Algebraic Geometry and Number

Theory Summer School Galatasaray

University Istanbul 2014 1st Edition
Hussein Mourtada
Visit to download the full and correct content document:
https://textbookfull.com/product/algebraic-geometry-and-number-theory-summer-scho
ol-galatasaray-university-istanbul-2014-1st-edition-hussein-mourtada/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Dynamics and analytic number theory proceedings of the

Durham Easter School 2014 1st Edition Badziahin

https://textbookfull.com/product/dynamics-and-analytic-number-
theory-proceedings-of-the-durham-easter-school-2014-1st-edition-
badziahin/

Math 223a Algebraic Number Theory notes Alison Miller

https://textbookfull.com/product/math-223a-algebraic-number-
theory-notes-alison-miller/

Algebraic Number Theory - A Brief Introduction 1st

Edition J.S. Chahal

https://textbookfull.com/product/algebraic-number-theory-a-brief-
introduction-1st-edition-j-s-chahal/

Algorithmic and Experimental Methods in Algebra

Geometry and Number Theory Gebhard Böckle

https://textbookfull.com/product/algorithmic-and-experimental-
methods-in-algebra-geometry-and-number-theory-gebhard-bockle/
Algebraic and Differential Methods for Nonlinear
Control Theory Elements of Commutative Algebra and
Algebraic Geometry Rafael Martínez-Guerra

https://textbookfull.com/product/algebraic-and-differential-
methods-for-nonlinear-control-theory-elements-of-commutative-
algebra-and-algebraic-geometry-rafael-martinez-guerra/

A Gateway to Number Theory Applying the Power of

Algebraic Curves 1st Edition Keith Kendig

https://textbookfull.com/product/a-gateway-to-number-theory-
applying-the-power-of-algebraic-curves-1st-edition-keith-kendig/

3264 and All That A Second Course in Algebraic Geometry

Intersection Theory Combinatorial Enumerative Geometry
final draft 1st Edition David Eisenbud

https://textbookfull.com/product/3264-and-all-that-a-second-
course-in-algebraic-geometry-intersection-theory-combinatorial-
enumerative-geometry-final-draft-1st-edition-david-eisenbud/

Abstract Algebra Applications to Galois Theory

Algebraic Geometry Representation Theory and
Cryptography Second Edition Celine Carstensen-Opitz

https://textbookfull.com/product/abstract-algebra-applications-
to-galois-theory-algebraic-geometry-representation-theory-and-
cryptography-second-edition-celine-carstensen-opitz/

Competitive Math for Middle School Algebra Probability

and Number Theory 1st Edition Vinod Krishnamoorthy

https://textbookfull.com/product/competitive-math-for-middle-
school-algebra-probability-and-number-theory-1st-edition-vinod-
krishnamoorthy/
 Progress in Mathematics
321

Hussein Mourtada
Celal Cem Sarıoğlu
Christophe Soulé
Ayberk Zeytin
Editors

Algebraic Geometry
and Number Theory
Summer School, Galatasaray University,
Istanbul, 2014
Progress in Mathematics
Volume 321

Series Editors
Antoine Chambert-Loir, Université Paris-Diderot, Paris, France
Jiang-Hua Lu, The University of Hong Kong, Hong Kong SAR, China
Yuri Tschinkel, Courant Institute of Mathematical Sciences, New York, USA

More information about this series at http://www.springer.com/series/4848

Hussein Mourtada • Celal Cem Sarıoğlu
Christophe Soulé • Ayberk Zeytin
Editors

Algebraic Geometry
and Number Theory
Summer School, Galatasaray University,
Istanbul, 2014
Editors
Hussein Mourtada Celal Cem Sarıoğlu
Institut de Mathématiques de Department of Mathematics
Jussieu - Paris Rive Gauche Dokuz Eylül University
Paris, France İzmir, Turkey
Christophe Soulé Ayberk Zeytin
Institut des Hautes Études Scientifiques Department of Mathematics
Bures-sur-Yvette, France Galatasaray University
Istanbul, Turkey

ISSN 0743-1643 ISSN 2296-505X (electronic)

Progress in Mathematics
ISBN 978-3-319-47778-7 ISBN 978-3-319-47779-4 (eBook)
DOI 10.1007/978-3-319-47779-4

Library of Congress Control Number: 2017939709

Mathematics Subject Classification (2010): 11E16, 11E41, 11M41, 11R23, 11S40, 13D02, 14C17, 14E30,
14G40, 14J60, 14J70, 55N25, 55R10, 11R29, 14M17

© Springer International Publishing AG 2017

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the
material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now
known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book are
believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors
give a warranty, express or implied, with respect to the material contained herein or for any errors or
omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.

Printed on acid-free paper

This book is published under the trade name Birkhäuser, www.birkhauser-science.com

The registered company is Springer International Publishing AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Contents
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii

List of Participants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi

K. Büyükboduk
p-adic Variation in Arithmetic Geometry: A Survey . . . . . . . . . . . . . . . . . 1

I.˙ Coskun
¸
The Birational Geometry of Moduli Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . 29

O. Debarre
On the Geometry of Hypersurfaces of Low Degrees
in the Projective Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

G. Freixas i Montplet
The Riemann–Roch Theorem in Arakelov Geometry . . . . . . . . . . . . . . . . 91

L.W. Tu
Computing the Gysin Map Using Fixed Points . . . . . . . . . . . . . . . . . . . . . . 135

Z. Wojtkowiak
On -adic Galois L-functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161

A. Zeytin
Class Number Problems and Lang Conjectures . . . . . . . . . . . . . . . . . . . . . . 211
Progress in Mathematics, Vol. 321, vii–x

c 2017 Springer International Publishing

Preface

The CIMPA summer school “Algebraic Geometry and Number Theory” (AGNT)
was held at Galatasaray University, Istanbul, between 2 and 13 June 2014. Sub-
jects ranging from Arakelov geometry and Iwasawa theory to classical projective
geometry and birational geometry were covered. All these active research domains
were presented by leading experts in their respective fields. The main aim of the
lectures was to introduce these contemporary research topics to graduate students.
Accordingly the lectures have incorporated the main ideas and techniques together
with motivating examples and guiding problems in this variety of subjects.
Each individual lecturer provided a detailed outline of the lecture together
with a list of prerequisites. In addition, lecturers have pre-organized the lectures
so as to use the two-week time in most efficient way.
The young participants have shown a considerable interest to the summer
school. We had to regretfully decline more than half of the applications to our
school. Those who participated have been actively involved in the lectures and
helped providing a lively atmosphere throughout the school. The final three days
of the summer school afternoon sessions were reserved for research talks. The talks
were mostly delivered by the young participants.
We are thankful to Galatasaray University not only for the financial, but also
for the logistic support. We would also like to thank CIMPA and TÜBİTAK for
their financial support.
Our gratitude goes also to the members of the scientific committee (Olivier
Debarre (École Normale Supérieure, Paris, France), Fouad El Zein (IMJ, Paris,
France), Monique Lejeune (Université de Versailles, France), Kamal Khuri-Makdisi
(LSMS, Beirut, Lebanon), Rahim Zaare-Nahandi (University of Tehran, Iran),
Chris Peters (Université de Grenoble, Institut Fourier, France), Christophe Soulé
(IHÉS, France), Loring W. Tu (Tufts University, Medford, USA), A. Muhammed
Uludağ (Galatasaray University, Turkey)) and those of the organizing committee
(Hakan Ayral (Galatasaray University, Turkey), Merve Durmuş (Yeditepe Univer-
sity, Turkey), Hussein Mourtada (Université de Paris VII, France), İrem Portakal
(Galatasaray University, Turkey), İsmail Sağlam (Koç University, Turkey), Celal
Cem Sarıoğlu (Dokuz Eylül University, Turkey), Ayberk Zeytin (Galatasaray Uni-
versity, Turkey)) for their efforts.
The last named editor was supported by TÜBİTAK grants 113R017 and
114R073 during the summer school and during the preparation of this volume.
viii Preface

We now offer a glance of the program:

Kazım Büyükboduk (Koç University, Turkey):
Arithmetic of Abelian varieties and Iwasawa theory
İzzet Coşkun (University of Illinois, Chicago, USA):
Birational geometry of moduli spaces
Olivier Debarre (École Normale Supérieure, Paris, France):
On the geometry of subvarieties of low degree in the complex
projective space
Sabir M. Gusein-Zade (Moscow State University, Moscow, Russia):
Singular points of complex hypersurfaces
Rahim Zaare-Nahandi (University of Tehran, Iran):
A brief introduction to computational commutative algebra
Chris Peters (Université de Grenoble, Institut Fourier, France):
Lectures on motivic aspects of Hodge theory
Christophe Soulé (IHÉS, France) and Gerard Freixas i Montplet
(Institut de Mathématiques de Jussieu, Paris, France):
Arakelov theory
Loring W. Tu (Tufts University, Medford, USA):
Sheaf cohomology
Sinan Ünver (Koç University, İstanbul, Turkey):
Arakelov geometry on arithmetic surfaces

The following is an outline of the individual chapters.

