jet), and P/x, = Uk' P~,.
We apply the theory developed in Section 8.1 to each piece Pk and p~,;
so, we obtain the theory of false modular forms for GLn(Op) x GLm(Op)
(Op 3":! Op 3":! Zp). Since p is unramified in F, we have 0 Q9zp W 3":! WI for the
set of embedding I = {(J' = id, c} of F into Q. Then we consider filtrations fila
and filc of c(A[pkJet) and c(A[pk'Jet) and test objects of the following type:
A(k,k') = {A, A,i,Tj(p), fila, filc}/R. Let M = Sh~) and S = M [i], where E
is a lift of the Hasse invariant. We write Tf~k,k' / Se for the etale covering over
Sf = S Q9w Wf (Wf = W/pfW) classifying the above test objects. Similarly,
Tf,k,k' classifies (A, A, i, Tj(p), (O/pk)n X (O/pk')m 3":! c·A[pkJet xc·A[pk'Jet). The
covering Tf,oo,oo/ Sf is an etale faithful covering with Galois group canonically
containing Mp = GLm(Op) x GLn(Op), and we have Tf,oo,oo/Mp 3":! Sf.
We have an action of the expanding semi-group on filtrations of Pk for

the etale sheaf Pk as described in Section 8.1.6 and on the abelian scheme X
through isogenies (this isogeny action On abelian schemes A was not specified
when we described the action on filtration in Section 8.1.6). If we have a p-
isogeny {3 : P00 -+ P00 preserving a filtration of P00, we may assume that
the matrix form of {3 is given by {3J = (ln o- J P~J) with respect to a base
compatible with the filtration. Then the action of {3J is to give a new filtration
On Poo .
Since we cannot separate Pk and P~, (which is sitting in the single universal
abelian scheme X), we define a m +J = diag[{3J,pl m ] and consider an isogeny of
type a m +;. We can thus interpret the operator action in terms of the sextuple
(A, A, i, rj\p) , filII, file) IF as follows. Take an isogeny a : A -+ Aa of type (Ym+J
as above (inducing a On PooffiP/x,). Then we get a new filtration a(JillI,jile).
The p-isogeny is insensitive to the level K-structure (for K maximal at p),
and Aa has an induced polarization, still written as A. Then we have
fl(tJ(p) x tm(p))(A,A,i,rj(p),fillI,jil e)
=L p~(a-l )f(Aa, A, i, rj(p) , a(Jil lI , file)). (8.23)
Thus we have a GL(n) x GL(m)-Hecke operator tJ(p) x tm(P) acting on the

coherent cohomology HO(Se,'<d.~). This is actually an over-simplified version.
The exact sequence Hom(P~, G m ) Y X[pk] -» Pk may not split over Se; so,
the isogeny a can be defined only over a finite flat extension s~rn+J of Se
(which could be inseparable over an etale extension of Se). In other words,
if we replace the sum L,a p~(a-l )f(Aa, A, i, rj(p) , a(JillI' file)) in (8.23) by
the trace Tr s;rn+J I SE (PK (a- I )f(Aa, A, i, rj(p) , a(JillI' file))) as in (4.18) of Sec-
tion 4.1.10, we can relate tJ to a global Hecke operator U. (p) which is divisible
by the degree of s~rn+J over the maximal etale cover of Se under s~rn+J .
For the moment, we pretend that the over-simplified version (8.23) is
valid (and we later justify our argument). Thus for a while, our argument
is heuristic. Here we resort to the theory described in Section 8.1. 7 applied to
GL m x GL n , although the construction of the idempotent e = ec is global (so
some modification is necessary). So, under the notation used in Section 8.1.7,
we relate ec to eCLn x eCL rn .
Since U(Qp) S:' GLm+n(Qp), we embed GL(n) x GL(m) into U(Qp) by
(x, y) f-t diag[x, y]. This implies the p-isogeny whose kernel sits in A[poo]et
(resp. A[poo]et) corresponds to x (resp. y). Then write P = Pm,n for the
upper triangular parabolic subgroup of U whose Levi subgroup is given by
the image of GL(n) x GL(m). Let Nm,n be the unipotent radical of Pm,n.
Write a J E GU(Qp) whose projection to Cp = C 0F Fp is given by
diag[l m +n - J ,p' I J ] and v(aj) = p. We then have Hecke operators
where Ne is the upper unipotent subgroup of G(Zp). Thus N = Ne/Nm,n

and GL(m) x GL(n) = Pm,n/Nm,n' Since we identify U(Qlp) with GLm+n(Qlp)
by projecting down C @IQJ Qlp to the first component Cp , as a double coset, we
see (symbolically)
if j > n,
if j -:; n,
where we mean, for example, by (t]-n(P) x tm(j)")), the double coset
in GLn(Fp) x GLm(Fp) for the upper triangular unipotent subgroup N c

GLn(Op) x GLm(Op). This shows that the Hecke operator U](p) induces
accordingly as j > n or not. By computation, we get the following heuristic

multiplicity formula for the weight /Lm,n E X(T) as in (5.38):
if j > n,
(8.24)
if j -:; n.
This formula suggests us that Uj(p) is divisible by /Lm,n(a]), which we justify

later.
Since the universal deformation space of (A,.>..,i,f)(p),jilp,jilp)/lF is iso-
morphic to S = Hom(7pA~t @zp TpA~t, Gm ), as already seen in Section 8.2.5,
we can analyze the effect of an isogeny a : A IS -+ A~ 5 of type a] on S of the
universal deformation A is ' which sends
Here we have formulated the group structure of Gm mUltiplicatively (con-

trary to the additive notation so far used; so, the right-hand side becomes
q(a(x),a(y))/p if we use additive notation). We need to extract apth root as
above by the following reasoning. The coordinate q E S measures the depth
of nonsplitting of the exact sequence Hom(P~, G m ) '-+ c(X[pk]) -* P k , and
the sequence for k = 1 is split if q is a p-power. Thus the isogeny a ex-
ists over Slip. The isogeny is defined over a smaller covering S[(q 0 a)l/p] =
Spj(Os[(q 0 a)l/p]) by definition; so, sa
/S is given by S[(q 0 a)l/p]. At this
point, we are taking pth roots, and hence pure inseparability (we pretended
not to have) comes in. Then the action of the isogeny a of type a] on S only
depends on its effect on 7pA~t and TpA~t, not on the individual choice a. This
means that the covering sa
over S carrying the isogeny a only depends on
the image of a in the Levi quotient of P. Indeed, taking a base (x,), of 1PA~t
and (Yk)k of 'TpA~t so that the matrix of the isogeny is exactly a J , the effect
on Q = (Qk,l) = (log(e(xk, yL))) is given by
(8.25)
and S1 has the degree of pure inseparability given by the value in (8.24).
Here is the justification of our argument. Write the multiplicative vari-
able on S as an m x n variable matrix t = (tk,l) (so Qk,l = 109(tk,I)). The
conjugation (10 i:) r-+ a J (10 i: )
a;1 induces Qk,l r-+ p-lQk,1 for some in-
dices (k, l). We split the set of indices (k, l) into a disjoint union I U J of two
subsets so that the conjugation by a J induces Qk,l r-+ p-lQk,1 if and only if
(k, l) E I. The covering S~] is given by Spf(8s[t~:nk,l)EI). Thus a formal
function on S has expansion L€ a€t€ E W~k}] ZIUJ. Writing
for ~ E
~(I) for the I-part of the index ~, a formal function f on S~] has expansion
f = L€ a€t€(I)/pH(J), and we have
TrsQ] /s(j) = J.lm,n(a J ) 2.: a€tW)/pH(J),

€:€(I)=O mod p
because J.lm, n (a j) is the degree of the purely inseparable (wildly ramified)
covering Sa.] -» S and Tr(t~n = J.lm,n(aJ)t~f or 0 accordingly as pli or not.
Replacing the term La. pK.(a- 1)f(Aa., A, i, fj(p), a(jil a , file)) in (8.23) by the
trace TrsQ /s(PK. (a- 1)f(Aa., A, i, fj(p), a(jil a , file))), we get the p-divisibility of
the operator UJ (p) as the (heuristic) multiplicity formula (8.24) suggests. This
justifies the heuristic argument we gave (the heuristic argument is actually
valid for £-adic etale cohomology with £ =I- p as already explained).
Let Se = S Q9w We. On the universal deformation A/ s ' the sheaf c(Lie(A))
over Be is given by c(Lie(A[pi]O)). By duality, '-:!2.Se = 0Se Q9Zp A[pe]et, which
again only depends on Ao[pi]et; so, the Hecke operator UJ(p) is still divisible
by J.lm,n(aj) = [Nm,n : a;1 Nm,naJ] on '-:!2.§ for all K, > O. Thus the action of
the correspondence of characteristic 0 on HO(Soo, '-:!2.K.) is exactly a multiple by
the number in (8.24) of the operator induced by the mod p correspondence,
which is an integral operator. From this our claim is clear.
In any case, we can divide the action of UJ (p) by the number in (8.24)
keeping the integrality of the operator on '-:!2.K..
Lemma 8.12 Let the notation be as above. We have a well-defined integral
operator J.lm,n(a J )-IUJ (p) on HO(Tm,n,'-:!2.K. Q9 Sls/w).
We then define
ec = lim (U(p)t!, (8.26)

n-+oo
The divisibility of T(~) by /1(~) for /1 in (5.40) for the symplectic group
GSp(2n) in the correspondence ring of Sh(p) (GSp(2n), X) /lFp has been proven
in a different manner in [DAV] VIl.4 (and also there is another proof of
Lemma 8.12 via the q-expansion principle in this case: [H02]).
As before, write Mi = Sh~)w 0w Wi for e= 1,2, ... ,00. As for TJ(p), if

Ii 2: /1m,n, /1m,n (crJ )-lTj (p) is integral, /1m,n (crJ )-lTj (p) == /1m,n (crJ )-lUJ (p)
mod p, and we have the well-defined projector
eo = lim (T(p)t!,
n--+oo
(8.27)
where T(p) = I1;:i

