An Introduction To Algebraic Geometry and Algebraic Groups 1st Edition Meinolf Geck Download PDF
An Introduction To Algebraic Geometry and Algebraic Groups 1st Edition Meinolf Geck Download PDF
An Introduction To Algebraic Geometry and Algebraic Groups 1st Edition Meinolf Geck Download PDF
com
https://ebookgate.com/product/an-introduction-to-
algebraic-geometry-and-algebraic-groups-1st-
edition-meinolf-geck/
https://ebookgate.com/product/an-algebraic-introduction-to-
complex-projective-geometry-commutative-algebra-1st-edition-
christian-peskine/
https://ebookgate.com/product/liu-algebraic-geometry-1st-edition-
qing-liu/
https://ebookgate.com/product/geometry-of-time-spaces-non-
commutative-algebraic-geometry-applied-to-quantum-theory-1st-
edition-audal/
https://ebookgate.com/product/actions-and-invariants-of-
algebraic-groups-1st-edition-walter-ferrer-santos/
Hodge Theory and Complex Algebraic Geometry I 1st
Edition Claire Voisin
https://ebookgate.com/product/hodge-theory-and-complex-algebraic-
geometry-i-1st-edition-claire-voisin-2/
https://ebookgate.com/product/computation-with-linear-algebraic-
groups-1st-edition-willem-adriaan-de-graaf/
https://ebookgate.com/product/algebraic-models-in-geometry-first-
edition-yves-felix/
https://ebookgate.com/product/semidefinite-optimization-and-
convex-algebraic-geometry-content-added-1st-edition-grigoriy-
blekherman/
https://ebookgate.com/product/affine-algebraic-geometry-
proceedings-of-the-conference-1st-edition-kayo-masuda/
OXFORD GRADUATE TEXTS IN MATHEMATICS
Series Editors
R. COHEN S.K. DONALDSON S. HILDEBRANDT
T.J. LYONS M.J. TAYLOR
OXFORD GRADUATE TEXTS IN MATHEMATICS
1
3
Great Clarendon Street, Oxford OX2, 6DP,
United Kingdom
Oxford University Press is a department of the University of Oxford.
It furthers the Universitys objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
c Oxford University Press, 2003
The moral rights of the author have been asserted
First published 2003
First published in paperback 2013
Impression: 1
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above
You must not circulate this work in any other form
and you must impose this same condition on any acquirer
British Library Cataloguing in Publication Data
Data available
ISBN 978–0–19–852831–9 (Hbk.)
ISBN 978–0–19–967616–3 (Pbk.)
Printed in Great Britain by
Biddles Ltd, www.biddles.co.uk
Preface
or variants thereof. Later on, it is very likely that the author will
fix an F -stable Borel subgroup, an F -stable maximal torus therein,
and so on. One of the aims of this book is to explain these notions
carefully, to prove some fundamental results about them, and to
illustrate the general theory with many examples which are worked
out in great detail. In fact, this book arose from the desire to provide
a general introduction into the background material from algebraic
geometry which is needed to enter this web of ideas. This introduc-
tion should be at the same time as elementary as possible and yet
lead to some substantial results.
One of the most striking advances in this area is the Deligne–
Lusztig theory of representations of finite groups of Lie type. In the
last three sections of this book, we will arrive at a point where we
can give a first introduction to these ideas and study one family of
finite groups of Lie type in detail. For those who wish to read more
on this subject, there should then be no difficulty in continuing with
Lusztig (1977), Carter (1985), or Digne and Michel (1991).
Each chapter of this book has an introduction which explains
in more detail the content and the main results; at the end, the
reader will find bibliographic remarks and exercises. Usually, there
are rather explicit hints to solutions, at least for those exercises
which are used in the text. As for the prerequisites, I assume a
good knowledge of the material in a standard algebra course (the
basics about groups, commutative rings, polynomials, and fields). In
Chapter 1, we will make use of elementary properties of tensor prod-
ucts of vector spaces and separable field extensions. In Chapter 3, we
shall need at one point the Schur–Zassenhaus theorem (on the exis-
tence and conjugacy of complements of abelian normal subgroups).
In Chapter 4, we assume that the reader is willing to accept a very
deep result (Grothendieck’s trace formula) without proof. In the last
parts of Chapter 4, we assume some familiarity with the charac-
ter theory of finite groups. Otherwise, the exposition is completely
self-contained and proceeds in a fairly elementary and concrete way,
giving ‘full proofs for everything’.
Among the potential readers I imagine a student who wishes
to get a starting point for reading more advanced textbooks or
more specialized research articles on groups of Lie type and their
representations. I could also imagine a lecturer or researcher with a
specific interest in finite simple groups, say, who wishes to have an
easily accessible reference for basic geometric facts about groups of
Preface vii
Lie type. And last but not least, I imagine someone who just wishes
to study some basic aspects in a beautiful area of mathematics,
where algebraic geometry and finite groups (which are, a priori, quite
unrelated areas!) get mixed.
In the elaboration of this book, I have used a number of sources.