Büyükboduk, in his chapter, gives an overview of Iwasawa theory and con-
siders the following issues:
• Kummer’s work on cyclotomic fields and Fermat’s last theorem: Kummer
congruences for the special values of the Riemann zeta-function,
• Iwasawa theory of cyclotomic fields,
• Iwasawa theory of other Galois representations (and motives), after Green-
berg and Mazur,
• Iwasawa theory of Galois deformations and applications (to the Taniyama-
Shimura conjecture and Fermat’s Last Theorem, the Sato-Tate, Birch and
Swinnerton-Dyer (and its p-adic variants), Artin conjectures and the Lang-
lands’ program),

Coşkun’s contribution is on birational geometry of moduli spaces. After in-

troducing the basic objects and techniques used in birational geometry and Mori
program, he demonstrates the theory through simple examples. He then studies
the birational geometry of the Hilbert scheme of points on P2 . He also discusses
 Preface ix

other moduli spaces and gives a guide to the literature in this subject. The text
contains many exercises.

Debarre gives a modern treatment of some problems of classical algebraic

geometry. In his text entitled “On the geometry of hypersurfaces of low degrees in
the projective space”, he considers the following issues:
• Projective spaces and Grassmannians, Schubert calculus,
• Projective lines on a hypersurfaces,
• Rationality of smooth cubic surfaces over algebraically closed fields. Segre
non-rationality criterion in the case of non-algebraically closed fields,
• The intermediate Jacobian, the Albanese variety, principally polarized abelian
varieties and theta divisors, Abel-Jacobi maps, conic bundles, and Prym va-
rieties. The Clemens-Griffiths proof of the fact that cubic hypersurfaces of
dimension 3 over the complex numbers are not rational,
• Cubic fourfolds.
Freixas’ contribution is a friendly introduction to Arakelov geometry. He be-
gins with the proof of the finiteness of class groups following Minkowski’s approach
but using the modern language of algebraic geometry. In this way, he motivates
the arithmetic Riemann-Roch theorem, whose formulation is the main goal of the
lectures. He then presents:
• The Grothendieck-Riemann-Roch theorem,
• The arithmetic Chow groups and characteristic classes,
• The arithmetic Riemann-Roch theorem,
• Some applications of the arithmetic Riemann-Roch theorem.
Tu, in his chapter, makes use of the localization formula in equivariant co-
homology to give a systematic method to compute the Gysin homomorphism in
the ordinary cohomology of a fiber bundle. Then he recovers as examples classical
push-forward formulas for generalized flag bundles.

The contribution by Wojtkowiak is around the idea of obtaining information

regarding certain arithmetic data by p-adic interpolation. The main aim is to
interpolate special values of the Dirichlet L-functions L(s, ψ); where ψ is a finite
order character on Z×p . This is achieved by constructing a very natural sequence of
measures. Besides constructing the measures, and relating the very first of these
measures to p-adic Dirichlet L-functions – which are the main theorems of the
paper – the author also proves various other facts about the measures.

In his chapter, Zeytin studies a very classical topic: integral binary quadratic
forms. A new approach to this subject using certain bipartite ribbon graphs called
çarks is given. The central result of the paper is the one to one correspondence
between integral points of an affine hypersurface admitting a certain group action
and indefinite binary quadratic forms of fixed discriminant. The author also studies
these hypersurfaces along the lines of Kobayashi hyperbolicity. More precisely, by a
x Preface

result of Demailly these hypersurfaces are Kobayashi hyperbolic and have finitely
many rational points as the class group is finite. These results are in accordance
with a conjecture of Lang relating arithmetic to hyperbolicity.

Hussein Mourtada (Université de Paris VII, Paris)

Celal Cem Sarıoğlu (Dokuz Eylül University, İzmir)
Christophe Soulé (IHÉS, Paris)
Ayberk Zeytin (Galatasaray University, İstanbul)
List of Participants

Nasrollah Nejad Abbas Assim Jilali

Deeba Afzal Berke Noyan Karagöz
Mirzaei Ahmad Shabbir Khurram
Yazdan Pour Ali Akbar Edwin León-Cardenal
Sadeghimanesh AmirHosein Eva Martinez Romero
Hakan Ayral Zadeh Dabbagh Mohammad
Haniya Azam Dicle Mutlu
Ayman Badawi Johannes Nagel
Sami Barhoumi Nhut Nguyen
Ilya Burenko Funda Özdemir
Kazım Büyükboduk Elif Segah Öztaş
Faryal Chaudhry Rumi Melih Pelen
İzzet Coşkun Chris Peters
Emre Coşkun İrem Portakal
Türkü Özlüm Çelik Roohi Seraji Reza
Nuray Çiçek Mohammad Reza Rezaeian
Yousef Dabborasad İsmail Sağlam
Olivier Debarre Badidja Salim
Merve Durmuş Celal Cem Sarıoğlu
Ahmad El Soufi Selcen Sayıcı
Fouad ElZein Jafari Sepehr
Altan Erdoğan Nadia Shoukat
Badr Eslam Cristophe Soulé
Mohamed Farahat Marseglia Stefano
Amir Behzad Farrahy Loring Tu
Gerard Freixas-i-Montplet Gülşen Ulucak
Özhan Genç Muhammed Uludağ
Fatma Gharbi Sinan Ünver
Sabir Gusein-Zade Uğur Ustaoğlu
Burçin Güneş Fırat Yaşar
Farid Hosseini Jafari Rahim Zaare-Nahandi
Kafian Attari Iman Mohammad Zaman Fashami
Bahmani Jaf’arloo Iman Ayberk Zeytin
Rida Irfan Aliabadi Zohreh
Canberk İrimağzı
Progress in Mathematics, Vol. 321, 1–27

c 2017 Springer International Publishing

p-adic Variation in Arithmetic Geometry:

A Survey
Kâzım Büyükboduk
Dedicated to the memory of Robert Coleman

Abstract. The main goal of this survey is to provide a general overview of the
theme of p-adic variation, both from a historical and technical view point. We
start off with Kummer’s work and Iwasawa’s treatment of cyclotomic fields,
which eventually paved the way to the modern p-adic variational techniques.
These methods have proved extremely powerful and enabled us to gain access
to some of the most important problems in mathematics, such as the Bloch-
Kato conjectures and Langlands’ Programme. We will point at a variety of
concrete applications in this vein.
Mathematics Subject Classification (2010). 11G05; 11G07; 11G40; 11R23;
14G10.
Keywords. Iwasawa theory, Modular motives.

1. Introduction
These notes grew out of a mini lecture series I gave in the CIMPA Summer School
“Algebraic Geometry and Number Theory” that took place in June 2014 in Istan-
bul. Evident from the scope of these notes, I was certainly over-ambitious for the
material I would like to introduce and I was not actually able to touch upon any
of the matters I hope to discuss in these notes. In this regard, this survey is rather
a continuation of what I have covered in these lectures.
The main goal of this survey is to provide a general overview of the theme
of p-adic variation, both from a historical and technical view point. Needless to
say, this survey is bound to be humble, due to the very large landscape that
the topic governs (from Bloch–Kato conjectures to the Langlands program) and
the ignorance of the author. However, we hope that it will still be useful as an
introductory guide to the literature for beginners.
Due to time and space restrictions, we shall not provide almost any proofs,
but rather refer the reader to the appropriate piece of the literature. We will still
discuss the idea behind various constructions whenever we are able to.
2 K. Büyükboduk

Although the technology developed to study the theme of p-adic variation in

arithmetic is referred to as Iwasawa theory in popular culture, its roots go way
back to Ernst Kummer. This is where we will start our discussion and brunch out
hopefully touching upon the following matters:
• Kummer’s work on cyclotomic fields and Fermat’s last theorem: Kummer
congruences for the special values of the Riemann zeta function.
• Iwasawa’s theory of cyclotomic fields.
• Iwasawa theory of other Galois representations (and motives), after Green-
berg and Mazur.
• Iwasawa theory of Galois deformations and applications (to the Taniyama–
Shimura conjecture and Fermat’s Last Theorem, the Sato–Tate, Birch and
Swinnerton-Dyer (and its p-adic variants), Artin conjectures and the Lang-
lands program).
Experts will already realize that it is absolutely unrealistic to discuss all these
matters as part of one short survey, although they are all tied together with the
idea of p-adic variation. We apologize in advance for the injustice we will make to
this marvelous circle of ideas.
Fix forever an odd prime p.

