Sharedwww / job / Prop.texOpen in CoCalc
Author: William A. Stein
1
\documentclass[12pt]{article}
2
3
\newcommand{\doc}{Detailed Research Plan}
4
\newcommand{\myname}{William A.\ Stein}
5
\newcommand{\phone}{(510) 883-9938}
6
\newcommand{\email}{{\tt was@math.berkeley.edu}}
7
\newcommand{\www}{{\tt http://www.math.berkeley.edu/\~{\mbox{}}was}}
8
\newcommand{\address}{2041 Francisco Street, \#14\\
9
Berkeley, CA 94709\\
10
USA}
11
12
13
\hoffset=-.1\textwidth
14
\textwidth=1.2\textwidth
15
\voffset=-.1\textheight
16
\textheight=1.2\textheight
17
18
\usepackage{fancyhdr,ifthen}
19
\pagestyle{fancy}
20
\cfoot{} % no footers (in pagestyle fancy)
21
% running left heading
22
\lhead{\bfseries\LARGE\em \noindent{}\hspace{-.2em}\myname{}
23
\hfill \doc\vspace{-.2ex}\\}
24
% running right heading
25
\newcommand{\spc}{.33ex}
26
%\rhead{\em {\small{\phone{}}} \hspace{\spc}$\bullet$\hspace{\spc} \email{}
27
% \hspace{\spc}$\bullet$\hspace{\spc} \www{}}
28
% adjust, because the header is now taller than usual.
29
\rhead{}
30
\setlength{\headheight}{7ex}
31
32
33
\usepackage{amsmath}
34
\usepackage{amsthm}
35
\usepackage{amssymb}
36
37
38
39
40
\newcommand{\Adual}{A^{\vee}}
41
\newcommand{\isom}{\cong}
42
%\newcommand{\ncisom}{\approx}
43
%\newcommand{\mysection}[1]{\vspace{12pt}\noindent{\large\bf #1}\vspace{1ex}}
44
\newcommand{\mysection}[1]{\section{#1}\vspace{-1.5ex}}
45
%\newcommand{\mysubsection}[1]{\noindent{\bf #1}}
46
\newcommand{\mysubsection}[1]{\subsection{#1}\vspace{-1ex}}
47
\newcommand{\Q}{\mathbf{Q}}
48
\newcommand{\Z}{\mathbf{Z}}
49
\newcommand{\R}{\mathbf{R}}
50
\newcommand{\C}{\mathbf{C}}
51
\newcommand{\PGL}{\mbox{\rm PGL}}
52
\newcommand{\GL}{\mbox{\rm GL}}
53
\newcommand{\Div}{\mbox{\rm Div}}
54
\newcommand{\GQ}{G_\Q}
55
\newcommand{\ra}{\rightarrow}
56
\newcommand{\tensor}{\otimes}
57
\DeclareMathOperator{\tor}{tor}
58
\font\cyr=wncyr10 scaled \magstep 1
59
\newcommand{\Sha}{\mbox{\cyr X}}
60
\font\cyrbig=wncyr10 scaled \magstep 2
61
\newcommand{\BigSha}{\mbox{\cyrbig X}}
62
\newcommand{\intersect}{\cap}
63
\newcommand{\comment}[1]{}
64
\theoremstyle{plain}
65
\newtheorem{theorem}{Theorem}
66
\newtheorem{corollary}[theorem]{Corollary}
67
\newtheorem{conjecture}[theorem]{Conjecture}
68
69
\begin{document}
70
71
72
\begin{center}
73
{\LARGE \bf \doc}\\
74
\end{center}
75
76
\mysection{Introduction}
77
My research program reflects the essential interplay between abstract
78
theory and explicit machine computation during
79
the latter half of the twentieth century;
80
it sits at the intersection of recent work of
81
B.~Mazur, K.~Ribet, J-P.~Serre, R.~Taylor, and A.~Wiles
82
on Galois representations attached to modular abelian varieties
83
(see \cite{ribet:modreps, serre:conjectures,
84
taylor-wiles:fermat, wiles:fermat})
85
with work of J.~Cremona, N.~Elkies, and J.-F.~Mestre
86
on explicit computations involving modular forms
87
(see \cite{cremona:algs, elkies:ffield}).