n /1m,n(crj)-lTj (p).
The proof of the p-integrality of the operator T(p) in (8.27) and the fact
that ec induces an isomorphism of H~rd(M/zp,w..I<) onto H~rd(Soo/Zp,w..I<) if
Ii » /1m,n is similar to the treatment in the Hilbert modular case (see the
proofs of Theorems 4.9 and 4.37), and the argument is also analogous to the
proof of Theorem 5.3. Indeed, flTj (p) is the sum of flUJ (p) and the terms
pl«{3-1)f(A{3,"') for nonetale isogenies {3 : A --+ A{3. The abelian variety
A{3 is a quotient of A by a finite flat group scheme (so well-defined over
Wi see [ABV] Section 12). Thus f(A{3,'" ) is well-defined over W. However,
/1m,n(crj)-lpl«{3-1) == 0 mod p (by the argument proving Theorem 5.3), and
hence we get the congruence /1m,n(crj)-lTj (p) == /1m,n(cr J )-lUJ (p) mod p.
The boundedness of the rank of the topological Betti cohomology (indepen-
dent of Ii) can be proven in the same way as in the Hilbert modular case (see
[H95] Theorem 1.1 for a proof). This shows the condition (F) in Section 8.1.7
and hence the isomorphism H~rd(M/zp,w..I<) S:! H~rd(Soo/Zp,w..I<) for I i » /1m,n.
There is an alternative argument showing the p-integrality of the operator
TJ(p) on w..1< over M. Indeed, the well-definedness of Tj(p) as a correspon-
dence on Sh~) xw ShY::) can be seen in the same manner as in the proof of
Theorem 6.28 where the Siegel modular case is considered. Once we have a
correspondence well-defined, we can (in principle) show the integrality in the
case of the Shimura variety of PEL type (see [DAV] Chapters VI and VII for
more details in the Siegel modular case).
8.3.2 Statements and Proof
We keep the notation introduced in the previous section. Let N be the upper
unipotent subgroup of GLm(Op) x GLn(Op). Applying the theory expounded
in Section 8.1 to the present setting, we can define the space of p-adic modular
forms V on Soo (which is the formal completion of S along Sd. We simply
write V for V N in this subsection. Thus V = ~ i (~k,kl V/tkl) for Vl,k,k ' =
HO(Tf,k,k l , OTe.k.kl )N. We also define V = ~i (~k,kl v'ii'j"kl). We now state
the final result on vertical control in this book.
Theorem 8.13 Let the notation be as above. Let (GU, V, D, F,p = pp) be as
in (ul-4) at the beginning of this section. Identify 7Lp = Op naturally. Suppose
that M jzp = Sh~)(GU,X)jZp for an open compact subgroup K c GU(A(=»)
maximal at p is proper smooth over 7Lp and that S jZp = M [j; 1 for a lift E
of the Hasse invariant. Let T be the diagonal torus of G Lm (Op) x G Ln (Op)
and N be the upper unipotent radical of G Lm (Op) x G Ln (Op). Regard T as
a diagonal torus of the unitary group U(Qp). We say K, E X(T) is positive if
K, is dominant with respect to the upper triangular Borel subgroup of UjZp =
GL(m + n) in the sense described in Section 8.1.2. We write X+(T) for the
set of positive weights in X (T).
(1) There exists a canonical inclusion for R = 7Lp and Qp/7Lp,
f3: EB HO(Mjzp'f:d.'" 0z p R) Y V 0z p R.
"'EX+(T)
(2) Im(f3)[~] n V is dense in V;

(3) Write U(p) = fL:S;]:S;m+n Uj(p) for the standard Hecke operators at p of
level p=; that is, u] (p) = /Lm,n (Ctj) -1 U] (p) for the character /Lm,n E X (T)
as in (5.38), and define the ordinary projector eo = limn--+= U(p)n! on V.
Then H~rd(S, f:d.'" 0z p Qp/7Lp) is a p-divisible module with finite co rank,
where as defined in Section 8.1.6, H'!iTd(X, 7) is eoHq(X, 7) if X = Sm or
T:/",n and e'GHq(X, 7) if X = Mm.
(4) If K, » /Lm,n, H~rd(S,f:d.'" 0 Qp/7Lp) (resp. H~rd(S=,f:d."')) is isomorphic
to H~rd(M, f:d.'" 0 Qp/7Lp) (resp. H~rd(M,f:d."')), where K, » /Lm,n means
K, - /Lm,n is sufficiently interior of the positive Weyl chamber X+(T);
(5) Let V;rd be the Pontryagin dual 7Lp[[T(7L p)ll-module of eV (which is iso-
morphic to the 7Lp-dual of eV). Then V;rd is 7Lp[[TTll-free of finite rank,
where TT is the maximal p-profinite subgroup of T(7Lp);
(6) If K, E X+(T), V;rd 0zp[[T(Zp)1l,,,, 7Lp ~ Homzp(H~rd(S=,f:d."'),7Lp).
Proof. We need to check the conditions described in Section 8.1.7 to obtain

the result as in the theorem. The boundedness condition (F) in Section 8.1. 7
is verified in [H95], Theorem 1.1 for our unitary group U in the same manner
as in the proof of Theorem 4.9 in the Hilbert modular case (using the gen-
eralized Eichler-Shimura map in Section 5.3), because U(7L p) ~ GLm+n(7Lp).
The hypothesis (Hp1) is clear because S = Sh~)[j;] is an affine scheme for
our projective scheme Sh~). The value fIUj(p)(A,)..,i,rt!p),fil,w) is the sum
(more precisely, the trace) of the value f(A/C,)..',i','f/,(P),jil',w') for etale
subgroups C of AlP]. Since the Hasse invariant is insensitive to etale isogeny
(by its definition), the commutativity condition (C) in Section 8.1.7 holds. 0
Although we restricted ourselves to the projective case here, a similar result

can be obtained in more general settings, once we replace f:d.'" by its cuspidal
subsheaf f:d.",. Even in the Hilbert modular case, as seen in Section 4.1.7, if the
base field is not equal to Ql, we can easily find an example that the cokernel
of the inclusion eHO(S/w,~'Jw) ®w IF Y eHO(S/F,~'JF) is a nonzero finite
module for K, densely populated in X(T). Thus to obtain the result similar to
the above theorem in the nonprojective case, we need to add some assumption,
such as replacing ~" by the cuspidal sheaf~" (although for U (n, n) /IQ and
GSp(2n)/1Q it seems the result is true even for ~"). Here we state a precise
result without proof only for the quasi-split unitary group G = GU(n, n) for
an inert prime p of F (when p splits in F, the result is basically included in
Theorem 8.13 in the sense that we need to replace~" by the cuspidal sheaf~"
and V and V by their cuspidal counterparts Vcusp and V cusp ). Thus D = F,
V = F 2n = Fn E9 Fn, and ((Xl, X2), (Yl, Y2)) = TrF/IQ(Xl tY2 - X2tYl). The
lattice L can be given so that Lp = 0; E9 0;. The balanced maximal parabolic
subgroup P = {( 0: :)} of the unitary group U = U(n, n) has the reductive
part isomorphic to ResF/IQ(GL(n)) via ResF/IQ(GL(n)) 3 a r--+ diag[a, ta-lJ E
P. We consider the diagonal torus T C ResF/IQ(GL(n)) and the unipotent
radical N of the upper triangular Borel subgroup of ResF/IQ(GL(n)). Write
No for the unipotent radical of the upper triangular Borel subgroup of G; so,
N = No/Nn,n for the unipotent radical Nn,n of P. In the same way as in the
Hilbert modular case (see Section 4.1. 7), we define the space of cuspidal p-adic
modular form VCU8P := Vc;;:sp (resp. its p-divisible counterpart: VCU8P := V~sp)
for the Igusa tower over Sh<t) for an open compact subgroup K of G(AJ=))
maximal at p. In this case, the single GLn(Op) acts on the tower (in place of
GLn(Op) x GLn(Op) in the case where p splits).
Theorem 8.14 Let G /IQ be the quasi-split unitary similitude group of signa-
ture (n, n). Suppose that p is inert in the imaginary quadratic field F. Let the
notation be as in Theorem 8.13. Let M/ op be the toroidal compactification of
Sh<:)(GU,X)/op, and put S = M[1;;J for a lift E of the Hasse invariant.
(1) We have an inclusion (3 : EB"EX+(T) HO(M/op'~" ®zp R) Y V ®zp R for
R = 'lLp and Qlp/'lLp with the identity ofq-expansion: (3(j)(q) = Rcan(j)(q).
(2) Im((3cu8P)[~J n V is dense in Vcusp , where (3cusp is the restriction of (3 to
EB"EX+(T) HO(M/op'~,,).
(3) Write U(p) = TIl::oJ::on u J (p) for the standard Hecke operators at p of level
p=, where, for a J = diag[lJ,p1n-J,p21j,p1n-JJ E G(Qlp),
uJ(p) = fLn,n(aJ)-l[No('lLp)aJNo('lLp)J.
Then u J(p) is well-defined over Op if K, ~ fLn,n. Define the ordinary pro-
jector eo = limn-t= U(p)n! on the cuspidal subspace Vcusp of V. Then
H~rd(S,~" ®zp Qlp/'lLp) is a p-divisible module with finite co rank.
(4) If K, is sufficiently regular, H~rd(S,~" ®zp Qlp/'lLp) is isomorphic to a direct
summand H~rd(M,~" ®zp Qlp/'lLp) of HO(M,~" ®zp Qlp/'lLp).
(5) Let V;rd be the Pontryagin dual Op[[T('lLp)JJ-module of eo (Vcusp). Then
V;rd is Op[[rTJJ-free of finite rank, where rT is the maximal p-profinite
subgroup ofT('lLp).
8.4 Irreducibility of Igusa Towers 363
V;rd Q9Op[[T(Zp)Il,1< Op s:! Homop(H~rd(5oo'~I<)' Op)
if K, 2 0, where H~rd(5oo,f!d.I<) = ~mH~rd(5m'~I<). Similarly, if K, »

/-In,nl we have
V;rd Q9op[[T(Zp)Il,1< Op s:! Homo p (H~rd(M, f!d.1<)' Op).
The proof for the cuspidal subs heaves in the nonprojective cases is basically
the same as in the projective case except for an extra effort necessary to verify
(Hp1), because in the nonprojective case, the ordinary locus 5 m of the smooth
toroidal compactification is no longer affine. This can be done via q-expansion
at all cusps of the Igusa tower (in the same manner as in the Hilbert modular
case, Section 4.1.7). As we show in the following section, the Igusa tower T~
over a geometrically irreducible component 5'1 of 51 for general unitary groups
may not be geometrically irreducible, although each irreducible component of
T~ contains a cusp. So, the naive use of q-expansion only at one cusp (often
00) in the elliptic modular case (done in Section 4.2.9) to show the integrality
of the Hecke operator no longer works in general, but we need to use one cusp
at each component. In any case, as we have done in the split case, one may use
the Serre-Tate deformation coordinates in place of the q-expansion in order
to prove the integrality of U (p) in the general case. See [H02] for details of
the arguments in the nonprojective case.
8.4 Irreducibility of Igusa Towers

We sketch a proof of the irreducibility of the generalized Igusa tower by using
the determination of the automorphism group of the arithmetic automorphic
function field by Shimura and others (cf. Theorem 7.7). In this section, we
deal with a general eM field F not necessarily imaginary quadratic and study
general groups GU of types A and C admitting Shimura varieties of PEL type.
The method we use is classical (based on the determination of the de-
composition group of p-adic valuations of the arithmetic automorphic func-
tion field) and goes back to a work of Deuring [Du] for the elliptic modular
function field. By this result, the q-expansion principle holds for p-adic mod-
ular forms on these groups. Shimura has now completed his computation of
Fourier coefficients of Eisenstein series of quasi-split groups having a balanced
parabolic subgroup of the form {(o:)} ([EPE] 18.14 and [AAF] 16.9-11). We
should therefore be able to construct the p-adic Eisenstein measure for such
quasi-split unitary groups, although we do not cover this topic in this book.
Since we can identify the p-adic completion of the space of classical automor-
phic forms with the space of p-adic automorphic forms by q-expansion, the
measure (if constructed) is now verified to have values in the space of p-adic
automorphic forms and could be specialized at a point of the Shimura variety
to obtain certain p-adic automorphic L-functions (this point is not verified in

the construction by Panchishkin in [P] of a "naive" Eisenstein measure for
Siegel modular varieties).
In this section, W is the pullback by ip : Q '-+ Qp of the p-adic integer ring
of the maximal unramified extension of Qp in Qp; so, it is a strict Henselization
of Z(p) inside ij, and its residue field JF is an algebraic closure of JFp . The
valuation of Wand W = ~nWlpnw is written as ordp' For the reflex field
E of (GU, X), we write V for W n E and r;, for the residue field of V.
8.4.1 Irreducibility and p-Decomposition Groups
Let the notation be as in Section Section 7.1, in particular 7.1.3, and we