2. Kummer congruences
(For the basic notions we refer to in this section that the reader may be unfamiliar
with, as well as basic results we shall be quoting, we advice the reader to have a
look at [Jan96, Lan94, Was97].)
Kummer’s attempt on Fermat’s Last Theorem did not result its complete1
resolution, its impact in arithmetic geometry has been immense. His starting point
was to consider the Fermat equation
Xp + Y p = Zp
(where XY Z = 0 and X, Y, Z are integers) as an equality that takes place in the
cyclotomic field Q(e2πi/p ), by rewriting the same equation in the form
X p = (Z − Y ) · (Z − e2πi/p Y ) · · · (Z − e2(p−1)πi/p Y ) . (2.1)
Note that each factor Z − e Y (k = 0, 1, . . . , p − 1) is an element of the ring2
2kπi/p

Z[e 2πi/p
]. One than tries to utilise the following rather elementary idea: If X, Y, Z
are all sufficiently reduced (and massaged, in a sense which we will not make
precise), then all the factors above will be pairwise coprime. Their product being
a pth power will force (by unique factorisation into primes in the ring Z[e2πi/p ]?)
1 In a certain sense of the word it did eventually succeed, as Kummer’s work paved the way for

the idea of p-adic variation, by which Andrew Wiles’ [Wil95, TW95] work was certainly heavily
influenced.
2 This ring happens to be ring of integers of the cyclotomic field Q(e2πi/p ), see [Lan94, §IV.1,

Theorem 3]. Exactly for this reason it is a Dedekind ring [Lan94, ] and possesses all the nice
properties we will be using in this section.
 p-adic Variation in Arithmetic Geometry: A Survey 3

each of these factors to be pth powers. It is then not so hard to reach a contradiction
at this point, and deducing Fermat’s Last Theorem.
We all very well know that something has to go wrong in the discussion
above. This is also secretly indicated there: For p ≥ 23, the rings Z[e2πi/p ] will
never be Unique Factorization Domains [Uch71], i.e., there never will be unique
factorization into primes in these rings. This is the flaw in the argument above
and Kummer’s contributions start with his attempt to mend this problem.
Kummer first defined what he called ideal numbers (called simply ideals of a
ring nowadays, after Dedekind) and proved the following theorem.
Theorem 2.1 (Kummer). Suppose K is a number field and OK is its ring of inte-
gers. The ring OK is a Dedekind ring. In particular, every ideal I ≤ OK may be
written uniquely (up to ordering) as a product of (finite number of) prime ideals.
Definition 2.2. Given a number field L, we let JK denote the group of fractional
ideals of K and PK ≤ JK the subgroup of principal fractional ideals. The quotient
ClK := JK /PK is called the ideal class group of K.
Remark 2.3. We will not go as far as providing the definitions of the concepts that
appear in Definition 2.2, but will rather refer the reader to the basic references we
have given at the start of this section. However, let us observe that the collection
IK of integral ideals of OK form a semigroup. It turns out that the collection of
fractional ideals turns out to be the group of fractions of this semigroup, and as a
matter of fact, this fact is equivalent to Kummer’s theorem above.
We may again attempt to run the argument above after writing the equation
(2.1) as an equality taking place in JK :
Xp = Z − Y  · Z − e2πi/p Y  · · · Z − e2(p−1)πi/p Y 
where A stands for the ideal generated by A. Theorem 2.1 now shows that each
principal ideal Z − e2kπi/p Y  = Apk is a pth power of an ideal Ak ≤ Z[e2πi/p ]. In
order to be able to go back to the realm of elementary number theory, one would
hope in the first place that each Ak = αk  is principal. In fact if this was really
so, we would have had
Z − e2kπi/p Y = αpk · uk , k = 0, 1, . . . , p − 1 (2.2)
2πi/p ×
for some uk ∈ Z[e ] , i.e., each factor Z − e 2kπi/p
Y would be a pth power
up to a unit in the ring Z[e2πi/p ]. This is a statement close enough to what we
erroneously attempted to work out above and it turns out that (2.2) does still
yield a contradiction which proves FLT.
The question that remains is therefore the following: Is there a way to decide
whether or not the ideals Ak are in fact principal, given the fact that their pth
powers are principal? Note that the first obstruction for an affirmative answer to
this question is already identified as part of Definition 2.2 above: The ideal class
group ClZ[e2πi/p ] , whose non-triviality means that the ring Z[e2πi/p ] possesses non-
principal ideals. The following is one of the main theorems of classical number
theory.
4 K. Büyükboduk

Theorem 2.4. For a number field K, the ideal class group ClK is always finite.
The size of this group is traditionally denoted by hK and called the class
number of K. We now have an element [Ak ] ∈ ClZ[e2πi/p ] (the class of ideals that
contains Ak ) and we are hoping that for some reason this class should better be the
trivial class (i.e., the class that contains only principal ideals). We are given that
[Apk ] = [Ak ]p is the trivial class in the finite group ClZ[e2πi/p ] . This immediately
yields the following:
Proposition 2.5. If p does not divide the order hZ[e2πi/p ] of the ideal class group of
the pth cyclotomic field, then each of the ideal classes [Ak ] are trivial. In turn, all
ideals Ak are principal and Fermat’s Last Theorem for exponent p holds true.
This motivates the following definition:
Definition 2.6. A prime number p is called regular if p does not divide hZ[e2πi/p ] ,
and called irregular otherwise.
If we wish to draw a concrete conclusion towards FLT from Proposition 2.5,
we would need a procedure to decide when a prime p is regular. That is exactly
where magic happens and Kummer proves the following magnificent result (Theo-
rem 2.8 below). Before stating this theorem, we first need to introduce completely
new object, from a far apart domain (at least at first sight) of Mathematics, which
surprisingly will 
be relate to the discussion above.
∞
Let ζ(s) = n=1 1/ns denote the Riemann zeta function. This is a complex-
valued function analytic (and non-vanishing) on the half-plane Re(s) > 1. Riemann
proved that it admits an analytic continuation to the whole complex plane (except
for a simple pole at s = 1) and verifies a functional equation3 relating the values
ζ(s) and ζ(1 − s). Since the days of Riemann, ζ(s) (and its analogues in other
contexts, which we shall also briefly talk about later on in Section 4) has attracted
a lot of attention and analysing ζ (or its ‘cousins’) to extract various arithmetic
information has become one of the central themes in number theory. Kummer’s
theorem we are about to present is a result also along these lines. Before we do
so, the following proposition (due to Euler, but formally proved by Riemann) is
a strong indication that the Riemann ζ-function, albeit being a purely analytic
object, might have reflections on the algebraic world:
Proposition 2.7. The values of ζ(s) at negative odd integers are all (non-zero)
rational numbers; whereas its values at a negative even integer are always zero.
It is not hard to compute these values explicitly, either: ζ(−1) = −1/12,
ζ(−3) = 1/120, ζ(−5) = −1/252, ζ(−7) = 1/240, ζ(−9) = −1/132, ζ(−11) =
691/32760, . . .
We are now ready to state Kummer’s results:

3 This functional equation had already been observed by Euler, before Riemann had formally

established the fundamentals of complex analysis.

 p-adic Variation in Arithmetic Geometry: A Survey 5

Theorem 2.8 (Kummer).

(i) The prime p is irregular iff p divides the numerator of ζ(r) for some negative
odd integer r.
(ii) For a positive even integer m, the prime p divides the denominator of ζ(1−m)
if and only if m ≡ 0 mod (p − 1). Furthermore, for two positive even integers
m ≡ n ≡ 0 mod (p − 1) we have ζ(1 − m) ≡ ζ(1 − n) mod p.
(iii) More generally, if m and n are positive integers with m ≡ n mod (p − 1)pα
and n ≡ 0 mod (p − 1) we have
(1 − pm−1 )ζ(1 − m) ≡ (1 − pn−1 )ζ(1 − n) mod pα+1 .
Using Kummer’s theorem one can check very fast that all primes less than 37
are all regular and therefore FLT with those exponents hold true. However, 691 is
visibly irregular, as a matter of fact 37, 59, 67, 101, 103, 131, 149 and 157 all are.
The sad news (at least from Kummer’s perspective it perhaps was) is that
we know that there are infinitely many irregular primes but we still do not know
till this day whether this is also the case for regular primes or not!