88
89
In 1969 B.~Birch~\cite{birch:bsd} described computations that
90
led to the most fundamental open
91
conjecture in the theory of elliptic curves:
92
\begin{quote}
93
I want to describe some computations undertaken by myself and
94
Swinnerton-Dyer on EDSAC
95
by which we have calculated the zeta-functions of certain elliptic
96
curves. As a result of these computations we have found an analogue
97
for an elliptic curve of the Tamagawa number of an algebraic group;
98
and conjectures (due to ourselves, due to Tate, and
99
due to others) have proliferated.
100
\end{quote}
101
The rich tapestry of arithmetic conjectures and theory
102
we enjoy today would not exist without the ground-breaking
103
application of computing by Birch and Swinnerton-Dyer.
104
Computations in the 1980s by Mestre were key in convincing
105
Serre that his conjectures on modularity of odd
106
irreducible Galois representations
107
were worthy of serious consideration (see~\cite{serre:conjectures}).
108
These conjectures have inspired much recent work;
109
for example, Ribet's proof of the
110
$\epsilon$-conjecture, which played an essential role in
111
Wiles's proof of Fermat's Last Theorem.
112
113
My work on the Birch and Swinnerton-Dyer
114
conjecture for modular abelian varieties and search for new
115
examples of modular icosahedral Galois representations
116
has led me to discover and implement algorithms for
117
explicitly computing with modular forms.
118
My research, which involves finding ways to compute
119
with modular forms and modular abelian varieties,
120
is driven by outstanding conjectures in number theory.
121
122
\mysection{Invariants of modular abelian varieties}
123
Now that the Shimura-Taniyama conjecture has been proved,
124
the main outstanding problem in the field
125
is the Birch and Swinnerton-Dyer conjecture (BSD conjecture),
126
which ties together the arithmetic invariants of an elliptic curve.
127
There is no general class of elliptic curves for which
128
the full BSD conjecture is known.
129
Approaches to the BSD conjecture that rely on congruences
130
between modular forms are likely to
131
require a deeper understanding of the analogous
132
conjecture for higher-dimensional abelian varieties.
133
As a first step, I have obtained theorems that make possible explicit
134
computation of some of the arithmetic invariants of modular
135
abelian varieties.
136
137
\mysubsection{The BSD conjecture}
138
By~\cite{breuil-conrad-diamond-taylor} we now know
139
that every elliptic curve over~$\Q$ is
140
a quotient of the curve~$X_0(N)$ whose complex points
141
are the isomorphism classes of pairs consisting of a
142
(generalized) elliptic curve and a cyclic subgroup of order~$N$.
143
Let~$J_0(N)$ denote the Jacobian of $X_0(N)$; this is an abelian
144
variety of dimension equal to the genus of~$X_0(N)$ whose points
145
correspond to the degree~$0$ divisor classes on~$X_0(N)$.
146
147
An {\em optimal quotient} of $J_0(N)$ is a quotient by an abelian subvariety.
148
Consider an optimal quotient~$A$ such that $L(A,1)\neq 0$.
149
By~\cite{kolyvagin-logachev:totallyreal},~$A(\Q)$ and~$\Sha(A/\Q)$
150
are both finite.
151
The BSD conjecture asserts that
152
$$\frac{L(A,1)}{\Omega_A} =
153
\frac{\#\Sha(A/\Q)\cdot\prod_{p\mid N} c_p}
154
{\# A(\Q)\cdot\#\Adual(\Q)}.$$
155
Here the Shafarevich-Tate group $\Sha(A/\Q)$ is a measure of the failure
156
of the local-to-global principle; the Tamagawa numbers~$c_p$ are the
157
orders of the component groups of~$A$; the real number~$\Omega_A$ is
158
the volume of~$A(\R)$ with respect to a basis of differentials having
159
everywhere nonzero good reduction; and~$\Adual$ is the dual of~$A$.
160
My goal is to verify the full
161
conjecture for many specific abelian varieties on a case-by-case basis.
162
This is the first step in a program to verify the above
163
conjecture for an infinite family of quotients of~$J_0(N)$.
164
165
\mysubsection{The ratio $L(A,1)/\Omega_A$}
166
Following Y.~Manin's work on elliptic curves,
167
A.~Agash\'e and I proved the following
168
theorem in~\cite{stein:vissha}.
169
\begin{theorem}\label{thm:ratpart}
170
Let~$m$ be the largest square dividing~$N$.
171
The ratio $L(A,1)/\Omega_A$ is a rational number that can be
172
explicitly computed, up to a unit (conjecturally $1$) in $\Z[1/(2m)]$.