suppose the simply connectedness condition (SC) for G l ; so, we exclude groups
of type D. Probably Case D could be treated similarly, but due to the lack
of the strong approximation theorem for Gl(A.(oo)), the description of the
connected components should be more complicated.
By the assumption (SC), the strong approximation theorem holds for
Gl(A.(oo)), and the geometrically connected components of Sh(GU,X)/E over
E are indexed by 1r01r(ZCU) for the cocenter ZCU := GUIG l of GU (see Sec-
tion 7.2.3 for 1r01r(ZCU)). Note that the group GU satisfies (Dl-4), (sq, and
(CT) if the involution ~ of C = End D (V) is not definite (positive or negative).
Hereafter we assume that the involution ~ is indefinite; so, dim She GU, X) > O.
Under this assumption, Theorem 7.7 holds.
Let j{ be the function field of a geometrically irreducible component of
Sh(G,X)/E for the reflex field E of (G,X). The field j{ is called the arith-
metic automorphic function field for G. We show that all elements in the
decomposition group of a well-chosen p-adic valuation of the field j{ have
their p-components in a maximal parabolic subgroup P of GU (in the Siegel
modular case, P is given by the Siegel parabolic subgroup; see Theorem 6.27).
For the reductive quotient M of P, M(Zp) gives a subgroup of the automor-
phism group of the Igusa tower over the Shimura variety. If the projection
YJ: GU(A.(=)) -t 1r01r(ZCU) is nontrivial over M(Zp), we cannot expect that
the Igusa tower over an irreducible base is irreducible and has Galois group
M(Zp), because some of the components belong to different generic geometri-
cally irreducible components. We expect therefore that the Galois group of the
irreducible component of the tower is isomorphic to Ml(Zp) for Ml = MnG l .
8.4.2 Closed Immersion into the Siegel Modular Variety
Hereafter we use the symbols (G, Xc) and (GU, Xcu) exclusively for those
defined in Section 7.1.3 (particularly, (7.1)). Write (GSp(2g)/'Q' X So' SJ g USJg )
for the pair described in (C) in Example 7.6. For simplicity, we write
Shc = Sh(G,Xc), Shcu = Sh(GU,Xcu), Sh CSp (2g) = Sh(GSp(2g),X).

Note that Q(GSp(2g), X) = GSp2g(A)+ by Theorem 6.26. We write J for

the arithmetic automorphic function field QCYC(9J1) in Theorem 6.26; so,
GSp2g(A(001)/Z(Q) ~ Aut(J). Similarly we write J(pl = Q(pl(9J1(pl) for the
subfield of J giving the function field of the generic fiber of the p-integral
model of level prime-to-p. Then Aut(J(pl) = GSp2g(A(pool)/Z('L(pl)'
We write ,It (resp. ,It(p)) for the arithmetic function field of the neutral
component of Shcu (resp. Sh~0). By our assumptions (SC) and (CT), we
have the group Q(GU, Xcu) for the pair (GU, Xcu) (as in Theorem 7.7) acting
on the field ,It with constant field e = ,It n ij which is an abelian extension of the
reflex field E. In particular, T(Q(GU, Xcu)) ~ Q(GU, Xcu )/GU(JR)+Z(Q) is
a subgroup of finite index of Aut(,lt).
Since the Shimura variety Shc of PEL type carries the universal abelian
scheme Ac = (Ac, AC, TJe), by the universality of the triple A CSp (2g)/1Ql over
Sh CSp (2g), we have a unique morphism
i: ShC/ E --+ ShCSp(2g)/E for 9 = dimsheu Acu
such that i* A CSp (2g) ~ Ac. Our construction of Shc in Sections 7.1.2 and
7.1.3 tells us the finiteness and unramifiedness of this morphism (Corol-
lary 7.2).
Since hx E Xc gives rise to the complex structure on VlR = V ®1Ql JR and
the fiber A of Ac at x has expression A(C) = VlR / L as a complex torus, the
morphism i is described as hx --+ if; 0 hx at the level of the symmetric space for
an embedding if; : G /lR Y GSp(2g) /lR, where G is the subgroup of GU with
rational similitude (defined in Section 7.1.1). Recall that X6 = X6u for which
we write X+. The natural morphism ic : Shc --+ Shcu with icAcu ~ Ac
identifies the neutral component of S hc and that of S hcu. This also follows
from the fact T(Q(GU, Xcu)) = E(G, Xc) (for E(G, Xc) in [MiS] 4.2) inside
G(A(ool) * Gad(Q)+ C GU(A(ool) * GUad(Q)+ = GU(A(ool)

Zc(Q) C(lQll+ Zcu(Q) CU(IQl)+ Zcu(Q) ,
because the left-hand side of the above inclusion is the closure of the image
of G(A =)
( 1
in
CU(A,(=l)
1 . Thus Q(GU, Xcu) acts on the neutral component
Zeu(1Ql
of Shc. The embedding if; is obtained by regarding a E G(Q) as a Q-linear
automorphism if; (a) of V preserving the symplectic form (-,.) in (L2) of Sec-
tion 7.1.1; so, if; takes G(Q) into GSp(2g). Thus if; is defined over Q. In other
words, if; is induced by C = EndD(V) y EndlQl(V), and if;(g)(i(x)) = i(g(x))
for x E X+ and 9 E GU(JR)+.
Using the description of complex points Shc(q in (7.7), it is easy to
verify that i is an embedding S hc (q y S h CSp (2g) (q of complex manifolds.
Thus i : Shc --+ Sh CSp (2g) is a closed immersion ([D1] 5.4) defined over E =
E(G, Xc) = E(GU, Xcu) compatible with reciprocity laws in Theorem 7.7
(4). In particular the reflex field E and the constant field QCYc of the arithmetic
function field J for (GSp(2g), X) are linearly disjoint over k = QCYc n E; so, we
may identify T(GSP2g(A)+) with an open subgroup of Aut(E6'), where E6' =

E@k 6' (the arithmetic automorphic function field for the neutral component
of ShCSp(2g)/E).
We now choose a lattice L satisfying the conditions (Ll-2) in Section 7.l.
We choose a Z-base of L so that the matrix of the alternating form (-, -) is
given by J g • Then we may assume that
Recall that the p-integral model She!} is defined over V = W n E. The mor-
phism i induces a finite unramified morphism of the p-integral models
.. Sh(p) Sh(p)
Z• C/V -+ CSp(2g)/V'
Since ¢ brings the connected component G(JR)+ into GSP2g (JR)+ , i brings the
geometrically irreducible component (the neutral component) 9J1g) = 9J1gb
of Shg) containing the image of X+ x 1 into the neutral component 9J1(p) =
~pfM9J1ir of sh~1p(29)/Q'
Supposing the condition (ord) requiring that the neutral component
9J1g) @v JF has dense p-ordinary locus Sl, we look at the stalk Os"x at a
closed point x E Sl(lFp ) in the neutral component of Sl' The formal comple-
tion Sl,x = Sp!(Os"x) along x is the formal deformation space of an ordinary
abelian variety Ax sitting on x. By the description of the deformation space
in Sections 8.2.4 and 8.2.5 (see the following subsection for more details of
this point), we conclude that i induces an embedding of Sl,X into the formal
completion along i(x) of the Siegel modular variety. Since i is unramified over
sh~1p(29) (see Corollary 7.2) and i is generically a closed immersion, i above
is a closed immersion over V of 9J1g) into the ordinary locus of 9J1W for the
residue field JF = lFp of W, although we only need the fact that i is a closed
immersion on an open dense sub scheme of Shg) faithfully flat over V. In any
case, the function field of the geometrically irreducible component of the open
dense subscheme coincides with the one of the entire neutral component, over
E or over K" where we write K, for the residue field of V.
Pick x E Sp2g(A(oo)) with xp E GSP2g(Zp). Then the scheme automor-
phism T(X) E Aut(Shcsp (2g)) induces an automorphism of the p-integral
model sh~1p(29)' which is in turn induced by an isogeny
a(x) : A CSp (2g) -+ T(X)* A CSp (2g) such that a(x) 0 TJ(p) = TJ(p) 0 x
by the universality of A CSp (2g)' Since an isogeny preserves p-ordinarity, T(X)

leaves stable the p-ordinary locus 6 1 of the neutral component 9J1)~ C
sh~1p(29) @z(p) JF. As we have seen in the proof of Theorem 6.28, if T(X)
IS m the decomposition group Vp of the valuation v on J defined there,

T( x) preserves (over the p-ordinary locus ( 1 ) the connected component
A CSp (2g) [POOl ° C A CSp (2g) [pool. This property is equivalent to the fact that
xp is in the Siegel parabolic subgroup Pg(7!..p) for Pg = {(a :n. Since
a(x)-l (A CSp (2g) [prlo) is a locally free subgroup scheme of rank (pr)g, if a(x)
sends the fiber of the connected component ACSp(2g) [POOl ° (of A CSp (2g) [POO])
at one point in 6 1 into another fiber of A CSp (2g) [POOl °, it preserves the en-
tire A CSp (2g) [POOl ° , because a(x) is induced by a global isogeny. This fact is
important in the determination of the p-decomposition group.
Since the Igusa tower can be created as a subscheme of a self product of
copies of the etale quotient of ACI6 [Po<l for the ordinary locus ACI6 of Acu
(see (8.1)), the embedding i extends to the embedding of the Igusa tower of
G into that of GSp(2g), and ¢-l(Pg) is a maximal parabolic subgroup of G
well-defined over 7!..p which gives rise to a unique maximal parabolic subgroup
Pc GU defined over 7!..p (i.e., the normalizer of ¢-I(Pg) in GU).
By Theorem 7.5, we have sh~b @v E = ShcujGU(7!..p). From this fact,
we have a unique geometrically irreducible component 9J1c = 9J1cu (inside
ic(Shc)) of Shcu covering 9J1~) Xw J( for the quotient field J( of W, which
is the neutral component of Shcu. Then i brings 9J1cu into the neutral com-
ponent 9J1 . J
= ~ M9J1 M C Sh CSp (2g) over Q.
-
8.4.3 Description of a p-Decomposition Group

Let QJ v be the valuation ring of J with maximal ideal mv with respect to the
valuation v introduced in the proof of Theorem 6.27. Thus for any f E J,
pO< f E QJ v for a suitable a E 7!... For Siegel modular forms f defined over QCYc,
we have v(J) = Infe; ordp(a(~, f)) for the q-expansion coefficients a(~, f) at
the cusp 00. We then consider the subring QJcu of J made of meromorphic
functions that are well-defined over i(9J1cu). Thus QJ cu is the localization of
the sheaf of ideals giving the closed subscheme i(9J1cu) inside 9J1,
and QJcu is
a local ring with maximal idealD = {J E QJculfl.(9J1Gu) = a}. The elements
f E QJcu are smooth over an open dense subscheme in the image of i.
Let QJ = QJ v n QJcu and X(p) = X n J(p) for any subring X of J. Since QJ v
and QJcu are normal, respectively, over QJ}f) and QJ~b, QJ is normal over QJ(p).
The scheme 9J1;% has a unique irreducible special fiber 9J1;~ = 9J1(p) @w IF.
As we show in the following subsection via the Serre-Tate local coordinates, at
each ordinary eM point x E 9J1~b(lF), the vertical modulo p special fiber 9J1;~
and the horizontal closed subscheme 9J1~b /W of 9J1;% intersect transversally
(i.e., the intersection is a local complete intersection at all ordinary points).
Thus mv n QJ(p) = (p) and D n QJ(p) are two distinct prime ideals in QJ(p) ,
and hence we have two prime ideals p = mv n QJ and q = QJ n D of QJ. The
localization of QJ at p (resp. q) is equal to QJ v (resp. QJcu). By definition, QJ j q
is a subring of st with st = Unp-n(QJjq). Let mcu be the prime ideal (p+q)jq
in QJjq c st. Let
Dcu,v = {o- E r(g(GU, X)) I o-(mcu) = mcu}