3. Iwasawa’s theory of cyclotomic fields

The ideal class group as we know it today was in fact studied before the notion
of an ideal was invented: These groups already appeared in Gauss’ treatment of
binary quadratic forms. The equivalence classes of binary quadratic forms with
a given discriminant D (where D is a square-free integer) form an abelian group
under Gauss’ composition law for binary quadratic forms, and this group √ turns
out to be closely related to the ideal class group of the number field Q( D).
√More
precisely, this group is exactly isomorphic to the ideal class group of√ Q( D) if
D < 0 and it is isomorphic to the so-called narrow-class group of Q( D), which
may have half the size of the original ideal class group.
Conjecture
√ 3.1 (Gauss). Let hD denote the size of the ideal class group of the field
Q( D).
G1 lim hD = ∞ .
D→−∞
G2 hD = 1 with D < 0 iff D = −3, −4, −7, −8, −11, −19, −43, −67, −163.
G3 hD = 1 for infinitely many D > 0.
• The first of Gauss’ conjectures is a theorem of Mordell, Deuring and Heil-
bronn [Deu34, Hei34], later Goldfeld [Gol76] proved an effective version of
G1 granted the Gross–Zagier theorem [GZ86]4 : Given any  > 0 there exists
an effectively computable constant c > 0 such that
hD > c(log |D|)1− .
4 This theorem is about another type of zeta function that we shall introduce later. This zeta

function is known as the Hasse–Weil zeta function of an elliptic curve. The celebrated theorem
of Gross–Zagier computes the first derivative of this zeta function at the value s = 1 in terms of
the height of a Heegner point.
6 K. Büyükboduk

• The second conjecture is known as Gauss’ class number one problem and it
was solved by Heegner, Baker and Stark [Hee52, Bak67, Sta67].
• The conjecture G3 is wide open till this day.
If we were to sloganize Gauss’ conjectures in one sentence, it would be the state-
ment that the class numbers of quadratic fields behave rather erratically in families
of extensions of fixed degree5 . Iwasawa on the other hand observed that situation
changes dramatically if one considers a certain tower extensions, the behavior of
the (p-parts of) class numbers along such families may be characterized only in
terms of the tower one considers. The main point we would like to emphasize at
this point is that Iwasawa’s conclusion (Theorem 3.2 below) is due to the fact that
the (pro-p parts of) ideal class groups along this tower in fact interpolate p-adic
analytically and Iwasawa theory aims to study this sort of p-adic analytic variation
in arithmetic families.
To state Iwasawa’s theorem let us introduce the tower we shall be interested
in. The main reference for this portion of our survey is [Was97, Chapters 7 and
13]. Let K be any number field and let K∞ be any Zp extension of K. That
is to say, the Galois group of K∞ /K is an infinite Galois extension with group
isomorphic to Zp . Every number field has at least one Zp extension. This we
can see using Class Field Theory6 but also more directly: Consider the extension
Q(μp∞ ) := ∪∞ n=1 Q(μpn ) obtained by adjoining the rational field all p-power roots
of unity. Observe that
Gal(Q(μp∞ )/Q) = lim Gal(Q(μpn /Q)) ∼
= lim(Z/pn Z)× ∼
= Z×
p.
←− ←−
n n

We have a decomposition
Z× ∼ ×
p = (Z/pZ) × (1 + pZp )
∞

(via the map ai pi → a0 and the Teichmüller character ω to split this surjection)
i=1
which yields the decomposition
Gal(Q(μp∞ )/Q) ∼
= Δ × Γ.
Here Δ = Gal(Q(μp )/Q) ∼ = (Z/pZ)× and Γ = Gal(Q(μp∞ )Δ /Q) ∼ = 1 + pZp ∼
= Zp ,
where Q(μp∞ ) is the fixed field of Δ (viewed as a subgroup of Gal(Q(μp∞ )/Q))
Δ

and the isomorphism 1+pZp ∼ = Zp is obtained via the p-adic logarithm. The exten-
sion Qcyc := Q(μp∞ )Δ is called the cyclotomic Zp -extension7 of Q. The following

5 We shall call such families horizontal families.

6 Class Field Theory in fact shows that there are r2 (K) + 1 + δ(K) independent Zp -extensions of
K, where δ(K) is Leopoldt defect and r2 (K) is the number of (half of) the complex embeddings
of K. See [Was97, Chapter 13] for details.
7 It is in fact the unique Z -extension of Q. Note in this case r (Q) = 0 and since the Leopoldt
p 2
conjecture is known for Q, δ(Q) = 0 as well.
 p-adic Variation in Arithmetic Geometry: A Survey 7

diagram summarizes the discussion in the paragraph above:

Q(μp∞ )
z HH
zz HH Δ
z HH
z HH
zzz H
zz
z Q∞
zzz 
z 
zzz 
zz 
Q(μpn ) Z× =Δ×Γ 
p

n−1
 
Z/p Z  Γ∼ = Zp
 
Q(μp ) Q
QQQ 
QQQ  
QQQ
Δ∼
Q
=(Z/pZ)× QQQQ 
Q 
Q
Now given a number field K, we set Kcyc := KQcyc , the cyclotomic Zp -extension
of K. This construction validates our claim that every number field possesses at
least one Zp -extension.
Given an arbitrary Zp -extension K∞ /K as above, let Kn /K denote the
unique subfield with Gal(Kn /K) ∼ = Z/pn Z. Existence and uniqueness of Kn fol-
low from the fact that the topologically cyclic group Zp has a unique (closed)
subgroup of index pn . We call Kn the nth layer in this extension and we set8
Γ := Gal(K∞ /K) and Γn := Γ/Γp ∼
n
= Gal(Kn /K). Here is our diagram that
summarizes this picture.
K∞ P
PPP pn+1 n+1
PΓPP ∼ =p Zp
PPP
PP
Kn+1
n n+1
Γ Γp /Γp ∼
=Z/pZ

Kn
nnnnn
nn
nnnnΓnn =Γ/Γpn ∼
=Z/pn Z
nn
K
We are now ready to state Iwasawa’s theorem. Let pen denote the exact power
of p dividing the class number hKn of the nth layer of the tower K∞ /K.

8 Iwasawa [Iwa69, Iwa71, Iwa73],when he initiated the arithmetic study of towers of extensions of
the sort K∞ , called these Γ-extensions. This is the reason for our tendency to name the relevant
Galois groups Γ.
8 K. Büyükboduk

Theorem 3.2 (Iwasawa). There exists integers λ = λ(K∞ /K), μ = μ(K∞ /K) ≥ 0
and ν = ν(K∞ /K), all independent of n and an integer n0 such that
en = λn + μpn + ν
for every n ≥ n0 .
The proof of this theorem relies on the following observation. Let An denote
the pro-p part of the ideal class group of Kn . Then An is naturally a Zp [Γn ]-
module. Furthermore, the Zp [Γn ]-action on An is compatible in that, it induces a
continuous action of the ring Λ := lim Zp [Γn ] on A∞ := limn An . Serre observed
←− ←−
that the ring Λ is (non-canonically) isomorphic to Zp [[X]], formal power series ring
in one variable with coefficients in Zp . This ring has many fantastic properties: It
is a complete local Noetherian ring, and it is regular therefore a UFD. Moreover,
the module theory over Λ is relatively easy and there is a structure theorem for
finitely generated Λ-modules, resembling the structure theorem for modules over
PIDs. All these facts assembled together shows that A∞ is a finitely generated
torsion Λ-module and there is a Λ-module map
 
A∞ −→ Λ/f Λ × Λ/pμ Λ (3.1)
(where the product is over a finite number of polynomials f ∈ Λ of positive degree,
such that the quotients Λ/f Λ are free Zp -modules) with finite kernel and cokernel.
This fact together with Iwasawa’s control theorem that the maps (A∞ )Γpn → An
have finite cokernels proves the theorem (with μ as in (3.1) and λ the sum of the
degrees of the polynomials f that appear in (3.1)). Here (A∞ )Γpn stands for the
n
largest quotient9 of A∞ on the which Γp acts trivially.
Remark 3.3. It is a folklore conjecture that the μ-invariant μ(Kcyc /K) vanishes for
any number field K. Note that this conjecture implies via (3.1) that the interpolated
class group A∞ is a finitely generated Zp -module. Washington and Ferrero proved
in [FW79] that this is the case for abelian extension of Q. The general case remains
widely open.
3.1. p-adic zeta functions and Iwasawa’s main conjectures
We recall the significant role that the special values of the zeta function played
in Kummer’s results1 Theorem 2.8. It turns out that Iwasawa’s theory has also
a counterpart in the zeta-world. The first hint towards the fact that the zeta-
values interpolate into a p-adic analytic function was secretly spelled out as part
9 It n
is helpful to notice that this quotient is given by A∞ /(γ p −1)A∞ , where γ is any topological
generator of Γ.
1 Equally importantly, one should also take a note of Dirichlet’s analytic class number formula,

cf., [Was97, Chapter 4]. These express the value of the Dedekind zeta function ζK (s) at s = 0 (or
equivalently, the residue at s = 1 by its functional equation) in terms of the class number and the
unit regulator of the field K. Due the space restrictions, we do not dwell further on this topic.
Note however that the Birch and Swinnerton-Dyer conjecture (as well as its p-adic variant) that
we will discuss in Section 4 are direct analogues of this class number formula in view of the very
general conjectures of Bloch and Kato.
 p-adic Variation in Arithmetic Geometry: A Survey 9