173
\end{theorem}
174
The proof uses modular symbols combined with an extension of the argument
175
used by Mazur in~\cite{mazur:rational} to bound the Manin constant.
176
The ratio $L(A,1)/\Omega_A$ is expressed as the lattice index of
177
two modules over the Hecke algebra.
178
I expect the method to give similar results
179
for special values of twists, and of
180
$L$-functions attached to eigenforms of higher weight.
181
I have computed $L(A,1)/\Omega_A$ for all optimal
182
quotients of level $N\leq 1500$; this table continues to be
183
of value to number theorists.
184
185
\mysubsection{The torsion subgroup}
186
I can compute upper and lower bounds on $\#A(\Q)_{\tor}$, but I can not
187
determine $\#A(\Q)_{\tor}$ in all cases. Experimentally, the deviation
188
between the upper and lower bound is reflected in congruences with
189
forms of lower level; I hope to exploit this in a precise way.
190
I also obtained the following
191
intriguing corollary that suggests cancellation between
192
torsion and~$c_p$; it generalizes to higher weight forms, thus
193
suggesting a geometric explanation for reducibility
194
of Galois representations.
195
\begin{corollary}
196
Let~$n$ be the order of the image of
197
$(0)-(\infty)$ in $A(\Q)$, and
198
let~$m$ be the largest square dividing~$N$.
199
Then $\displaystyle n\cdot L(A,1)/\Omega_A$ is an integer,
200
up to a unit in $\Z[1/(2m)]$.
201
\end{corollary}
202
203
\mysubsection{Tamagawa numbers}
204
\begin{theorem}\label{thm:tamagawa}
205
When $p^2\nmid N$, the number~$c_p$ can be explicitly computed
206
(up to a power of~$2$).
207
\end{theorem}
208
I prove this in~\cite{stein:compgroup}. Several related problems
209
remain: when $p^2 \mid N$ it may be possible to compute~$c_p$
210
using the Drinfeld-Katz-Mazur interpretation
211
of~$X_0(N)$; it should also be possible to use my
212
methods to treat optimal quotients of $J_1(N)$.
213
214
I was surprised to find that systematic computations using this
215
formula indicate the following conjectural refinement of a result of
216
Mazur~\cite{mazur:eisenstein}.
217
\begin{conjecture}
218
Suppose~$N$ is prime and~$A$
219
is an optimal quotient of $J_0(N)$. Then $A(\Q)_{\tor}$ is
220
generated by the image of $(0)-(\infty)$
221
and $c_p = \#A(\Q)_{\tor}$. Furthermore,
222
the product of the~$c_p$ over all optimal factors
223
equals the numerator of $(N-1)/12$.
224
\end{conjecture}
225
I have checked this conjecture for all $N\leq 997$ and,
226
up to a power of~$2$, for all $N\leq 2113$.
227
The first part is known when~$A$ is an elliptic
228
curve (see~\cite{mestre-oesterle:crelle}).
229
Upon hearing of this conjecture, Mazur proved it
230
when all ``$q$-Eisenstein quotients'' are simple.
231
There are three promising approaches to finding
232
a complete proof. One involves the explicit
233
formula of Theorem~\ref{thm:tamagawa};
234
another is based on Ribet's level lowering theorem,
235
and a third makes use of a simplicity result of Merel.
236
237
Theorem~\ref{thm:tamagawa}
238
also suggests a way to compute Tamagawa numbers of
239
motives attached to eigenforms of higher weight.
240
These numbers appear in the conjectures of Bloch and Kato,
241
which generalize the BSD conjecture to motives (see~\cite{bloch-kato}).
242
243
244
\mysubsection{Upper bounds on $\#\BigSha$}
245
V.~Kolyvagin and K.~Kato~\cite{kolyvagin:structureofsha, scholl:kato}
246
obtained upper bounds on~$\#\Sha(A)$.
247
To verify the full BSD conjecture for certain abelian
248
varieties, it is necessary is to make these bounds explicit.
249
Kolyvagin's bounds involve computations with Heegner points,
250
and Kato's involve a study of the Galois representations
251
associated to~$A$. I plan to carry out such
252
computations in many specific cases.
253
254
\mysubsection{Lower bounds on $\#\BigSha$}
255
One approach to showing that~$\Sha$ is as large as predicted
256
by the BSD conjecture is suggested by Mazur's notion of
257
the visible part of~$\Sha$ (see~\cite{cremona-mazur, mazur:visthree}).