Dv = {o- E r(GSp2g(A)+) n Aut(EJ/ E) I o-(mv) = mv}
Do. = {o- E r(GSp2g(A)+) nAut(EJ/E)1 o-(D) = D}.
Since the localization of mat p (resp. q) is equal to mv (resp. mcu), we see

that
Dv n Do. = {o- E r(GSp2g(A)+) n Aut(m/V) I o-(q) = q and o-(p) = p}.
The subgroup Dv is an open subgroup of Dp (in Theorem 6.28) of finite index,

and they are equal if E n QCYc = Q. We want to show first
Theorem 8.15 Suppose that the group GU admits the Shimura variety of
PEL-type Shg& described in (A) and (C) of Example 7.6 and that the Hasse
invariant does not vanish identically on Shg& 0w IF. Then r brings
{g E Q(GU,X)I¢(gp) E Pg(Qp) and o-(g) preserves the valuation ordp on e}

onto Dcu,v, where 0- : Aut(it) -+ Gal(e/ E) is the restriction map. In particu-
lar, r-l(Dcu,v) = ¢-l(Dv), andDcu,v is isomorphic to r(¢(g(GU, X))) n Dv
in Aut (EJ/ E).
Proof. We' prove the first assertion (because the last one follows from the
first). The function field it = mcu / D is characterized by the reciprocity law
(4) of Theorem 7.7. Indeed, r(g) on it is induced by rJ f-t rJ 0 9 for the level
structure rJ of test objects. This definition of r for the Siegel modular function
field is the same for Siegel modular test objects, which implies that
(T(¢(g)) f) 0 i =T(g) (f 0 i) for each f E mcu and 9 E Q(GU, X). (8.28)
For each special point x E X+ and the reciprocity map rx : Gal(E~b / Ex) -+
Tx(A)/Tx(Q), i(x) E X+ is a special point for 9Jl, and by definition in Sec-
tion 7.2.2, we find ri(x) = ¢ 0 r x . Then we confirm from the local reciprocity
law (rc1-2) that
C(¢(T", (17))) 1) 0 i(x) =T(T",(17)) (f 0 i)(x)
for each f E mcu and 0- E Gal(E~b / Ex). Since Tx(A) generates GU(A)
(moving around all special points x E X+), from the density of special points
in 9Jlcu, we confirm (8.28). Therefore i induces a well-defined homomorphism
¢: r(g(GU, X)) -+ Aut(JE/E).
From (8.28), we find Do. ::::J r(¢(g(GU, X))). Therefore, r(¢(g)) preserves
the prime ideal D c m cu , and for each 0- E Dv n Do., we have o-(p + q) c
p + q, which shows that 0- as an automorphism of it preserves mcu. We get
Recall the lattice L of V (we have chosen as above) satisfying the conditions
(Ll-2) in Section 7.1 for GU and GSp(2g) at the same time. We have the
maximal compact subgroup GU(Zp) = {x E GU(Qp)lxLp = Lp} of GU(Qp)
and G(Zp) = GU(Zp) n G(Qp) of G(Qp), and we may assume that G(Zp) =
¢-1(GSP2g(Zp)) (choosing a suitable Z-base of L). If a prime factor Po of P
in 0 0 = OFo splits into pp in 0, we have the Po factor of GU(Zp) isomorphic
to G Ln (Op) x 0; for a suitable n. If Po Ip is a prime factor of P in 0 0 giving
a prime p = PoO of 0, the po-factor of GU(Zp) is isomorphic to GUo(Oo,p)
(see Lemma 8.10), where GUo is the quasi-split unitary group defined over
0 0 ®z Z(p) given by
GUo(R) = {a E GL 2m (p)(R®oo O)laJm(p)ta c = v(a)Jm(p) for v(a) E RX}

(8.29)
for 0 0 ®z Z(p)-algebras R (m(p) is independent of the inert p). Similarly we

define
(8.30)
In other words, GU(Zp) is a hyperspecial maximal compact subgroup of

GU(Qp) in the sense of [Ttl].
Pick a E Vcu,v. Suppose that a = T(g) with gp E GU(Zp). Thus a pre-
serves the function field ji(p) = jiCU(Zp) of the generic fiber of the Kottwitz
model S hr;;~. Since S h~) (GU, X cu ) /wne carries a universal abelian scheme
Acu having the dense open ordinary locus Ac8, we have the isogeny Ac8 ;::,;
i* AC~P(29)' In particular, a E Dcu,v preserves the p-ordinary locus. The ex-
act sequence Ac8[Pn]o Y Ac8[Pn] --» Ac8[Pn]et is the pullback of the cor-
respon d'mg one Aord
CSp(2n) [pn]o Y Aord CSp(2n) [pn]et by z.. A s we
CSp(2n) [pn] --» Aord
have already remarked, if the action of x E GSP2g (Zp x Ai. (00)) sends one fiber
of the above exact sequence to another fiber of the same exact sequence (over
the given connected component of Sh csp (2g)), we know that xp E Pg(Zp). This
tells us that ¢(gp) E Pg(Zp) for the Siegel parabolic subgroup Pg = {( 0 : )}.
When gp tf. GU(Zp), by choosing an element 7r E Pg(Q), we find ¢(g)7r has
a p-component in GSP2g(Zp), because of the density of Pg(Q) in Pg(Qp) and
the Iwasawa decomposition GSP2g(Qp) = GSP2g(Zp) .Pg(Qp) (see [EPE] 7.2).
Still T( ¢(g)p) keeps some fibers of the etale-connected exact sequence; so, by
the same argument, we find that ¢(gp) E Pg(Qp). This combined with Theo-
rem 6.28 proves the desired result. 0
8.4.4 Irreducibility Theorem in Cases A and C
We describe the parabolic subgroup P c GU(Qp) given by ¢-l(Pg). First

suppose that we are in Case A. Recall the ordinarity condition we assumed in
addition to (unr) in Section 7.1.1:
(ord) The Hasse invariant of Acu does not vanish identically on 9J1~~/IF'
See Lemma 8.10 for an explicit description of GU satisfying this condition.

Then the signature of the Hermitian pairing ( , ) on L 00' ~ depends only
on the p-adic place Va of F induced by ip 0 rT. Let us explain this. Take the
idempotent c = diag[1, 0, ... ,0] = (6 OnO_l) E OD,p with c* = c, and regard
it as the endomorphism of the Barsotti-Tate group Acu [pool. For the prime
ideal I' of 0 corresponding to V, if we write m(p) for the generic rank of the
maximal etale quotient of c·Acu[poo], the signature is given by (m(p), m(pC))
for the automorphism c of F induced by complex conjugation. Here pC is
induced by ip 0 rTC. We split the set S of all p-adic places of F into a disjoint
union Esp U E;p U Ei so that Esp U E;p is made up of all prime factors of
p split over F o, E t is made up of the rest (inert primes), and m(p) 2:: m(pC)
for all I' E Esp. For I' E E i , m(p) = m(pC) because I' = pC (see Lemma 8.10).
We write this signature as (m(p), m(pC)) for each p-adic place I' E S. Then
n = m(p) + m(pC) is independent of I' E S. We write Po for the restriction I'
to the (totally real) subfield Fo of F fixed by c. Let E = Esp U E t • Then E
is in bijection with the set So of all p-adic places of Fo (by I' H Po). By (um)
and (ord), we may identify
for the group Uo in (8.30). Define a parabolic subgroup P c GU(Qp) by the

normalizer of
(8.31)
Thus if I' E E" the po-component of P is the upper triangular balanced

parabolic subgroup of GUo(Fo,p).
We assume that we are in Case C and that (GU, X) as in (C) of Exam-
ple 7.6 yields a Shimura variety of PEL type. Then GU(~) ~ G8P2n(~)IF.
In this case, we define P to be the Siegel parabolic subgroup Pn , because
GU(Qp) ~ TIpEs G8P2n(Fp).
We now return to the general case where GU can be of type A or C. Recall
the p-adic valuation ordp of the strict Henselization W c Q of Z(p) coming
from the embedding ip : ij Y Qp. Let V = Vcu be the discrete valuation of
.It over ordp such that the residue field of V is an algebraic extension of the
function field of an irreducible component of the Igusa tower over the ordinary
locus 8 1 of 9Jl~bll<' Recall the constant field t =.It n ij. Let
v= {rT E Aut(.It/E) I Vcu OrT = Vcu and rT E T(Q(GU,X))} ,

1) = {rT E Gal(t/E) I ordp OrT = ordp }.
Thus V is the decomposition (or monodromy) group of Vcu inside Im(T).

We can make explicit the valuation Vcu as follows. We first deal with
the unitary case. We pick a eM point x E 9Jlg'b(W), and suppose that the
abelian scheme A = Ax sitting over x is p-ordinary. Since x is a W-point, by

the universality of 9J12?ir, Ax/w is an abelian scheme.
We have identified L with zg x zg to embed G 1 into Sp(2g) by cp. Let
L~rd be the first factor Z~ (which is a totally isotropic sublattice under the
alternating form (-,.) in (Ll-2) of Section 7.1.1). Since A x / w is ordinary, we
can fix an ordinary level structure TJ~rd : L~rd(6)l1p= ~ Ax[poolo, which induces
by duality composed with the polarization TJ~t : L~t (6) (Qp/Zp) ~ Ax [poolet
for L~t = Lp / L~rd. Thus we have a lift of the point x to the Igusa tower
Too = IsomoD(L~rd(6)l1p=,AGg[PoolO), which we again denote by x. Then via
the Serre-Tate deformation theory, we computed in Section 8.2 the formal
completion of Sh K / W along x E ShK(W) as the formal deformation space
Scu/w of Ax. The result is as follows. Write -y;;et(A) (p E S) for the p-adic
Tate module of the p-power torsion points of Ax (IF). By (8.20) and (8.21),
pEE
for p-adic W-algebras R, where we have written L"j; = (Lp/ L~rd) (6)0 017
and L"j;e = (Lp/ L~rd) (6)0 0Ee. Here 017 = I1 PE E Op. Regarding Ax as
a member A 2 (x) of the universal family over the (geometrically irreducible)
Siegel modular variety 9J1;~, by (8.22), the formal completion 5 along x in
ShK'(GSp(2g), X)(W) is given by
S(R) = Hom(Sym2(7;et(A)), Gm(R)) ~ Hom(Sym2(L~t), Gm(R))