of Theorem 2.8(iii): One may read these congruences as a statement of p-adic

continuity.
Throughout this section, we shall study the field K = Q(μp ) and the Zp -
extension we shall consider will be the cyclotomic Zp -extension Qμp∞ of Q(μp ).
Let ω : (Z/pZ)× → Z× p denote the Teichmüller character (that maps a
residue class a mod p to the unique (p − 1)st root of unity in Z× p ). Let χcyc :
Gal(Q(μp∞ )/Q) → Z× p denote the cyclotomic character, giving the action of the
Galois group on the p-power of roots of unity μp∞ . After identifying Cp (the com-
pletion of the algebraic closure Qp of Qp ) with C, we may think of ω also as a Dirich-
let character, as well as a character of the Galois group Gal(Q(μp )/Q) ∼ = (Z/pZ)× .
For every integer r we may therefore form the Dirichlet series
∞
ω r (n)
L(s, ω r ) := .
i=1
ns
Observe that for r ≡ 1(mod p − 1) we have L(s, ω r ) = ζ(s). Dirichlet L-functions
enjoy similar properties to those of the Riemann zeta function, we refer the reader
to [Was97, Chapter 4] for those that concern us here.
Theorem 3.4 ([KL64, Iwa69]). For every integer r there exists a p-adic meromor-
phic (analytic unless r ≡ 1(mod p − 1)) function Lp (s, ω r ) on the open p-adic ball
1
{s ∈ Cp : |s|p < p1− p−1 } which is characterized by the following interpolation
property:

(1 − pn−1 )L(1 − n, ω r−n ) , if n ≡ r (mod p − 1),
Lp (1 − n, ω ) =
r
L(1 − n, ω r−n ) , otherwise,
for positive integers n. As a matter of fact the function Lp (s, ω r ) is an Iwasawa
function r ≡ 1( mod p − 1): It is the Amice transform of the measure μKL r ∈ Λ on
Γ, i.e., it has the property that

r
Lp (s, ω ) = χ1−s KL
cyc dμr .
Γ
The function Lp (s, ω r ) is called the Kubota–Leopoldt p-adic L-function and
μKL
r the Kubota–Leopoldt measure.
Remark 3.5. We leave it as an exercise to verify that Λ may be identified with
measures on Γ.
Remark 3.6. Note that the factor 1 − pn−1 is the reciprocal of the p-Euler factor
of the L-function L(1 − n, ω r−n ) = ζ(1 − n) when n ≡ r (mod p − 1). For general r,
note that the Euler factor at s = 1 − n of the Dirichlet L-series L(s, ω r−n ) is (1 −
ω r−n (p)pn−1 )−1 = 1. Hence the Kubota–Leopoldt p-adic L-function interpolates
the special values of the L-series after removing the Euler factors at p.
Let M∞ denote the maximal pro-p extension of Q(μp∞ ) unramified outside
p and set X∞ := Gal(M∞ /Q(μp∞ )). It is an easy exercise to see that
Gal(Q(μp∞ )/Q) = Δ × Γ
10 K. Büyükboduk

has a natural action on X∞ . Let Ln be the maximal pro-p extension of Q(μpn )

unramified everywhere (in other words, Ln is the maximal p-extension contained
in the Hilbert class field of Q(μpn )). Class field theory yields an isomorphism
between the p-part An of the ideal class group of Q(μpn ) and the Galois group
Gal(Ln /Q(μpn )). Set A∞ := limn An . Evidently, each Ln is a subfield of M∞
←−
and therefore their joint L∞ = ∪Ln as well. This discussion allows us to view
A∞ ∼= Gal(L∞ /Q(μpn )) as a Λ-submodule of X∞ .
Theorem 3.7 (Mazur–Wiles [MW84], Iwasawa’s main conjectures). For every in-
r
teger r ≡ 1(mod p − 1), the characteristic ideal of the torsion Λ-module Xω
∞ is
generated by the Kubota–Leopoldt p-adic measure μKL r .
r
In the statement above, Xω r
∞ stands for the ω -isotypical part of the module
X∞ under the Δ-action we have noted in the paragraph above. The theorem above
should be thought of as a generalized class number formula packaging information
about the (ω r -parts of the) pro-p ideal class groups of the fields Q(μpn ) as one
climbs up the cyclotomic tower into a p-adic analytic identity.
The original proof of Mazur and Wiles in [MW84] uses the Eisenstein ideal
method, building on Ribet’s techniques. See [FKS14] for an exposition of the cir-
cle of ideas that go into the proof from a philosophical perspective. Later on,
Rubin (see the appendix of [Lan90]) gave a different (and in a certain sense of the
word, elementary) proof of the main conjectures using the cyclotomic unit Euler
system, based on Kolyvagin’s ideas. What is significant about the latter approach
is that Rubin was able to generalize his results in [Rub88, Rub91] so as to apply
for an imaginary quadratic field K (on replacing the cyclotomic units with elliptic
units) and prove the analogues the main conjectures for the unique Z2p -extension
of K. This result has marvelous consequences towards the Birch and Swinnerton-
Dyer conjecture for elliptic curves that has complex multiplication by K. We shall
come back to this point in the next section.
For a variety of results towards the main conjectures over more general totally
real and CM fields, the reader should consult [Wil90], [HT93, HT94, Hid06, Hid09],
[Mai08], [Hsi14], [Büy14], [Büy17].

4. Iwasawa theory of elliptic curves (and other geometric objects)

Let E be an elliptic curve2 defined over Q. For an arbitrary number field K, we
write E(K) for the Mordell–Weil group of E over the field K, which is simply the
group of K-rational points on E. The Mordell–Weil group E(K) is known to be a
finitely generated abelian group, so that
E(K) ∼ = Zralg (K) ⊕ E(K)tors
where E(K)tors is a finite group and ralg (K) is a non-negative integer called the
algebraic rank of E. We have a pretty satisfactory understanding of the behavior
2 We refer the reader for almost everything basic we shall need regarding elliptic curves in this

portion of our survey to [Sil09, Sil94].

 p-adic Variation in Arithmetic Geometry: A Survey 11

of the torsion subgroup E(K)tors (in fact, for example in the case when K = Q,
we understand what all the possibilities this group could possibly be, thanks to
Mazur’s phenomenal theorem [Maz77]). On the other hand, the Mordell–Weil rank
rlag is still mysterious and understanding this invariant is the subject of many
major open problems in number theory. One of these is the celebrated conjecture
of Birch and Swinnerton-Dyer, which we are to introduce next.
Just as we may attach a Dirichlet character an L-series, we may also3 asso-
ciate E an L-function, called the Hasse–Weil L-function of the elliptic curve E
and denoted by L(E/K, s). We refer the reader to [Sil09, Section C.16] for a defini-
tion of this object. And just the same way these Dirichlet L-series control various
arithmetic data associated to the Dirichlet character in question, we may inquire
whether or not the special values of L(E/K, s) have an arithmetic significance.
The Birch and Swinnerton-Dyer conjecture (BSD for short) claims that they do
and in fact furnishes us with a efficient algorithm to at least determine whether a
given elliptic curve carries an infinite number of K-rational points or not.
Conjecture 4.1 (Birch and Swinnerton-Dyer). The order of vanishing of L(E/K, s)
at s = 1 agrees with the Mordell–Weil rank ralg (K) of E/K. In particular, the
Mordell–Weil group E(K) is finite if and only if L(E/K, s) = 0.
Although we shall not record it here, we remark that the full BSD makes also
a strikingly precise prediction for the ralg (K)th derivative of L(E/K, s) in terms
of the arithmetic invariants of the elliptic curve E.
Remark 4.2. For a general number field K, we do not know whether L(E/K, s) is
defined at s = 1 or not: The defining (infinite) Euler products for L(E, s) converges
only on the right half-plane Re(s) > 2. When K = Q, it follows from Wiles’ proof
of the Taniyama–Shimura conjecture in [Wil95, TW95] that L(E/Q, s) admits an
analytic continuation to the full complex plane. For a general totally real number
field K, Taylor [Tay06] has proved that L(E/K, s) has a meromorphic continuation
to C and this is pretty much all we know for a generic (i.e., non-CM) elliptic
curve. Both of these spectacular results are corollaries of various modularity lifting
theorems, which dwell heavily on p-adic variation. We shall take the a few baby
steps towards these results in the next section.
We now discuss what extend of our knowledge towards BSD. We first make
a very honest remark: Almost nothing is known over non-abelian number fields
K and even over the field K = Q, nothing at all is known when the order of
vanishing ran (K) := ords=1 L(E/K, s) is greater than 1. This discussion should
make the limits of our exposition explicit.
4.1. The case K = Q and ran = 1
When the analytic rank is at least one, the Birch and Swinnerton-Dyer conjecture
predicts that there should be points of infinite order in the Mordell-Weil group of
3 The underlying theme is the notion of a motive: Associated to each of these objects, one may

define a motive, to which one attaches an L-function in a natural way. See [Del79, Ser86].
12 K. Büyükboduk