258
Let~$\Adual$ be the dual of~$A$.
259
The visible part of $\Sha(\Adual/\Q)$ is the
260
kernel of $\Sha(\Adual/\Q)\ra \Sha(J_0(N))$.
261
Mazur observed that if an element of order~$p$
262
in~$\Sha(\Adual/\Q)$ is visible,
263
then it is explained by a jump in the rank of Mordell-Weil
264
in the sense that there is another abelian subvariety $B\subset J_0(N)$
265
such that $p \mid \#(\Adual\intersect B)$ and the rank of~$B$ is positive.
266
I think that this observation can be turned around: if there is
267
another abelian
268
variety~$B$ of positive rank such that $p\mid \#(\Adual\intersect B)$,
269
then, under mild hypotheses, there is an element of~$\Sha(\Adual/\Q)$
270
of order~$p$. Thus the theory of congruences between modular
271
forms can be used to obtain a lower bound on~$\#\Sha(\Adual/\Q)$.
272
I am trying to use the cohomological methods of~\cite{mazur:tower}
273
and suggestions of B.~Conrad and Mazur to prove the following conjecture.
274
\begin{conjecture}
275
Let~$\Adual$ and~$B$ be abelian subvarieties of~$J_0(N)$.
276
Suppose that $p\mid \#(\Adual\intersect B)$, that~$p\nmid N$,
277
and that~$p$ does not divide the
278
order of any of the torsion subgroups
279
or component groups of~$A$ or~$B$. Then
280
$(B(\Q) \oplus \Sha(B/\Q))\tensor\Z/p\Z
281
\isom (\Adual(\Q)\oplus \Sha(\Adual/\Q) )\tensor\Z/p\Z$.
282
\end{conjecture}
283
Unfortunately, $\Sha(\Adual/\Q)$ can fail to be
284
visible inside~$J_0(N)$.
285
For example, I found that the BSD conjecture
286
predicts the existence of invisible elements of odd order in~$\Sha$
287
for at least~$15$ of the~$37$ optimal quotients of prime
288
level $\leq 2113$.
289
For every integer~$M$ (Ribet~\cite{ribet:raising} tells us which~$M$
290
to choose), we can consider the images of~$\Adual$ in $J_0(NM)$.
291
There is not yet enough evidence to conjecture the existence of
292
an integer~$M$ such that all of $\Sha(\Adual/\Q)$ is visible in
293
$J_0(NM)$.
294
I am gathering data to determine whether or not
295
to expect the existence of such~$M$.
296
297
\mysubsection{Motivation for considering abelian varieties}
298
If $A$ is an elliptic curve, then explaining~$\Sha(A/\Q)$ using
299
only congruences between elliptic curves is bound to fail.
300
This is because pairs of nonisogenous elliptic curves
301
with isomorphic $p$-torsion are, according to
302
E.~Kani's conjecture, extremely rare.
303
It is crucial to understand what happens in all dimensions.
304
305
Within the range accessible by computer,
306
abelian varieties exhibit more richly textured
307
structure than elliptic curves.
308
For example, I discovered a visible element of prime order $83341$
309
in the Shafarevich-Tate group of an abelian variety of
310
prime conductor~$2333$; in contrast, over all optimal elliptic
311
curves of conductor up to $5500$, it appears that
312
the largest order of an element of a Shafarevich-Tate
313
group is~$7$.
314
315
316
\mysection{Conjectures of Artin, Merel, and Serre}
317
318
\mysubsection{Icosahedral Galois representations}
319
E.~Artin conjectured in~\cite{artin:conjecture}
320
that the $L$-series associated to any continuous irreducible
321
representation $\rho:\GQ\ra \GL_n(\C)$, with $n>1$, is entire.
322
Recent exciting work of Taylor and others suggests that a complete proof
323
of Artin's conjecture, in the case when $n=2$ and~$\rho$ is odd,
324
is on the horizon.
325
This case of Artin's conjecture is known when the image
326
of~$\rho$ in $\PGL_2(\C)$ is solvable
327
(see~\cite{tunnell:artin}), and in infinitely
328
many cases when the image of~$\rho$ is not solvable (see~\cite{bdsbt}).
329
330
In 1998, K.~Buzzard suggested a way to combine the
331
main theorem of~\cite{buzzard-taylor}, along with
332
a computer computation, to deduce modularity of certain
333
icosahedral Galois representations. Buzzard and I recently obtained
334
the following theorem.