r)p
where the symbol Sym 2 indicates the symmetric square tensor over Zp
and L~t = Lp/ L~rd. The morphism i : 9J12?) --+ 9J1(p) restricted to Scu
is the embedding induced by the projection Sym2(L~t) ---» cL"j; (6)0 cL"j;e,
which associates a symmetric form iJt(x, y) = 'IjJ(x, y) + 'IjJ(y, x) E S(R) with
each OD-bilinear form 'IjJ : -y;et(A) x -y;et(A) --+ Gm(R) in Scu(R). Note
that the 0 D-linear form 'IjJ is determined by its restriction to the subspace
I1 PE E (c· -y;;et(A) xc' -y;;~t(A)); so, 'IjJ --+ iJt gives rise to the closed immersion
of Scu into S. The restriction f H flscu for f E 0 s gives rise to a surjective
ring homomorphism 7r : 0 s ---» 0Scu with Ker(7r) = D0s' which shows that
the morphism i is a closed immersion over V of the ordinary locus of Sh2?)
into that of the Siegel modular variety (as claimed in the previous subsection).
Writing 0Scu = W[[t]] for variables t = (tp"])PEE",] for 1 ~ i ~ m(p) and
1 ~ j ~ m(pC), we can make v = vcu explicit. Write R(P) for the function field
of the geometrically irreducible component 9J12?ir of Sh2?ir. Recall the subring
'2.J of the Siegel modular function field ~ and its prime ideal q (vanishing on
the image of the unitary Shimura variety in the Siegel modular variety) in
Theorem 8.15. Expanding f E (Q1jq) n jt(p) c 0Sau into I:~ a~t~, we have
a valuation veu(f) = Inf~ ordp(a~). Then Veu over jt is an extension of the
above Veu to jt(p) (whose residue field gives rise to the function field of an
irreducible component of the Igusa tower Too).
Recall the function field ~(p) of the generic fiber of the geometrically ir-
reducible component 9J1(p) of sh~1p(29)' Similarly as above, i(x) in the Igusa
tower over sh~1p(29) (W) gives a point in the irreducible component con-
taining the infinity cusp (as we have already proven the irreducibility in
this case: Theorem 6.27). Thus by Theorem 6.28, the original v defined for
~(p) in the proof of Theorem 6.27 can be rewritten in the same manner as
VeSp(2g)(f) = Inf~ ordp(c~), where we have expanded f E ~(p) into I:~ c~T~
for the variables T with Os = W[[T]]. This is because we can measure the
divisibility by p of an algebraic function on a geometrically irreducible smooth
scheme over W by the expansion at any closed point of the scheme with re-
spect to the local coordinate over W. In our case, one point is given by the
cusp 00 and another given by i(x) (since i(x) and 00 are sitting on the same
geometrically irreducible component of the special fiber). In particular, choos-
ing T suitably, Beu is a complete intersection defined by T' = 0 for a suitable
subset T' C T of variables in T. The generic transversality of 9J1~b/w and
9J1;~ inside 9J1;% claimed in the previous subsection follows from this ex-
pression. By this fact, the transversality is indeed valid in the entire ordinary
locus of 9J1;%. This also shows that, choosing the extension of Veu defined
above for jt(p) to jt compatibly with v on ~, the localization of Q1jq at meu
is the valuation ring of the extension Veu to Jt. In other words, the decom-
position group D for Veu is given by the stabilizer Deu,v of meu studied in
Theorem 8.15.
The symplectic case is similar. In this case, we have by (8.22),
Beu(R) = II Hom (Sym 2(c. Tpet(A)),Gm(R)) ,

p
where !l runs over all p-adic places of the base (totally real) field F and
the symbol Sym 2 indicates the symmetric square tensor over G p • The same
argument as above proves that the decomposition group for Veu is given by
the stabilizer of !leu in Theorem 8.15. We get for groups of type A and C:
Theorem 8.16 Suppose (unr) and (ord). Then the group D is the image
(under T) of a conjugate in Q(GU, X) under Aut(9J1eu) of
where (1 is as in Theorem 7.7 (2) and P is as in (8.31) in the unitary case

and is the Siegel parabolic subgroup in the symplectic case.
Proof. This directly follows from Theorem 6.28 and Theorem 8.15, because
the valuation we created, vcu, is given by v = VCSp(2g) on J induced from the
q-expansion by the choice of our embedding ¢ : GU --+ GSp(2g). For general
Vcu not necessarily related to Siegel modular q-expansions, its decomposition
group in Aut(J't) is the conjugate of the one specifically chosen. Since the image
of Q(G, X) in Aut(J't) is a normal subgroup of finite index, the conjugated
image of P remains in that of Q(G, X). 0
We consider the universal abelian scheme Act S over Sl = shgb [-k 1Q9w IF
for the Hasse invariant H. Then we recall the tgusa tower over Sl as follows.
We write Act[pm]/~I for the maximal etale quotient of Act [pm]. Then we
define a constant group scheme C m by
in Case A
in Case C.
The Igusa tower is given by T m / sl = Isomo (Cm/sl,c(Act[pm]et)/sJ,

where GU(Qp) ~ GSP2n(F Q91Ql Qp) in Case C.
Corollary 8.17 Let M be the reductive part of the parabolic subgroup P u in
(8.31). Define Ml = M(7l..p) n G 1 (7l.. p) (so, Ml = M(7l..p) in Case C). Then
under the hypotheses of Theorem 8.16, each geometrically connected compo-
nent T~ of the generalized Igusa tower Too over a geometrically irreducible
component S'1 of Sl has Galois group over S'1 isomorphic to Mi. In particu-
lar, Too xS I S'1 is irreducible ifGU is of type C, and if Sh(p)(GU,Xcu) is not
projective, each geometrically irreducible component of Too X SI S'1 contains at
least one cusp.
Except for the ramified primes p in F /Q in the Hilbert modular case, this
corollary includes all earlier results on mod p irreducibility under (um) as
special cases and also can be applied to projective Shimura varieties.
Proof. We take a characteristic 0 lift E of the Hasse invariant and consider

S = shgb[-,§;-]/w. We put Sm = S Q9w Wm and Soo = ~mSm, which is
independent of the choice of E. We recall that 9J1gb/w = ~s9J1s/w for
9J1s with 9J1s(C) = rs\x+, where rs = SU(Q) n S· GU(lR.) and S runs
over all open compact subgroups S c GU(A(oo)) maximal at p. We may
assume that S'1 = 9J1gb [-,§;-] Q9w IF by the existence of a smooth projective
compactification of 9J1gb/w ([Fl). Each element g E P acts on Sm, because
it preserves the ordinary locus. Suppose that GU is of type A. By (5.34),
det(f:!"a) ~ det(f:!!.O"C) on rs \X+. Note that 9J1 s / w is strongly quasi-projective
(by the existence of the projective smooth toroidal compactification; see [Fl).
Then by [NMD] Theorem 5 in Section 8.2, Pic 9Jls / w is a quasi-projective
scheme. Since f:!!.O",f:!!.O"c E Pic9Jls/w(W), we find that det(f:!!.O") ~ det(f:!!.O"C) Q9.c
over W up to .c E Picw(W) which is trivial. Thus we get det(f:!!.O")/w ~
det(~aC)IW' Since over Soo, we have det(~a)lw = det(c:T(Acu[poo]et)) ®zp

OSoo (where det(c:T(Acu[poo]et)) = 1\'J';(p) det(c:T(Acu[poo]et)) for the ideal
p of F induced by ipoIJ), we find from the faithful flatness of OSoo over 7Lp that
det(c:Acu[pn]et) ~ det(c:Acu[pCn]et) over S~ for all m and n. This implies
that for 9 E M(7Lp) to be in P, it has to belong to Ml which is given by
Ml = {{ (gp)p E TIp GLm(p)(Op)1 det(gp) = det(gpc) lip E S} in Case A,

TIpEs GLn(Op) in Case C.
In Case C, there is no restriction. The fact Ml = M(7Lp) n P can also be

checked by computing 9(GU, X) explicitly. Since Ml C P, Ml '---+ D by
Theorem 8.16, and this shows Ml = Gal(T~jS~) for the irreducible com-
ponent T:'xols f of Too whose decomposition group is D. In Case C, the Ga-
lois group is full, and hence Too x s, Sf is irreducible. In Case A, we have
7ro(T:XO) = M(7Lp)jMl for T:XO = Too Xs , Sf, because M(7Lp) acts transitively
on 7ro(T:XO). Then the ordinary locus of the minimal compactification T~ of
T:XO has to have an irreducible component containing a cusp, because T:XO is
a Galois covering of Sf whose compactification has a cusp. Since M(7Lp) acts
transitively on the irreducible components of T:XO, every component contains
a cusp. 0
By the above result, in the symplectic case, the q-expansion principle holds
as usual (as already stated for the Hilbert modular and the Siegel modular
cases). In the unitary case, to determine a mod p (or p-adic) automorphic form
by q-expansion, one needs an expansion at one cusp for each geometrically
irreducible component of the Shimura variety before the reduction modulo p.
We leave for the reader a precise formulation of this fact.
References
Books
[AAF] G. Shimura, Arithmetic2ty in the Theory of Automorphic Forms, Math-
ematical Surveys and Monographs 82, American Mathematical Society,
Providence, RI, 2000.
[AAG] S. S. Gelbart, Automorphic Forms on Adele Groups, Annals of Math.
Studies 83, Princeton University Press, Princeton, NJ, 1975.
[AAQ] M.-F. Vigneras, Anthmetique des algebres de quaternions, Lecture Notes
in Mathematics 800, Springer, New York, 1980.
[ABV] D. Mumford, Abelian Varieties, TIFR Studies in Mathematics, Oxford
University Press, New York, 1994.
[ACM] G. Shimura, Abelian Varieties w2th Complex Multiplication and Modular
Functions, Princeton University Press, Princeton, NJ, 1998.
[AGD] A. Borel and G. D. Mostow, editors, Algebraic Groups and Discontinuous
Subgroups, Proc. Symp. Pure Math. IX, American Mathematical Society,
Providence, RI, 1966.
[ALF] K. Iwasawa, Algebraic FUnctions, Translation from the 1973 Japanese
edition by Goro Kato. Translations of Mathematical Monographs, 118,
American Mathematical Society, Providence, RI, 1993.
[ALG] R. Hartshorne, Algebraic Geometry, Graduate Texts in Mathematics 52,
Springer, New York, 1977.
[AME] N. M. Katz and B. Mazur, Arithmetic Moduli of Elliptic Curves, Annals
of Math. Studies 108, Princeton University Press, Princeton, NJ, 1985.
[ARE] J.-P. Serre, Abelian £-Ad2c Representations and Elliphc Curves, Ben-
jamin, New York, 1968.
[ARG] G. Cornell and J. H. Silverman, editors, Arithmetic Geometry, Springer,
New York, 1986.
[BALl N. Bourbaki, Algebre, Hermann, Paris, 1958.
[BCM] N. Bourbaki, Algbre Commutative, Hermann, Paris, 1961-1965.
[BLI] N. Bourbaki, Groupes et Algebres de L2e, Hermann, Paris, 1972-1985.
[BNT] A. Weil, Basic Number Theory, Springer, New York, 1974.
[CBT] W. Messing, The Crystals Associated to Barsoth-Tate Groups; With Ap-
plications to Abelwn Schemes, Lecture Notes in Mathematics 264, New
York, Springer, 1972.
376 References
[CDR] A. Borel and N. Wallach, Continuous Cohomology, Discrete Subgroups,