E. In the particular case when the analytic rank is exactly one, Heegner points
supply us systematically with these points of infinite order: This is the celebrated
theorem of Gross and Zagier.
Let us briefly recall the definition of Heegner points on the elliptic curve E.
To start, the work of Breuil, Conrad, Diamond and Taylor [BCDT01] (generalizing
[Wil95, TW95] to give the full statement of the Taniyama–Shimura conjecture)
shows that every elliptic curve over Q admits a modular parametrization
φ : X0 (N ) −→ E,
namely a non-zero morphism between the modular curve X0 (N ) of level N and the
elliptic curve E which is defined over Q. (As a side note, we should perhaps record
that the proof of this spectacular fact is based on the theme
√ of p-adic variation
as well.) Consider a quadratic imaginary field F = Q( −D) with discriminant
−D. For the sake of simplicity we shall assume that D = 3, 4 (so as to ensure
that OF× = {±1}). Fix an integer N which verifies the Heegner hypothesis, namely
that every prime dividing N splits in K/Q. This shows that there exists and
ideal N ≤ OF for which OF /N ∼ = Z/N Z. For a square-free positive integer n
that is prime to N , consider an order O of F of conductor n. Define the ideal
Nn := N ∩ O of O. This is an invertible ideal of O and it verifies N−1 ∼
n /O = Z/N Z.
In other words, N−1n /O ⊂ N −1
O/O is a cyclic subgroup of order N . As a matter
of fact, complex analytic description of elliptic curves shows us that A := C/O is
an elliptic curve (a priori defined over C) and N −1 O/O is simply the N -torsion
submodule A[N ] of the elliptic curve E. The moduli interpretation of the curve
X0 (N ) then shows that we may view the pair xn := (C/O, N−1 n /O) (which we
call an enhanced elliptic curve with Γ0 (N )-level structure) as a point on the curve
X0 (N )(C). Theory of Complex Multiplication shows that the points xn are in fact
algebraic: They are defined over the ring class field F (n) of conductor n of the
imaginary quadratic field F , i.e.,
xn = (C/O, N−1
n /O) ∈ X0 (N )(F (n)) .

Composed with the modular paramatrization φ, these yield points yn := φ(xn ) ∈

E(F (n)) on our elliptic curve itself! This system of points have many wonderful
properties which the number theorists have explored in depth since Brian Birch
introduced them some time 50 years ago. To this day, much of what we are able
to say in the case when the Hasse–Weil L-function vanishes at s = 1 rests upon
Heegner points (and related constructions).
Under the Heegner hypothesis, the Hasse–Weil L-function L(E/F, s) is forced
to vanish (to odd order) at s = 1. Based on the BSD conjecture, we are then lead
to believe that there should exist a point in E(F ) of infinite order. We may cook
up a candidate out of Heegner points: The point y1 ∈ E(F (1)) is a point on E
which is defined over the Hilbert class field F (1) of F . Taking traces, this yields a
point yF := TrF (1)/F (y1 ) ∈ E(F ). The following theorem is rather a consequence
of the celebrated Gross–Zagier formula (see [GZ86]), that expresses the derivative
L (E/F, 1) in terms of the Néron–Tate height of the Heegner point yF (in a manner
 p-adic Variation in Arithmetic Geometry: A Survey 13

compatible with the leading term formula predicted as part of the full BSD which
we have not made explicit here).
Theorem 4.3 (Gross–Zagier). Under the Heegner hypothesis, the point yF has in-
finite order iff L (E/F, 1) is non-zero, i.e., iff the analytic rank of E/F is exactly
one. In particular,
ran (E/F ) = 1 =⇒ ralg (E/F ) ≥ 1 .
Not far later than this work, Kolyvagin assembled a powerful machine (called
the method of Euler systems) which takes as an input the full collection of Heegner
points yn and gives out the following more precise version of Theorem 4.3.
Theorem 4.4 (Kolyvagin, see [Kol90, Kol91a, Kol91b]). Suppose that the field
F verifies the Heegner hypothesis and ran (E/F ) = 1. Then the algebraic rank
ralg (E/F ) also equals one, as predicted by the BSD conjecture. Moreover, the Tate–
Shafarevich group is finite as well.
Remark 4.5. Although these statements are given over an imaginary quadratic
field F , one may choose this auxiliary field in a suitable way (and use along the
way the results of [BFH90, MM91]) to formulate consequences towards the BSD
conjecture for E/Q: For r = 0 or 1,
ran (E/Q) = r =⇒ ralg (E/Q) = r
and the Tate–Shafarevich group (over Q) is finite.
Remark 4.6. The reader may rightfully worry that until this point in this section,
we have not made a single reference to the Iwasawa theory of elliptic curves nor
any sort of p-adic variational technique. Although Kolyvagin’s methods rely on p-
descend techniques, they do not have p-adic analytic flavour. In contrast, though,
the converse theorems to Kolyvagin’s theorems (due to Rubin [Rub92] in the CM
case, and for a non-CM elliptic curve established in [SU14, Ski14a, Ski14b, Wan12,
Ven16b] under varying set of assumptions ([SU14, Wan12] treats the case r = 0))
very strongly appeal to techniques from Iwasawa theory.
One should also note that in the (non-generic) case when E has CM, there
was much earlier progress on the BSD conjecture, due to [CW77, RW82] (and this
approach was ultimately perfected in [Rub91]). Let us very briefly overview the
main idea (that very much dwells on Iwasawa theoretic techniques). Let E/Q be an
elliptic curve that has CM by a quadratic imaginary field F . A fundamental object
in the study of elliptic curves is the p-adic Tate module Tp (E) := limn E(Q)[pn ],
←−
the projective limit of the p-power torsion on E. This is a free Zp -module equipped
with a continuous action of the absolute Galois group GQ of Q. By a theorem of
Faltings [Fal83], the p-adic Tate module (for a single prime p) determines E up
to isogeny (even when E has no CM), so it should not come as a surprise that
we allow ourselves to forget pretty much everything about the elliptic curve E
besides this simple linear-algebra object! The fact that E has CM translates into
the fact that Gal(F (E[p∞ ])/F ) (where F (E[p∞ ]) is the smallest field through
14 K. Büyükboduk

whose Galois group over F the action of GF on Tp (E) factors) is an abelian

group4 . Furthermore, the field F (E[p∞ ]) contains the unique Z2p -extension F∞ of
F . This essentially means that the action on the Galois module Tp (E) is given, up
to twisting by a character of Gal(F∞ /F ), simply by a Dirichlet character. Thus
takes us back to the realm of classical Iwasawa theory.
4.2. p-adic Birch and Swinnerton-Dyer conjecture and the case ran = 0
Alternative to the Heegner point Euler system utilized by Kolyvagin, there is an
equally fruitful approach developed by Kato [Kat04]. Kato’s ideas, albeit being
rather intricate and much more technically challenging (that we could hope to
explore and discuss here), are in fact rather flexible and has been revisited and
enhanced by many mathematicians in order to treat fundamental arithmetic prob-
lems not only concerning elliptic curves but also many other interesting algebro-
geometric objects. Having said that, we sadly will not have the chance to discuss
the entirety of these exciting developments (other than just citing them) besides
those have to do with elliptic curves. We shall also finally get to the point where
we will introduce Mazur’s version of Iwasawa theory of elliptic curves that he
developed in [Maz72] and along these lines, the p-adic BSD conjecture of Mazur–
Tate–Teitelbaum [MTT86].
Throughout this section, E/Q is an elliptic curve and p ≥ 3 is a fixed prime.
For simplicity, let us assume that E has ordinary reduction at p (although much of
what we are to discuss here has also a supersingular counterpart; see [Pol03, Kob03,
PR04]). Let α denote the (p-adic) unit root of the Hecke polynomial X 2 − ap X + p
and β = p/α denote the other root. In the case when E has split (resp., non-split)
multiplicative reduction at p we have α = 1 (resp., α = −1).
4.2.1. The p-adic L-function. Recall the cyclotomic Zp -extension Qcyc . We shall
denote its nth layer by Qn . Identifying Cp with C as above, the characters we shall
consider will take values both in C and Cp . Fix forever a generator {ζpn } ∈ lim μpn .
←−
Theorem 4.7 (Mazur–Tate–Teitelbaum, [MTT86]). For E, p and α given as above,
there exists a p-adic measure LE ∈ Λ on Γ which is characterized by the following
interpolation property: For every character χ of Γ of finite order we have:
⎧
⎪
⎪ (1 − α−1 )2 L(E, 1)/ΩE , if χ = 1 and E has good
 ⎪
⎪
⎨ reduction at p,
χ dLE = (1 − α−1 )L(E, 1)/ΩE , if χ = 1 and E has multiplicative
⎪
⎪
Γ ⎪
⎪ reduction at p,
⎩ −n
α τ (χ)L(E, χ−1 , 1)/ΩE , if χ has conductor pn > 1.
In the theorem above, ΩE is the canonical period (that, among other things,
ensures that L(E, χ−1 , 1)/ΩE is algebraic for every χ as above) and

τ (ψ) := ψ(γ) ζpγn
γ∈Gal(Q(μpn )/Q)

is the Gauss sum of the character ψ of conductor pn .