335
\begin{theorem}
336
The icosahedral Artin representations of conductor~$1376=2^5\cdot 43$
337
are modular.
338
\end{theorem}
339
We expect our method to yield several more examples.
340
These ongoing computations are laying a small part of the technical
341
foundations necessary for a full proof of the Artin conjecture
342
for odd two dimensional~$\rho$, as well as stimulating
343
the development of new algorithms for computing with modular forms
344
using modular symbols in characteristic~$\ell$.
345
346
\mysubsection{Cyclotomic points on modular curves}
347
If~$E$ is an elliptic curve over~$\Q$
348
and~$p$ is an odd prime, then the $p$-torsion
349
on~$E$ can not all lie in~$\Q$; because of
350
the Weil pairing the $p$-torsion generates a field
351
that contains~$\Q(\mu_p)$.
352
For which primes~$p$ does
353
there exist an elliptic curve~$E$ over $\Q(\mu_p)$
354
with all of its $p$-torsion rational over $\Q(\mu_p)$?
355
When $p=2,3,5$ the corresponding moduli space has genus zero
356
and infinitely many examples exist. Recent work of L.~Merel,
357
combined with computations he enlisted me to do,
358
suggest that these are the only primes~$p$ for
359
which such elliptic curves exist.
360
In~\cite{merel:cyclo}, Merel exploits
361
cyclotomic analogues of the techniques used
362
in his proof of the uniform boundedness conjecture
363
to obtain an explicit criterion that can
364
be used to answer the above question for many primes~$p$,
365
on a case-by-case basis.
366
Theoretical work of Merel, combined with my computations of
367
twisted $L$-values and character groups of tori,
368
give the following result (see~\cite[\S3.2]{merel:cyclo}):
369
\begin{theorem}
370
Let $p \equiv 3\pmod{4}$ be a prime satisfying
371
$7 \leq p < 1000$. There are no elliptic curves
372
over $\Q(\mu_p)$ all of whose $p$-torsion is rational
373
over $\Q(\mu_p)$.
374
\end{theorem}
375
The case in which~$p$ is congruent to~$1$ modulo~$4$ presents
376
additional difficulties that involve showing that~$Y(p)$ has no
377
$\Q(\sqrt{p})$-rational points. Merel and I hope to tackle these
378
difficulties in the near future.
379
380
381
\mysubsection{Serre's conjecture modulo $pq$}
382
Let~$p$ and~$q$ be primes, and consider a continuous
383
representation $\rho:\GQ\ra \GL(2,\Z/pq\Z)$
384
that is irreducible in the sense that its reductions
385
modulo~$p$ and modulo~$q$ are both irreducible.
386
Call~$\rho$ {\em modular} if there is a modular
387
form~$f$ such that a mod~$p$ representation attached to~$f$
388
is the mod~$p$ reduction of~$\rho$, and ditto for~$q$.
389
I have carried out specific computations suggested by Mazur
390
in hopes of determining when one should expect
391
that such mod~$pq$ representations are modular; the computation
392
suggests that the right conjectures are elusive.
393
Ribet's theorem (see~\cite{ribet:raising})
394
produces infinitely many levels $pq\ell$ at which there is
395
a form giving rise to $\rho$~mod~$p$ and another
396
giving rise to $\rho$~mod~$q$; we hope to determine if for some~$\ell$
397
there is a single form giving rise to both reductions.
398
399
\mysection{Genus one curves}
400
The index of an algebraic curve~$C$ over~$\Q$ is the
401
order of the cokernel of the degree map $\Div_\Q(C)\ra\Z$;
402
rationality of the canonical divisor implies that the index
403
divides $2g-2$, where~$g$ is the genus of~$C$.
404
When $g=1$ this is no condition at all; Artin conjectured, and
405
Lang and Tate~\cite{lang-tate} proved, that for every integer~$m$
406
there is a genus one curve of index~$m$ over some number field.
407
Their construction yields genus one curves over~$\Q$ only for a few
408
values of~$m$, and they ask whether one can find genus one curves
409
over~$\Q$ of every index. I have answered
410
this question for odd~$m$.
411
\begin{theorem}
412
Let~$K$ be any number field. There are genus one curves over~$K$
413
of every odd index.
414
\end{theorem}
415
The proof involves showing that enough cohomology classes in
416
Kolyvagin's Euler system of Heegner points do not vanish
417
combined with explicit Heegner point computations.