and Representations of Reductive Groups, Annals of Math. Studies 94,
Princeton University Press, Princeton, NJ, 1980.
[CFT] E. Artin and J. Tate, Class Field Theory, Benjamin, New York, 1968.
[CGP] K. S. Brown, Cohomology of Groups, Graduate Texts in Mathematics
87, Springer, New York, 1982.
[CPS] G. Shimura, Collected Papers, I, II, III, IV, Springer, New York, 2002.
[CRT] H. Matsumura, Commutative Ring Theory, Cambridge Studies in Ad-
vanced Mathematics 8, Cambridge Univ. Press, New York, 1986.
[CSM] C.-L. Chai, Compactijication of Siegel Moduli Schemes, LMS Lecture
Note Series 107, Cambridge University Press, New York, 1985.
[DAV] G. Faltings and C.-L. Chai, Degeneration of Abelian Varieties, Springer,
New York, 1990.
[DGH] J. Tilouine, Deformations of Galois Representations and Hecke Algebras
(published for the Mehta Research Institute of Mathematics and Mathe-
matical Physics, Allahabad), Narosa Publishing House, New Delhi, 1996.
[ECH] J. S. Milne, Etale Cohomology, Princeton University Press, Princeton,
NJ, 1980.
[EGA] A. Grothendieck and J. Dieudonne, Elements de Geometrie Algebrique,
Publications IRES 4 (1960), 8 (1961), 11 (1961), 17 (1963), 20 (1964),
24 (1965), 28 (1966), 32 (1967).
[EPE] G. Shimura, Euler Products and Eisenstein Series, CBMS Regional Con-
ference Series 93, American Mathematical Society, Providence, RI, 1997.
[FAN] D. Ramakrishnan and R. J. Valenza, Fourier Analysis on Number Fields,
Graduate Texts in Mathematics 186, Springer, New York, 1999.
[FGA] A. Grothendieck, Fondements de la Geometrie Algebrique, Seminaire
Bourbaki expo no.149 (1956/57), 182 (1958/59), 190 (1959/60), 195
(1959/60),212 (1960/61), 221 (1960/61), 232 (1961/62), Benjamin, New
York, 1966.
[GAN] M. Lazard, Groupes Analytiques p-Adiques, Publications IHES 26, 1965.
[GCH] J.-P. Serre, Galo~s Cohomology, in Monographs in Mathematics,
Springer, New York, 2002.
[GCS] M. Harris and R. Taylor, The Geometry and Cohomology of Some Simple
Shimura Varieties, Annals of Math. Studies 151, Princeton University
Press, Princeton, NJ, 200l.
[GIT] D. Mumford, Geometric Invariant Theory, Ergebnisse 34, Springer, New
York, 1965.
[GME] H. Hida, Geometric Modular Forms and Elliptic Curves, World Scientific,
Singapore, 2000.
[HAL] P. J. Hilton and U. Stammback, A Course in Homological Algebra, Grad-
uate Texts in Mathematics 4, Springer, New York, 1970.
[HMS] P. Deligne, J. S. Milne, A. Ogus and K.-y. Shih, Hodge Cycles, Motives,
and Shimura Varieties, Lecture Notes in Mathematics 900, Springer,
New York, 1982.
[IAT] G. Shimura, Introduction to the Arithmet~c Theory of Automorphic Func-
tions, Princeton University Press, Princeton, NJ, and Iwanami Shoten,
Tokyo, 1971.
[ICF] L. C. Washington, Introduction to Cyclotomic Fields, Graduate Text in
Mathematics, 83, Springer, New York, 1980.
References 377
[LAP] H. Yoshida, Lectures on Absolute CM Period, Mathematical Surveys and

Monographs 106, American Mathematical Society, Providence, RI, 2003.
[LFE] H. Hida, Elementary Theory of L-Functions and Eisenstein Series,
LMSST 26, Cambridge University Press, Cambridge, England, 1993.
[MFG] H. Hida, Modular Forms and Galois Cohomology, Cambridge Studies
in Advanced Mathematics 69, Cambridge University Press, Cambridge,
England, 2000.
[MFM] T. Miyake, Modular Forms, Springer, New York-Tokyo, 1989.
[MTV] U. Jannsen, S. Kleiman and J.-P. Serre, Motives, Proc. Symp. Pure Math.
55 Part 1 and 2, American Mathematical Society, Providence, RI, 1994.
[NAZ] N. Koblitz, p-Adic Numbers, p-Adic Analysis, and Zeta Functions, Grad-
uate Text in Mathematics, 58, Springer, New York, 1977.
[NMD] S. Bosch, W. Liitkebohmert, and M. Raynaud, Neron Models, Springer,
New York, 1990.
[RAG] J. C. Jantzen, Representations of Algebraic Groups, Academic Press, Or-
lando, FL, 1987.
[RTP] G. Faltings and G. Wiistholtz et aI, Rational Points, Aspects of Mathe-
matics E6, Friedr. Vieweg & Sohn, Braunschweig, 1992.
[SFT] G. E. Breadon, Sheaf Theory, McGraw-Hill, New York, 1967.
[SGA] A. Grothendieck, Revetements Etale et Groupe Fondamental, Seminaire
de geometrie algebrique, Lecture Notes in Mathematics 224, Springer,
New York, 1971.
[SGL] H. Hida, On the Search of Genuine p-Adic Modular L-Functions for
GL(n), Mem. SMF 67, 1996.
[TEB] G. Kempf, F. Knudson, D. Mumford and B. Saint-Donat, Toroidal Em-
beddings I, Lecture Notes in Mathematics 339, Springer, New York, 1973.
[TFG] A Grothendieck and J. P. Murre, The Tame Fundamental Group of a
Formal Neighborhood of a Divisor with Normal Crossings on a Scheme,
Lecture Notes in Mathematics 208, Springer, New York, 1971.
Articles
[BeZ] 1. N. Bernstein and A. V. Zelevinsky, Induced representations of reductive
p-adic groups, I, Ann. Sci. Ec. Norm. Sup. 4th series 10 (1977), 441-472.
[BI] D. Blasius, Elliptic curves, Hilbert modular forms, and the Hodge con-
jecture, in Contributions to Automorphic Forms, Geometry, and Number
Theory, a supplement of the American Journal of Mathematics, Johns
Hopkins University Press, Baltimore, MD, 2004.
[BlR] D. Blasius and J. D. Rogawski, Motives for Hilbert modular forms, In-
ventiones Math. 114 (1993), 55-87.
[BlRl] D. Blasius and J. D. Rogawski, Zeta functions of Shimura varieties, In
"Motive": [MTV], Proc. Symp. Pure Math. 55 (1994) Part 2, 525-571.
[Bo] A. Borel, Reduction theory of arithmetic groups, Proc. Symp. Pure Math.
9 (1966), 20-25 (CEuvres II 679-684, No.74).
[BoS] A. Borel and J. P. Serre, Corners and arithmetic groups, Comment.
Math. Helv. 48 (1974), 436-491 (Borel's CEuvres III 244-299 No.98).
[Br] L. Breen, Tannakian categories, Proc. Symp. Pure Math. 55 (1994) Part
I ([MTV] I), 337-376.
[BrCDT] C. Breuil, B. Conrad, F. Diamond, and R. Taylor, On the modularity of
elliptic curves over Q or Wild 3-adic exercises, Journal AMS 14 (2001),
843-939.
378 References
[BuDST] K. Buzzard, M. Dickinson, N. Shepherd-Barron, and R. Taylor, On icosa-

hedral Artin representations, Duke Math. J. 109 (2001), 283-318.
[BuT] K. Buzzard and R. Taylor, Companion forms and weight one forms, Ann.
of Math. 149 (1999), 905-919.
[C] C.-L. Chai, Arithmetic minimal compactification of the Hilbert-
Blumenthal moduli spaces, Appendix to [Will, Ann. of Math. 131
(1990),541-554.
[Co] R. F. Coleman, p-adic Banach spaces and families of modular forms,
Inventiones Math. 127 (1997), 417-479.
[D] P. Deligne, Variete abeliennes ordinaires sur un corps fini, Inventiones
Math. 8 (1969), 238-243.
[D1] P. Deligne, Travaux de Shimura, Sem. Bourbaki, Exp. 389, Lecture Notes
in Math. 244 (1971),123-165.
[D2] P. Deligne, Varietes de Shimura: Interpretation modulaire, et techniques
de construction de modeles canoniques, Proc. Symp. Pure Math. 33.2
(1979), 247-290.
[D3] P. Deligne, A quoi servent les motifs?, Proc. Symp. Pure Math. 55 (1994)
Part I ([MTV] I), 143-161
[DM] P. Deligne and D. Mumford, The irreducibility of the space of curves of
given genus, Pub!. LH.E.S. 36 (1969), 75-109.
[DR] P. Deligne and K. A. Ribet, Values of abelian L-functions at negative
integers over totally real fields, Inventiones Math. 59 (1980), 227-286.
[DT] M. Dimitrov and J. Tilouine, Variete de Hilbert et arithmetique des
formes modulaires de Hilbert pour r 1 (c, N), to appear in Geomet'T"'tc As-
pects of Dwork's Theory, A Volume in Memory of Bernard Dwork (edited
by A. Adolphson, F. Baldassarri, P. Berthelot, N. Katz, and F. Loeser),
Walter de Gruyter, Berlin, 2004.
[Du] M. Deuring, Die Typen der Multiplikatorenringe elliptischer Funktio-
nenkorper, Abhandlungen Math. Sem. Hansischen Universitat 14 (1941),
197-272.
[F] K. Fujiwara, Arithmetic compactifications of Shimura varieties, Master's
Thesis (University of Tokyo), 1989.
[F1] K. Fujiwara, Deformation rings and Hecke algebras in totally real case,
preprint, 1999.
[GM] F. Q. Gouvea and B. Mazur, Families of modular forms, Math. Compo
58 (1992), 793-805.
[G] B. B. Gordon, Canonical models of Picard modular surfaces, CRM Pub-
lications 13 (1992), 1-29.
[Gu] F. Q. Gouvea, Non-ordinary primes: a story. Experiment. Math. 6 (1997),
195-205.
[H79] H. Hida, On abelian varieties with complex multiplication as factors of
abelian variety attached to Hilbert modular forms (Master's Thesis at
Kyoto University, 1977), Japan. J. Math. 5 (1979), 157-208.
[H81] H. Hida, On abelian varieties with complex multiplication as factors of
the Jacobians of Shimura curves (Doctor's Thesis at Kyoto University,
1980), Amer. J. Math. 103 (1981), 727-776.
[H85] H. Hida, A p-adic measure attached to the zeta functions associated with
two elliptic modular forms. I, Inventiones Math. 79 (1985), 159-195.
[H86a] H. Hida, Iwasawa modules attached to congruences of cusp forms, Ann.
Sci. Ec. Norm. Sup. 4th series 19 (1986), 231-273.
References 379
[H86b] H. Hida, Galois representations into GL2(Zp[[X]]) attached to ordinary