4 This is also the key observation that leads to Kronecker’s Jugendtraum, see [Sil94].
 p-adic Variation in Arithmetic Geometry: A Survey 15

Definition 4.8. The Mazur–Tate–Teitelbaum p-adic L-function is defined by setting


1−s
Lp (E, s) := χcyc dLE .
Γ

Being the p-adic Mellin-transform of a measure, this function is p-adic analytic in

an open ball containing s = 1. Rohrlich in [Roh84] proved that L(E, χ−1 , 1) = 0
for almost all χ. It therefore follows that LE ≡ 0 as well.
Based on our previous experience with the Kubota–Leopoldt p-adic L-func-
tion, it would not be unreasonable to expect that the measure LE (and the p-adic
L-function Lp (E, s)) carries a great deal of information regarding the behavior
of the elliptic curve E along the cyclotomic tower. We shall see that this indeed
is the case and explain the techniques to extract this information. We shall also
see how this will lead us to some spectacular consequences towards the Birch and
Swinnerton-Dyer conjecture itself (Corollary 4.18 and Remark 4.19 below).
Remark 4.9. Throughout this paragraph, we assume that the elliptic curve E has
split multiplicative reduction at p. In this set up, the interpolation formula for the
Mazur–Tate–Teitelbaum p-adic measure shows that Lp (E, 1) = 0, regardless of
the value of the Hasse–Weil L-function L(E, s). In this situation we say that the
Mazur–Tate–Teitelbaum p-adic L-function has a trivial zero (some authors prefer
the name exceptional zero). This fact (along with numerical evidence) lead Mazur,
Tate and Teitelbaum to formulate the following conjecture:
ords=1 Lp (E, s) = 1 + ords=1 L(E, s) . (4.1)
When L(E, 1) =  0, the assertion (4.1) was proved6 by Greenberg and Stevens
[GS93]. In the situation when ran = 1, the assertion (4.1) follows from [Büy16,
Ven16a].
We set rp := ords=1 Lp (E, s) and call it the p-adic analytic rank of E.
Conjecture 4.10 (p-adic BSD).
⎧
⎨ ralg , if E has good ordinary or non-split
rp = multiplicative reduction at p,
⎩
1 + ralg , if E has split multiplicative reduction at p.
As opposed to the rather limited set of results we have for the original BSD
(namely, the results at hand are restricted to the case ran ≤ 1), we have the
following theorems towards its p-adic variant:

6 In their proof, Greenberg and Stevens make fabulous use of p-adic variation: What they first

compute is various directional derivatives of the two-variable Mazur–Kitagawa function associ-

ated to a Hida family passing through the modular form fE (given by Wiles’ modularity theorem)
“at the point (2, 1)”. This allows them to calculate the derivative along the “cyclotomic direc-
tion” and deduce that Lp (E, 1) = 0, as desired. We shall revisit these themes in the following
section, see particularly Section 5.2.
16 K. Büyükboduk

Theorem 4.11 (Kato, [Kat04] and Rubin, [Rub91]). In the situation of Conjec-
ture 4.10,
⎧
⎨ ralg , if E has good ordinary or non-split
rp ≥ multiplicative reduction at p,
⎩
1 + ralg , if E has split multiplicative reduction at p.
In fact, combined with the work of Bertrand [Ber82] on the non-vanishing of
p-adic height pairings, one may deduce that
ran = 1 ⇐⇒ rp = 1 ⇐⇒ ralg = 1
in case the elliptic curve E has CM. (We remark that a CM elliptic curve may
only have additive reduction at primes of bad reduction.)
Theorem 4.12 (Skinner–Urban, [SU14]). Suppose E does not have CM, has good
ordinary reduction at p and is semistable. Suppose further that certain p-adic height
pairing (see [Sch82, Sch85, Sch83]) is non-degenerate and the Tate–Shafarevich
group is finite. Then Conjecture 4.10 holds true for p ≥ 11.
The main idea behind the proofs of these two theorems is to extract infor-
mation regarding the big Mordell–Weil group E(Qcyc ) = ∪E(Qn ) from LE , then
specialize to n = 1 (which corresponds to the Mordell–Weil group E(Q) at the
base level). This is achieved through Iwasawa–Mazur main conjectures for the el-
liptic curve E over the cyclotomic Zp -tower. Before we give the formulation of
this conjecture, we first need to introduce Selmer groups, which are fundamental
objects of interest in arithmetic geometry.
Definition 4.13. For an arbitrary number field L and positive integer m, let
Selpm (E/L)

 H 1 (Lv , E[pm ])
:= ker H (L, E[p ]) −→
1 m
Kum
v im(E(Lv )/pm E(Lv ) → H 1 (Lv , E[pm ])
be the pm -Selmer group. Here the direct product is over all primes of L and
Kum : E(Lv )/pm E(Lv )→H 1 (Lv , E[pm ])
is the Kummer map. We define the pro-p Selmer group Selp (E/L) by setting
Selpm (E/L) := lim Selpm (E/L) .
−m
→
We finally define the Iwasawa theoretic Selmer group Selp (E/Qcyc ) by
Selp (E/Qcyc ) = lim Selp (E/Qn ) .
−→n

Remark 4.14. The main interest in the Selmer group lies behind the following
classical exact sequence:
0 −→ E(L) ⊗ Qp /Zp −→ Selp (E/L) −→ IIIE/L [p∞ ] −→ 0 .
 p-adic Variation in Arithmetic Geometry: A Survey 17

The Selmer group is more amenable to calculations (and also generalizations) as

compared to the Mordell–Weil group E(L), so we are lead to think of it as a calcu-
lable approximation to the Mordell–Weil group. In the sequence above IIIE/L [p∞ ]
is the pro-p part of the Tate–Shafarevich group. This object is conjectured to be
always finite and it is perhaps the most notorious part of the whole theory.
Conjecture 4.15 (Iwasawa–Mazur conjecture). The characteristic ideal of the Pon-
tryagin dual Selp (E/Qcyc )∨ of the Iwasawa theoretic Selmer group is generated by
the Mazur–Tate–Teitelbaum p-adic measure LE .
Remark 4.16. As we have seen in Remark 4.14, one should think of Selp (E/Qcyc )∨
as an approximation to the big Mordell–Weil group E(Qcyc ) ⊗ Zp .
Thanks to the spectacular works of Kato and Skinner-Urban (as well as the
results of Rubin in the case when E has CM), the Iwasawa–Mazur conjecture is
mostly proved.
Theorem 4.17 (Kato, Skinner–Urban; Rubin). If the elliptic curve E has good
ordinary reduction at p (and verifies mild technical hypotheses) then the Iwasawa–
Mazur main conjecture holds true.
We would like to show with an example how to use Iwasawa theoretic machine
to deduce results towards the original Birch and Swinnerton-Dyer conjecture. This
we shall do in the proof of the following corollary.
Corollary 4.18. In the setting of Theorem 4.17,
ran = 0 ⇐⇒ ralg = 0 ,
as predicted by the Birch and Swinnerton-Dyer conjecture.
Sketch of the proof. In the situation of Theorem 4.17, the interpolation formula
for the p-adic L-function shows that ran = 0 iff Lp (E, 1) = 0. This amounts to
saying that LE does not belong to the augmentation ideal J := ker(Λ → Zp ).
By the main conjecture, this is equivalent to the assertion that the characteristic
ideal of Selp (E/Qcyc )∨ is not contained in J . By the structure theorem for finitely
generated torsion Λ-modules7 , all this is equivalent to the assertion that we have
a map 
Selp (E/Qcyc )∨ −→ Λ/f Λ
with finite kernel and cokernel, where the product is over a finite set of distin-
guished polynomials all coprime to J . This in turn is equivalent to saying that
the quotient
Selp (E/Qcyc )∨ /J Selp (E/Qcyc )∨ ≈ Selp (E/Q)∨
(where “≈” means isomorphism up to finite kernel and cokernel and follows from
Mazur’s control theorem) is finite. 
7 The fact that Selp (E/Qcyc )∨ is Λ-torsion follows from the main conjecture and Rohrlich’s
theorem (which shows that LE = 0).
18 K. Büyükboduk

Remark 4.19. Although one does not deduce the following converse of Theorem 4.4
(due to Skinner [Ski14a]) directly from Theorem 4.17, its proof still is purely p-adic:
In the situation of Theorem 4.17, if ralg = 1 and the p-part of the Tate–Shafarevich
group is finite, then ran = 1. (See also [Ven16b] for a similar result in the case of
split-multiplicative reduction.)