418
I hope to show that curves of every index occur, and
419
to determine the consequences of my
420
nonvanishing result for Selmer groups. This can be viewed
421
as a contribution to the problem of understanding $H^1(\Q,E)$.
422
423
424
425
\providecommand{\bysame}{\leavevmode\hbox to3em{\hrulefill}\thinspace}
426
\newcommand{\closer}{\vspace{-1.5ex}}
427
\begin{thebibliography}{10}
428
429
\bibitem{agashe}
430
A.~Agash\'{e}, \emph{On invisible elements of the {T}ate-{S}hafarevich group},
431
C. R. Acad. Sci. Paris S\'er. I Math. \textbf{328} (1999), no.~5, 369--374.
432
\closer
433
434
\bibitem{stein:vissha}
435
A.~Agash\'{e} and W.\thinspace{}A. Stein, \emph{Visibility of
436
{S}hafarevich-{T}ate groups of modular abelian varieties}, in preparation
437
(1999).
438
\closer
439
440
\bibitem{artin:conjecture}
441
E.~Artin, \emph{{\"U}ber eine neue {A}rt von {L}-{R}eihen}, Abh. Math. Sem. Univ. Hamburg \textbf{3} (1923), 89--108.
442
\closer
443
444
\bibitem{birch:bsd}
445
B.\thinspace{}J. Birch, \emph{Elliptic curves over \protect{${\mathbf{Q}}$:
446
{A}} progress report}, 1969 Number Theory Institute (Proc. Sympos. Pure
447
Math., Vol. XX, State Univ. New York, Stony Brook, N.Y., 1969), Amer. Math.
448
Soc., Providence, R.I., 1971, pp.~396--400.
449
\closer
450
451
\bibitem{bloch-kato}
452
S.~Bloch and K.~Kato, \emph{\protect{${L}$}-functions and \protect{T}amagawa
453
numbers of motives}, The Grothendieck Festschrift, Vol. \protect{I},
454
Birkh\"auser Boston, Boston, MA, 1990, pp.~333--400.
455
\closer
456
457
\bibitem{breuil-conrad-diamond-taylor}
458
C.~Breuil, B.~Conrad, F.~Diamond, and R.~Taylor, \emph{On the modularity of
459
elliptic curves over~\protect{$\Q$}},
460
in preparation.
461
\closer
462
463
\bibitem{bdsbt}
464
K.~Buzzard, M.~Dickinson, N.~Shepherd-Barron, and R.~Taylor, \emph{On
465
icosahedral {A}rtin representations},
466
available at {\tt http://www.math.harvard.edu/\~{}rtaylor/}.
467
\closer
468
469
\bibitem{buzzard-taylor}
470
K.~Buzzard and R.~Taylor, \emph{Companion forms and weight one
471
forms}, Annals of Math. (1999).
472
\closer
473
474
\bibitem{cremona:algs}
475
J.\thinspace{}E. Cremona, \emph{Algorithms for modular elliptic curves}, second
476
ed., Cambridge University Press, Cambridge, 1997.
477
\closer
478
479
\bibitem{cremona-mazur}
480
J.\thinspace{}E. Cremona and B.~Mazur, \emph{Visualizing elements in the
481
\protect{Shafarevich-Tate} group}, Proceedings of the Arizona Winter School
482
(1998).
483
\closer
484
485
\bibitem{elkies:ffield}
486
N.\thinspace{}D. Elkies, \emph{Elliptic and modular curves over finite fields
487
and related computational issues}, Computational perspectives on number
488
theory (Chicago, IL, 1995), Amer. Math. Soc., Providence, RI, 1998,
489
pp.~21--76.
490
\closer
491
492
\bibitem{kolyvagin:structureofsha}
493
V.\thinspace{}A. Kolyvagin, \emph{On the structure of {S}hafarevich-{T}ate
494
groups}, Algebraic geometry (Chicago, IL, 1989), Springer, Berlin, 1991,
495
pp.~94--121.
496
\closer
497
498
\bibitem{kolyvagin-logachev:totallyreal}
499
V.\thinspace{}A. Kolyvagin and D.\thinspace{}Y. Logachev, \emph{Finiteness of
500
\protect{$\Sha$} over totally real fields}, Math. USSR Izvestiya \textbf{39}
501
(1992), no.~1, 829--853.