cusp forms, Inventiones Math. 85 (1986), 545-613.
[H88] H. Hida, On p-adic Hecke algebras for G L2 over totally real fields, Ann.
of Math. 128 (1988), 295- 384.
[H89a] H. Hida, On nearly ordinary Hecke algebras for GL(2) over totally real
fields, Adv. Studies in Pure Math. 17 (1989), 139-169.
[H89b] H. Hida, Nearly ordinary Hecke algebras and Galois representations
of several variables, Proc. JAMl Inaugural Conference, Supplement to
Amer. J. Math. (1989), 115-134.
[H91] H. Hida, On p-adic L-functions of GL(2) x GL(2) over totally real fields,
Ann. Inst. Fourier 41 (1991),311-391.
[H94] H. Hida, On the critical values of L-functions of GL(2) and GL(2) x
GL(2), Duke Math. J. 74 (1994), 431-529.
[H95] H. Hida, Control theorems of p-ordinary cohomology groups for SL(n),
Bull. SMF 123 (1995), 425-475.
[H98] H. Hida, Automorphic induction and Leopoldt type conjectures for
GL(n), Asian J. Math. 2 (1998),667-710.
[H02] H. Hida, Control theorems of coherent sheaves on Shimura varieties of
PEL-type, Journal of the Inst. of Math. Jussieu, 2002 1, 1-76 (preprint
downloadable at www.math.ucla.edu{bida).
[H03a] H. Hida, p-Adic automorphic forms on reductive groups, to ap-
pear in The Proceedings of the Automorphic Semester of the
Year 2000 at l'Institut Henri Poincare (preprint downloadable at
www.math.ucla.edu{bida).
[H03b] H. Hida, Automorphism groups of Shimura varieties of PEL type,
preprint, 2003 (preprint downloadable at www.math.ucla.edu{bida).
[HM] H. Hida and Y. Maeda, Non-abelian base-change for totally real fields,
Special Issue of Pacific J. Math. in memory of Olga Taussky Todd, 189-
217, 1997.
[HT] H. Hida and J. Tilouine, Anticyclotomic Katz p-adic L-functions and
congruence modules, Ann. Sci. Ec. Norm. Sup. 4th series 26 (1993),
189-259.
[HI] D. Hilbert, Mathematical problems, Bull. Amer. Math. Soc. 37 (2000),
407-436.
[Hz] A. Hurwitz, Uber die Entwicklungkoeffizienten der lemniskatischen Funk-
tionen, Math. Ann. 51 (1899), 196- 226 (Werke II, 342-373).
[1] J. Igusa, Kroneckerian model of fields of elliptic modular functions, Amer.
J. Math. 81 (1959),561-577.
[K] N. M. Katz, p-adic properties of modular schemes and modular forms,
Lecture notes in Math. 350 (1973), 70189.
[K1] N. M. Katz, Higher congruences between modular forms, Annals of Math.
101 (1975),332-367.
[K2] N. M. Katz, Serre-Tate local moduli, In "Surfaces Algebriques," Lecture
Notes in Math. 868 (1978), 138-202
[K3] N. M. Katz, p-adic L-functions for CM fields, Inventiones Math. 49
(1978), 199-297.
[Kn] M. Kneser, Strong approximation, in [AGD]: Proc. Symp. Pure Math.
IX (1966), 187-196.
[Kn1] M. Kneser, Hasse principle for HI of simply connected groups, in [AGD]:
Proc. Symp. Pure Math. IX (1966), 159-163.
380 References
[Ko] R. Kottwitz, Points on Shimura varieties over finite fields, J. Amer. Math.
Soc. 5 (1992), 373-444.
[L] M. J. Larsen, Arithmetic compactification of some Shimura surfaces,
CRM Publications 13 (1992), 31-45.
[LaR] R. P. Langlands and M. Rapoport, Shimuravarietiiten und Gerben, J.
Reine. Angew. Math. 378 (1987), 113-220.
[MaM] Y. Matsushima and S. Murakami, On vector bundle valued harmonic
forms on symmetric spaces, Ann. of Math. 78 (1963), 365-416.
[Mi] J. Milne, Canonical models of (mixed) Shimura varieties and automor-
phic vector bundles, Perspective Math. 10 (1990), 283-414.
[Mil] J. Milne, The points on a Shimura variety modulo a prime of good re-
duction, CRM Publications 13 (1992) ([ZFP]), 151-253
[Mi2] J. Milne, Shimura varieties and motives, Proc. Symp. Pure Math. 55
(1994) ([MTV]) Part 2, 447-523.
[MiS] J. Milne and K-y. Shih, Automorphism groups of Shimura varieties and
reciprocity laws, Amer. J. Math. 103 (1981), 911-935.
[Mt] T. Miyake, On automorphism groups of the fields of automorphic func-
tions, Ann. of Math. 95 (1972), 243-252.
[Mu] D. Mumford, Biextensions of formal groups, in Algebraic Geometry, Ox-
ford Univ. Press, Oxford (1969), 307-322.
[Mu1] D. Mumford, An analytic construction of degenerating abelian varieties
over complete rings, Compositio Math. 24 (1972), 239-272.
[P] A. A. Panchishkin, On the Siegel-Eisenstein measure and its applica-
tions, in "Proceedings of the Conference on p-adic Aspects of the Theory
of Automorphic Representations (Jerusalem, 1998)". Israel J. Math. 120
(2000), part B, 467-509.
[Ra] M. Rapoport, Compactifications de l'espace de modules de Hilbert-
Blumenthal, Compo Math. 36 (1978), 255-335.
[Ri] K. A. Ribet, P-adic interpolation via Hilbert modular forms, Proc.
Symp. Pure Math. 29 (1975), 581-592
[SBT] 1. Shepherd-Barron and R. Taylor, Mod 2 and mod 5 icosahedral repre-
sentations, J. AMS 10 (1997), 283-298.
[Sc] A. J. Scholl, Motives for modular forms, Inventiones Math. 100 (1990),
419-430.
[Se] J.-P, Serre, Geometrie algebrique et geometrie analytique, Annales Inst.
Fourier 6 (1955), 1-42 (CEuvres I 402-443, No. 32).
[Se1] J.-P. Serre, Le probleme des groups de congruence pour SL2, Ann. of
Math. 92 (1970), 487-527 (CEuvres II 487-527, No. 86).
[Se2] J.-P, Serre, Proprietes galoisiennes des points d'ordre fini des courbes
elliptiques, Inventiones Math. 15 (1972), 259-331 (CEuvres III 1-73, No.
94).
[Se3] J.-P. Serre, Formes modulaires et fonctions zeta p-adiques, Lecture notes
in Math. 350 (1973), 191-268 (CEuvres III 95-172, No. 97).
[Se4] J.-P. Serre, Quelques applications du theoreme de densite de Chebotarev,
Publ. IRES. 54 (1981), 323-401 (CEuvres III 563-641, No. 125).
[Sh] G. Shimura, On analytic families of polarized abelian varieties and auto-
morphic functions, Ann. of Math. 78 (1963), 149-192 ([63b] in [CPS] I).
[Sh1] G. Shimura, On modular correspondences for Sp(N, £::) and their con-
gruence relations, Proc. NAS 49 (1963), 824-828 ([63d] in [CPS] I).
References 381
[Sh2] G. Shimura, A reciprocity law in non-solvable extensions, J. Reine

Angew. Math. 221 (1966), 209-220 ([66a] in [CPS] I).
[Sh3] G. Shimura, Moduli and fibre system of abelian varieties, Ann. of Math.
83 (1966), 294-338 ([66b] in [CPS] I).
[Sh4] G. Shimura, Construction of class fields and zeta functions of algebraic
curves, Ann. of Math. 85 (1967),58-159 ([67b] in [CPS] II).
[Sh5] G. Shimura, On canonical models of arithmetic quotients of bounded
symmetric domains, Ann. of Math. 91 (1970), 144-222; II, 92 (1970),
528-549 ([70a-bJ in [CPS] II).
[Sh6] G. Shimura, On some arithmetic properties of modular forms of one and
several variables, Ann. of Math. 102 (1975),491-515 ([75c] in [CPS] II).
[Sh7] G. Shimura, The special values of the zeta functions associated with
cusp forms, Comm. Pure and App!. Math. 29 (1976), 783-804 ([76b] in
[CPS] II).
[Sh8] G. Shimura, On certain zeta functions attached to two Hilbert modular
forms: 1. The case of Hecke characters, II. The case of automorphic forms
on a quaternion algebra, I: Ann. of Math. 114 (1981), 127-164; II: ibid.
569-607 ([81b-c] in [CPS] III).
lSi] K.-Y. Shih, Existence of certain canonical models, Duke Math. J. 45
(1978), 63-66.
[SkW] C. M. Skinner and A.J. Wiles, Residually reducible representations and
modular forms. Pub!. IRES. 89 (1999), 5-126.
[Sp] T. A. Springer, Galois cohomology of linear algebraic groups, in [AGD],
Proc. Symp. Pure Math. IX (1966), 149-158.
[T] J. Tate, p-divisible groups, Proc. Conf. on local filds, Driebergen 1966,
Springer 1967, 158-183.
[Tl] J. Tate, Endomorphisms of abelian varieties over finite fields, Inventiones
Math. 2 (1966), 134-144
[T2] J. Tate, Conjectures on algebraic cycles in l'-adic cohomology, in [MTV]
Part 1, Proc. Symp. Pure Math. 55 (1994), 71-82.
[Ta] R. Taylor, On Galois representations associated to Hilbert modular
forms, Inventiones Math. 98 (1989), 265-280.
[Tal] R. Taylor, Icosahedral Galois representations, Pacific J. Math. (1997),
Olga Tauski-Todd Memorial Issue 337-347.
[Ta2] R. Taylor, On icosahedral Artin representations II, Amer. J. Math. 125
(2003), 549-566.
[TaW] R. Taylor and A. Wiles, Ring theoretic properties of certain Hecke alge-
bras, Ann. of Math. 141 (1995), 553-572.
[TiU] J. Tilouine and E. Urban, Several-variable p-adic families of Siegel-
Hilbert cusp eigensystems and their Galois representations, Ann. Scient.
Ec. Norm. Sup. 4th series 32 (1999), 499-574.
[Tt] J. Tits, Classification of algebraic semi-simple groups, Proc. Symp. Pure
Math. 9 (1966), 33-62.
[Ttl] J. Tits, Reductive groups over local fields, in "Automorphic forms, repre-
sentations and L-functions", Proc. Symp. Pure Math. 33 Part 1, (1979),
29-69.
[Wa] D. Wan, Dimension variation of classical and p-adic modular forms, In-
ventiones Math. 133 (1998), 449-463.
[We] A. Weil, Algebras with involutions and the classical groups, J. Indian
Math. Soc. 24 (1960), 589-623 (CEuvres II 413-447, [1960b]).
382 References
[Wi] A. Wiles, On ordinary A-adic representations associated to modular

forms, Inventiones Math. 94 (1988), 529-573.
[Wi1] A. Wiles, The Iwasawa conjecture for totally real fields, Ann. of Math.
131 (1990), 493-540.
[Wi2] A. Wiles, Modular elliptic curves and Fermat's last theorem, Ann. of
Math. 141 (1995), 443-551.
[Z] J. G. Zarhin, Isogenies of abelian varieties over fields of finite character-
istics, Math. USSR Sbornik 24 (1974), 451-461.
[Zl] J. G. Zarhin, Endomorphisms of abelian varieties over fields of finite
characteristics, Math. USSR Izvestija 9 (2) (1975), 255-260.
[Zi] T. Zink, Uber die Schlechte Reduktion einiger Shimuramanifaltigkeiten,
Compositio Math. 45 (1981), 15-107.
Symbol Index
(-, ')N, 103 J, J(p), 365

(z),180 Gk(rdN); R), 78, 86
(3), 125, 173 Gk(r1 (N); 1fp), Grd(rdN); R), 87
G"JK, E; R), hd'Jt, x; R), 177
00*, 167 S:i, 6
.:1(E,w),52 h~ord(K, E; R), 189
.:1 0('Jt) , 165 H,q,233
.:10 (pOO'Jt) , 168 H~ms(P, Pi), Isoms(P, Pi), 297
.:1 H , 235 H(X, Y)/s, W· s ' (X, Y)/s, 263
.:1 N , 235 i oo , ip, 12
Ig, Igs0(p r 'J1) , Ig K , 164
E~, 170 Jac(V), 45
ES, 170 .R, .R(p), 365
i\('Jt), Tl('Jt), r('Jt), 108 [KgK'], 165
/-LN,71 M(c, r(N)), 112
/-L'J1, 103 9J1 d ,N, 277
Tio(T), 69 9J1cu, 9J1~ir, 366, 367
Til(S,S), Til(X,X), 133,216 Pic(V), Pico (V), 45
p~, 7 Picx/s, 270
"-'.k,119 R, R(S, .:1 s ), 227, 228
"-'.;, "-'.'" E' 171 ResR' / RH/ R', 97
w(f), 246, 248 SCH/ B , B-SCH, 68
Sd'Jt) , si('Jt), S('Jt), 108
B-ALG, ALG/ B , 68 S(E), 252
A E ,354 §/1H:,305
C(1l,b,qw),110 Sh(G, X), Sh(p), 303, 309, 313, 325
c('\), 100 S",(K,E;R),177
D(m,n),239 s;:ord(K, E; R), 189
Erl ('J1), 103 Tate u ,b(q),116
i Tv(Y), Uv(Y), 174, 188
E x / s , 264
e(z), e(x), 24, 103 Vcusp(i'o('Jt)), Vcusp(ro('Jt)), 125
ET/ x ,216 Xc, Xcu, 305, 306
F~, 100 3,105
Statement Index
(E), 143 (ord), 353, 369