5. Deformations of Galois representations and Iwasawa theory

Inspired by Hida’s seminal work on p-adic families of p-ordinary modular forms
[Hid86a, Hid86b], Mazur gave a general framework for deformations of Galois
representations. The main insight is that natural geometric objects should fit in p-
adic analytic families and (many of the) arithmetic invariants associated with these
geometric objects should also move continuously in this family. We shall briefly
recall Hida’s theory below in Section 5.2 (which shows that any p-ordinary Hecke
eigenform fits in a p-adic family of p-adic modular forms) and also record some
of its many striking applications (one of which being the proof of the Taniyama–
Shimura conjecture in [Wil95, TW95]).
We first provide a quick overview of Mazur’s theory of Galois deformations;
see [Maz89, dSL97, Gou01] for details.

5.1. Deformations of Galois representations

Let Φ be a finite extension of Qp and O be the ring of integers of Φ. Let  ∈ O
be a uniformizer, and let k = O/ be its residue field. Consider the following
category C:
• Objects of C are complete, local, Noetherian commutative O-algebras A with
residue field kA = A/mA isomorphic to k, where mA denotes the maximal
ideal of A.
• A morphism f : A → B in C is a local O-algebra morphism.
Let Σ be a finite set of places of Q that contains p and ∞. Let GQ,Σ denote
the Galois group of the maximal extension QΣ of Q unramified outside Σ. Fix an
absolutely irreducible, continuous Galois representation ρ : GQ,Σ → GLn (k) and
let T be the representation space (so that T is an n-dimensional k-vector space on
which GQ,Σ acts continuously).
Let Dρ : C −→ Sets be the functor defined as follows. For every object A
of C, Dρ (A) is the set of continuous homomorphisms ρA : GQ,Σ → GLn (A) that
satisfy ρA ⊗A k ∼= ρ, taken modulo conjugation by the elements of GLn (A). For
every morphism f : A → B in C, Dρ (f )(ρA ) is the GLn (B)-conjugacy class of
ρA ⊗A B.
Theorem 5.1 (Mazur). The functor Dρ is representable.
In other words, there is a ring R(ρ) ∈ Ob(C) and a continuous representation
ρ : GQ −→ GLn (R(ρ))
Another random document with
no related content on Scribd:
distributing a Project Gutenberg™ electronic work under this
agreement, disclaim all liability to you for damages, costs and
expenses, including legal fees. YOU AGREE THAT YOU HAVE NO
REMEDIES FOR NEGLIGENCE, STRICT LIABILITY, BREACH OF
WARRANTY OR BREACH OF CONTRACT EXCEPT THOSE
PROVIDED IN PARAGRAPH 1.F.3. YOU AGREE THAT THE
FOUNDATION, THE TRADEMARK OWNER, AND ANY
DISTRIBUTOR UNDER THIS AGREEMENT WILL NOT BE LIABLE
TO YOU FOR ACTUAL, DIRECT, INDIRECT, CONSEQUENTIAL,
PUNITIVE OR INCIDENTAL DAMAGES EVEN IF YOU GIVE
NOTICE OF THE POSSIBILITY OF SUCH DAMAGE.

1.F.3. LIMITED RIGHT OF REPLACEMENT OR REFUND - If you

discover a defect in this electronic work within 90 days of receiving it,
you can receive a refund of the money (if any) you paid for it by
sending a written explanation to the person you received the work
from. If you received the work on a physical medium, you must
return the medium with your written explanation. The person or entity
that provided you with the defective work may elect to provide a
replacement copy in lieu of a refund. If you received the work
electronically, the person or entity providing it to you may choose to
give you a second opportunity to receive the work electronically in
lieu of a refund. If the second copy is also defective, you may
demand a refund in writing without further opportunities to fix the
problem.

1.F.4. Except for the limited right of replacement or refund set forth in
paragraph 1.F.3, this work is provided to you ‘AS-IS’, WITH NO
OTHER WARRANTIES OF ANY KIND, EXPRESS OR IMPLIED,
INCLUDING BUT NOT LIMITED TO WARRANTIES OF
MERCHANTABILITY OR FITNESS FOR ANY PURPOSE.

1.F.5. Some states do not allow disclaimers of certain implied

warranties or the exclusion or limitation of certain types of damages.
If any disclaimer or limitation set forth in this agreement violates the
law of the state applicable to this agreement, the agreement shall be
interpreted to make the maximum disclaimer or limitation permitted
by the applicable state law. The invalidity or unenforceability of any
provision of this agreement shall not void the remaining provisions.

1.F.6. INDEMNITY - You agree to indemnify and hold the

Foundation, the trademark owner, any agent or employee of the
Foundation, anyone providing copies of Project Gutenberg™
electronic works in accordance with this agreement, and any
volunteers associated with the production, promotion and distribution
of Project Gutenberg™ electronic works, harmless from all liability,
costs and expenses, including legal fees, that arise directly or
indirectly from any of the following which you do or cause to occur:
(a) distribution of this or any Project Gutenberg™ work, (b)
alteration, modification, or additions or deletions to any Project
Gutenberg™ work, and (c) any Defect you cause.

Section 2. Information about the Mission of

Project Gutenberg™
Project Gutenberg™ is synonymous with the free distribution of
electronic works in formats readable by the widest variety of
computers including obsolete, old, middle-aged and new computers.
It exists because of the efforts of hundreds of volunteers and
donations from people in all walks of life.

Volunteers and financial support to provide volunteers with the

assistance they need are critical to reaching Project Gutenberg™’s
goals and ensuring that the Project Gutenberg™ collection will
remain freely available for generations to come. In 2001, the Project
Gutenberg Literary Archive Foundation was created to provide a
secure and permanent future for Project Gutenberg™ and future
generations. To learn more about the Project Gutenberg Literary
Archive Foundation and how your efforts and donations can help,
see Sections 3 and 4 and the Foundation information page at
www.gutenberg.org.
 Section 3. Information about the Project
Gutenberg Literary Archive Foundation
The Project Gutenberg Literary Archive Foundation is a non-profit
501(c)(3) educational corporation organized under the laws of the
state of Mississippi and granted tax exempt status by the Internal
Revenue Service. The Foundation’s EIN or federal tax identification
number is 64-6221541. Contributions to the Project Gutenberg
Literary Archive Foundation are tax deductible to the full extent
permitted by U.S. federal laws and your state’s laws.

The Foundation’s business office is located at 809 North 1500 West,

Salt Lake City, UT 84116, (801) 596-1887. Email contact links and up
to date contact information can be found at the Foundation’s website
and official page at www.gutenberg.org/contact

Section 4. Information about Donations to

the Project Gutenberg Literary Archive
Foundation
Project Gutenberg™ depends upon and cannot survive without
widespread public support and donations to carry out its mission of
increasing the number of public domain and licensed works that can
be freely distributed in machine-readable form accessible by the
widest array of equipment including outdated equipment. Many small
donations ($1 to $5,000) are particularly important to maintaining tax
exempt status with the IRS.

The Foundation is committed to complying with the laws regulating

charities and charitable donations in all 50 states of the United
States. Compliance requirements are not uniform and it takes a
considerable effort, much paperwork and many fees to meet and
keep up with these requirements. We do not solicit donations in
locations where we have not received written confirmation of
compliance. To SEND DONATIONS or determine the status of
compliance for any particular state visit www.gutenberg.org/donate.

While we cannot and do not solicit contributions from states where

we have not met the solicitation requirements, we know of no
prohibition against accepting unsolicited donations from donors in
such states who approach us with offers to donate.

International donations are gratefully accepted, but we cannot make

any statements concerning tax treatment of donations received from
outside the United States. U.S. laws alone swamp our small staff.

Please check the Project Gutenberg web pages for current donation
methods and addresses. Donations are accepted in a number of
other ways including checks, online payments and credit card
donations. To donate, please visit: www.gutenberg.org/donate.

Section 5. General Information About Project

Gutenberg™ electronic works
Professor Michael S. Hart was the originator of the Project
Gutenberg™ concept of a library of electronic works that could be
freely shared with anyone. For forty years, he produced and
distributed Project Gutenberg™ eBooks with only a loose network of
volunteer support.

Project Gutenberg™ eBooks are often created from several printed

editions, all of which are confirmed as not protected by copyright in
the U.S. unless a copyright notice is included. Thus, we do not
necessarily keep eBooks in compliance with any particular paper
edition.

Most people start at our website which has the main PG search
facility: www.gutenberg.org.

This website includes information about Project Gutenberg™,

including how to make donations to the Project Gutenberg Literary
Archive Foundation, how to help produce our new eBooks, and how
to subscribe to our email newsletter to hear about new eBooks.