502
\closer
503
504
\bibitem{lang-tate}
505
S.~Lang and J.~Tate, \emph{Principal homogeneous spaces over abelian
506
varieties}, Amer. J. Math. \textbf{80} (1958), 659--684.
507
\closer
508
509
\bibitem{mazur:tower}
510
B.~Mazur, \emph{Rational points of abelian varieties with values in towers of
511
number fields}, Invent. Math. \textbf{18} (1972), 183--266.
512
\closer
513
514
\bibitem{mazur:eisenstein}
515
\bysame, \emph{Modular curves and the \protect{Eisenstein} ideal}, Inst. Hautes
516
\'Etudes Sci. Publ. Math. (1977), no.~47, 33--186 (1978).
517
\closer
518
519
\bibitem{mazur:rational}
520
\bysame, \emph{Rational isogenies of prime degree (with an appendix by {D}.
521
{G}oldfeld)}, Invent. Math. \textbf{44} (1978), no.~2, 129--162.
522
\closer
523
524
\bibitem{mazur:visthree}
525
\bysame, \emph{Visualizing elements of order three in the {S}hafarevich-{T}ate
526
group}, preprint (1999).
527
\closer
528
529
\bibitem{merel:cyclo}
530
L.~Merel, \emph{Sur la nature non-cyclotomique des points d'ordre fini des
531
courbes elliptiques}, preprint (1999).
532
\closer
533
534
%\bibitem{mestre:graphs}
535
%J.-F. Mestre, \emph{La m\'ethode des graphes. \protect{Exemples} et
536
% applications}, Proceedings of the international conference on class numbers
537
% and fundamental units of algebraic number fields (Katata) (1986), 217--242.
538
%\closer
539
540
\bibitem{mestre-oesterle:crelle}
541
J.-F. Mestre and J.~Oesterl{\'e}, \emph{Courbes de {W}eil semi-stables de
542
discriminant une puissance \protect{$m$}-i\`eme}, J. Reine Angew. Math.
543
\textbf{400} (1989), 173--184.
544
\closer
545
546
\bibitem{ribet:modreps}
547
K.\thinspace{}A. Ribet, \emph{On modular representations of \protect{${\rm
548
{G}al}(\overline{\bf{Q}}/{\bf {Q}})$} arising from modular forms}, Invent.
549
Math. \textbf{100} (1990), no.~2, 431--476.
550
\closer
551
552
\bibitem{ribet:raising}
553
\bysame, \emph{Raising the levels of modular representations}, S\'eminaire de
554
Th\'eorie des Nombres, Paris 1987--88, Birkh\"auser Boston, Boston, MA, 1990,
555
pp.~259--271.
556
\closer
557
558
\bibitem{scholl:kato}
559
A.\thinspace{}J. Scholl, \emph{An introduction to {K}ato's {E}uler systems},
560
Galois Representations in Arithmetic Algebraic Geometry, Cambridge University
561
Press, 1998, pp.~379--460.
562
\closer
563
564
\bibitem{serre:conjectures}
565
J-P. Serre, \emph{Sur les repr\'esentations modulaires de degr\'e \protect{$2$}
566
de \protect{${\rm{G}al}(\overline{\bf {Q}}/{\bf {Q}})$}}, Duke Math. J.
567
\textbf{54} (1987), no.~1, 179--230.
568
\closer
569
570
\bibitem{stein:compgroup}
571
W.\thinspace{}A. Stein, \emph{Component groups of optimal quotients of
572
{J}acobians}, preprint (1999).
573
\closer
574
575
\bibitem{taylor-wiles:fermat}
576
R.~Taylor and A.\thinspace{}J. Wiles, \emph{Ring-theoretic properties of
577
certain \protect{H}ecke algebras}, Ann. of Math. (2) \textbf{141} (1995),
578
no.~3, 553--572.
579
\closer
580
581
\bibitem{tunnell:artin}
582
J.~Tunnell, \emph{Artin's conjecture for representations of octahedral type},
583
Bull. Amer. Math. Soc. (N.S.) \textbf{5} (1981), no.~2, 173--175.
584
\closer
585
586
\bibitem{wiles:fermat}
587
A.\thinspace{}J. Wiles, \emph{Modular elliptic curves and \protect{F}ermat's
588
last theorem}, Ann. of Math. (2) \textbf{141} (1995), no.~3, 443--551.
589
\closer
590
591
\end{thebibliography}
592
593
594
595
\end{document}
596