(Pl-3), 59, 350
(Abel), 69 (pel-4), no
(Al-4), (A4'), 277, 278 (Pc1-4),283
(Bl-2),319 (pol), 134, 312, 315
(e),345 (pos),305
(el-3),211 (QS), 219
(eT),322 (q-exp) , 119
(Dl-4),321 (rc1-2),325
(dl-2), 342 (rml-4),99
(det), 99, 268, 308 (S), 218
(E), 119, 338 (Sl-3),218
(El-3),68 (sl-3), 164
(exO-3), 170 (SAl-3), 193
(F), 345 (SB),203
(Fl-2),212 (S81-2),201
(GO-2), 76 (Se),322
(G3), 79 (se), 305
(Hpl-2),335 (ul-4),355
(Ism), 106 (unr),304
(Ll-2),304 (Vl-4),37
Subject Index
Abelian scheme, 98, 264 direct product, 210

of CM type, 4 direct sum, 210
formal,284 fibered category, 98, 99
ordinary, 101 fibered product, 210
ordinary semi-abelian, 118 fibered sum, 210
polarization of, 99, 306 final object, 210
Raynaud extension of, 284 Galois, 212
rigidity of endomorphisms, 264 initial object, 210
semi-abelian scheme, 116, 118, 265 test object, 86
Weil pairing of, 103, 350 Character,
with complex multiplication, 4 arithmetic, 176
Abel's theorem, 56 Hecke, 176
Algebra totally even, 177
double coset, 165, 227 CM field, 23, 154
Hecke, 177, 189, 191 CM point, 4, 154, 303
Iwasawa,8 Cohomology,
opposite, 351 interior, 233
Automorphic, Compactification,
factor, 6 minimal, 284
function field, 364 toroidal, 284
Automorphic form, Cone,
central character, 171 decomposition, 110
classical, 330, 335 simplicial, 110
cusp form, 119 Congruence subgroup, 6
false, 329, 335 cocompact, 6
A-adic,91 Correspondence
p-adic analytic family of, 91 algebraic, 164
Jacquet-Langlands and Shimizu, 202
true, 330, 335
Curve,
Igusa,82
Cartier duality, 71
Cases A, B, C, D, 316 Differential,
Category, holomorphic, 46
categorical quotient, 211 of the first kind, 46
388 Subject Index
Divisor, 45 unitary, 224

algebraic equivalence of, 45 Group scheme,
degree of, 45 Barsotti-Tate, 346
linear equivalence of, 45 cyclic subgroup, 126, 127
Duality, etale cyclic subgroup, 126, 127
Cartier, 71 formal,346
locally free, 71
Elliptic curve, 21, 68 p-divisible, 346
j-invariant of, 22
Tate, 78 Hasse,
Weierstrass equation of, 52 invariant, 118
Elliptic modular form, principle, 319
p-adic, 10 Hecke,
q-expansion of, 63 algebra for GL(2), 177
character, 176
Frobenius, module, 7
absolute, 80 nearly ordinary Hecke algebra, 189
element, 20 operator, 128, 165
relative, 80 polynomial, 235
Function field, universal p-nearly ordinary Hecke
arithmetic automorphic, 364 algebra, 191
smooth model of, 42 Hermitian form,
Functor, anisotropic, 237
exact functor, 212 isotropic, 238
fiber functor, 99 null space, 236
fundamental, 212 radical, 236
left exact, 212 signature, 237
Picard, 69, 270 totally isotropic, 238
representability, 72, 75, 209 Hermitian symmetric domain
representable, 209 bounded realization, 240
right exact, 212 unbounded realization, 240
Fundamental group,
of (C, F), 212 Igusa,
path,215 curve, 82
pro-object, 213 tower, 119, 295, 330
Induction,
Group, algebrogeometric, 231
Barsotti-Tate, 346 continuous, 333
cocenter, 326 regularly induced, 227
expanding semi-, 165, 228 smooth,226
formal,346 Involution,
maximal at p, 307 of the first and the second kind, 304
p-divisible, 346 positive, 304
Picard,45 Isogeny, 265
quasi-split, 218 degree of, 265
special unitary, 305 etale, 168
split, 218 prime-to-E, 138
symplectic, 224 symmetric, 99
types A, B, C, D, 316 Iwasawa function, 11
Subject Index 389
Jacobian, 45 adapted to a differential, 51

Jacquet module, 226 Path, 215
Picard,
Level structure, functor, 69, 270
r-structure, 124 group, 45
r(N), r 1 (1J1)-structure, 102, 103 scheme, 69
Fo(IJ1)-structure, 127 Pliicker coordinate, 256
Polarization, 99, 307
Modular form, c-polarization, 100
central character, 171 D-linear, 307
cusp form, 119 symmetric, 99
A-adic form, 91 Polygon, 235
p-adic analytic family, 91 Hodge, 235
Module, Newton, 235
of finite corank, 339 Positive,
Hecke,7 involution, 304
Jacquet, 226 totally, 100
Tate, 307
Modulus, Moduli, 75 Radical,
coarse, 72, 75 of Hermitian space, 236
fine, 72 Raynaud extension, 284
representability, 72, 75, 209 Reflex,
Morphism, field, 132, 306, 325
epimorphism, 209 Representation,
monomorphism, 209 admissible, 226
separable, 216 continuously induced, 333
nearly p-ordinary, 234
Nearly ordinary, regularly induced, 227
Hecke algebra, 189 schematic, 76, 332
representation, 234 smooth, 226
universal Hecke algebra, 191 universal, 331
Norm, Residue,
reduced, 199 field, 41
Null space, 236 theorem, 48
Riemann-Roch theorem, 49
Object, Rigidity,
connected, 212 of endomorphisms, 264
final, 210
fundamental pro-object, 213 Scheme,
initial, 210 abelian, 98, 264
test, 86 ample invertible sheaf on, 276
Operator, Hilbert, 261
diamond, 125, 171, 173 moduli, 75
Hecke, 128, 165 ordinary abelian, 101
Picard, 69
p-adic, quasi-projective, 261
analytic family of, 91 semi-abelian, 116, 118, 265
ring, 127 stable point of, 273
Parameter, strongly projective, 265
390 Subject Index
strongly quasi-projective, 265 residue, 48

Wei! restriction of, 97 Riemann~Roch, 49
Sheaf, vertical control, 9
ample invertible, 276 Trace,
jppj, 346 reduced, 199
of cusp forms, 119
Shimura variety, U niformizer, 42
abelian type, 326
CM point of, 4, 154, 303 Valuation ring, 37
geometrically irreducible component discrete, 37
of, 141 DVR,37
neutral component of, 142 uniformizer of, 42
reflex field, 132, 306, 325
special point of, 325 Weierstrass equation, 52
Weight, 5
Tate, dominant, 5, 231
Barsotti~, 346 double digit, 171
curve, 78 positive, 331
module, 307 regular, 231
semi-abelian scheme, 116, 118, 265 Wei!,
Test object, 86 pairing, 103, 350
Theorem, restriction, 97
Abel's, 56
of Chai~Faltings, 283, 297 Zariski topology, 44
of Faltings~Zarhin, 141 Zariski~Riemann space, 39

(Springer Monographs in Mathematics) Haruzo Hida - P-Adic Automorphic Forms On Shimura Varieties-Springer (2004)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(Springer Monographs in Mathematics) Haruzo Hida - P-Adic Automorphic Forms On Shimura Varieties-Springer (2004)

Uploaded by

Copyright:

Available Formats

Springer Monographs in Mathematics

p-Adic Automorphic Forms

Mathematics Subject Classification (2000): llF33, llF41, llF46, llG18, 14G35

Library of Congress Cataloging-in-Publication Data

ISBN 978-1-4419-1923-6 Printed on acid-free paper.

© 2004 Springer-Verlag New York, LLC

Springer- Verlag is a part of Springer Science+Business Media

Los Angeles, California,

2 Geometric Reciprocity Laws. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 17

2.5.3 Moduli of Weierstrass Type. . . . . . . . . . . . . . . . . . . . . . . .. 54

4 Hilbert Modular Varieties ................ . . . . . . . . . . . . . . . .. 97

4.1.10 Hecke Operators on Geometric Modular Forms ........ 127

5 Generalized Eichler-Shimura Map ......................... 225

6 Moduli Schemes ........................................... 251

7 Shimura Varieties .......................................... 303

8 Ordinary p-Adic Automorphic Forms ...................... 329

8.2 Deformation Theory of Serre and Tate ..................... 345

References . .................................................... 375

Symbol Index . ................................................. 383

Statement Index . .............................................. 385

Subject Index ................................................. 387

H. Hida, p -Adic Automorphic Forms on Shimura Varieties

Since this is a description of the field automorphism group Gal(Qab /Q) by

1.1 Automorphic Forms on Classical Groups

Let G m be a classical (reductive) group and G1/rQ be the derived group of G.

which can be realized on the space of homogeneous polynomials of degree n so

When G 1 = SL(2) /Q' then Co = S02(lR) c C = T(C) whose irreducible

Hd(Xr, LC(K; R)), GK(r; R) := HO(X r , i±J/R)'

and the original Eichler-Shimura map is given by sending J to the cohomology

1.2 p-Adic Interpolation of Automorphic Forms

We now interpolate these two families of cohomology groups

Item (3) is often called a vertical control theorem (abbreviated as VCT) of

One would expect that vo rd := Homw(Vo rd , W) is W[[T(Zp)]]-projective

Though f is a functorial morphism (as we said a "rule"), hereafter abusing

f(q) = f(Tate(q)oo, ¢~an, ¢J!r) = a(n, f)qn.

We may identify the Fourier expansion of f E Gk(Tl (N); q with the q-

The algebra homomorphism r;, : A ---+ W is the "evaluation" at U = UK - I!

1.3 p-Adic Automorphic L-functions

Fix embeddings ioo : Q -+ C and ip : Q '-+ Qp- Since the specialization

L(s, ).,(11:)) = l))n- S ,

Zp x SpJ(lI)(W) 3 (s, P) t-+ Lp(s, P, r) = P(L)..,r(u S - 1))

satisfies the identity: Lp(m, 11:, r) = Lh(~,~~~')) (up to a standard modi-

1.4 Galois Representations

1.5 Plan of the Book

forms. Chapter 2 is from the notes of a graduate course I gave at UCLA

Geometric Reciprocity Laws

H. Hida, p -Adic Automorphic Forms on Shimura Varieties

is minimal, the algebrogeometric automorphism group Aut(V) of the variety

2.1 Sketch of Classical Reciprocity Laws

2.1.1 Quadratic Reciprocity Law

For an integer n (p t n), the Legendre symbol (~) is defined by

if x 2 == n mod p has a solution,

(~) is a character of the finite multiplicative group (ZjpZV.

(~) = (_1)(p-l)(q-l)/4 (~) (Legendre, 1785),

has an equivalent formulation due to Euler

(~) = (p;) (Euler, 1744),

where p* = (_1)(p-l)/2 p . For a general odd positive integer n prime to p,

(q)=qq' {==} R/(q)~lFqEBlFq {==} (~*) =1,

2.1.2 Cyclotomic Version

In the mid nineteenth century, Kummer extended the quadratic reciprocity

where a is the generator of Gal(Q[lLpl/Q)/(¢q)' Thus one feature of the reci-

2.1.3 Geometric Interpretation

To further generalize the law, we need to ponder a philosophical reason why

¢ : AX /Qx = ZXQx/Qx~~ ~ ZX /Qx n (ZX x ~~) = Zx.