As far as algebraic geometry and commutative algebra are concerned,
my favourite references are Atiyah and Macdonald (1969), Mumford
(1988), and Shafarevich (1994). The standard textbooks on algebraic
groups are Borel (1991), Humphreys (1991), and Springer (1998),
which contain much more material than we can cover here. Our
treatment owes a lot to the lecture notes of Steinberg (1974) and
the article by Steinberg (1977), in which Steinberg gave new and, in
many cases, more elementary proofs for standard results on algebraic
groups. For the more specific theory of finite groups of Lie type and
their characters (following Deligne–Lusztig), the standard reference
is Carter (1985). Finally, the work of George Lusztig on characters
of finite groups of Lie type has had the most important influence on
the development of my own understanding of all these matters.
For the experts, let me try to mention some points which I
consider to be the distinctive features of this book.
(1) The running examples in this book are the ‘classical groups’,
that is, groups of matrices which leave a non-degenerate bilinear or
quadratic form invariant. We will not discuss the classification of
such forms, for which one can consult Dieudonné (1971) or Grove
(2002), for example. Here, we introduce three families of groups
(which are precisely the ‘groups of Lie type’ Bm , Cm , Dm ) corre-
sponding to three fixed choices of a symmetric or alternating form.
We will give a complete treatment of these groups in the frame-
work of so-called ‘groups with a BN -pair’, a most useful concept
introduced by Tits in the 1960s. In this approach, for example, the
Dickson pseudodeterminant for orthogonal groups in characteristic 2
is not needed at all; see Section 1.7.
(2) In this book, we exclusively work in the framework of affine
varieties. There is no mention at all of general (quasi-projective)
varieties, sheaves of functions, or such things. We do introduce pro-
jective space, but this will only play an auxiliary role, as a topological
space on which an algebraic group acts. All the properties of pro-
jective varieties, grassmannian varieties, and flag varieties that are
needed can be quite naturally obtained by working directly with
viii Preface
Index 305
1
Algebraic sets and algebraic groups
h
g := fh+1 − xi X dh+1 −di fi ∈ I \ (f1 , . . . , fh ).
i=1
V(I(V )) = V̄ .
where the first equality holds by Remark 1.1.4 and the second equal-
ity holds since each fi is irreducible. Finally, we have (f1 )∩· · ·∩(fr ) =
(f1 · · · fr ) since f1 · · · fr is the least common multiple of f1 , . . . , fr .
Vf := {v ∈ V | f(v) = 0} ⊆ V.
The set Ṽf is algebraic in kn+1 and the projection map πf : Ṽf → V ,
(v, y) → v, defines a bijection between Ṽf and Vf .
all (v, y) ∈ Ṽf . We must show that g ∈ (I, fY − 1). To see this,
r
i=0 gi Y , where gi ∈ k[X1 , . . . , Xn ]; we set
let uswrite g = i
Xα | Xβ ⇒ α β.
LT(f)
f˜ := f − fi .
LT(fi )
f = Y · f1 + 0 · f2 + (−X − Y ) and f = 0 · f1 + X · f2 + 0.
s+n s+n 1
a
HF{0} (s) = = = (s + n)(s + n − 1) · · · (s + 1)
s n n!
n factors
1 n 1 n + 1 n−1
n
1
= s + s + ··· + s + 1.
n! n! 2 i
i=1
Proof First note that the polynomial a HPI (t) is unique, once we
know that it exists. If I = {0} or I = k[X1 , . . . , Xn ], then everything
follows from the explicit computations in Example 1.2.12. So let us
now assume that {0} = I = k[X1 , . . . , Xn ]. Let G be a Groebner
basis of I (with respect to a graded lexicographic order) and write
{LM(g) | g ∈ G} = {X β | β ∈ M }. We set
For any s 0, let C(I)s be the set of all α ∈ C(I) such that
|α| s. We claim that a HFI (s) = |C(I)s | for all s 0. Indeed,
by Lemma 1.2.13, we have a HFI (s) = a HF(LT(I)) (s). Furthermore,
by Example 1.2.9(c), {X β | β ∈ M } is a Groebner basis of (LT(I)).
Hence, by Lemma 1.2.3, we have that dim LT(I)s equals the number
of all α such that |α| s and X β | X α for some β ∈ M . Thus,
we have a HFI (s) = |C(I)s | as required. Now we continue in four
steps.
Step 1. Given a subset J ⊆ {1, . . . , n} and a function τ : J → Z0 ,
we define C(J, τ ) := {(α1 , . . . , αn ) ∈ Zn0 | αj = τ (j) for all j ∈ J}.
Now we claim that there exists a finite collection J of pairs (J, τ ) as
above such that
C(I) = C(J, τ ). (∗)
(J,τ )∈J
we have C(I) = β∈M C(β). Now note that if (J, τ ) and (J , τ ) are
two pairs as above, then
⎧
⎪
⎨∅ if τ (j) = τ (j) for some
C(J, τ ) ∩ C(J , τ ) = j ∈ J ∩ J ,
⎪
⎩
C(J ∪ J , τ0 ) otherwise,
Step 2. We claim that, for any pair (J, τ ) as in Step 1, there exists
a polynomial FJ,τ (t) ∈ Q[t] such that
deg FJ,τ (t) = n − |J| and FJ,τ (s) = |C(J, τ )s |
for all s |τ | := j∈J τ (j). Indeed, let γ = (γ1 , . . . , γn ) ∈ Zn0 be
such that γj = 0 for j ∈ J and γj = τ (j) for j ∈ J. Then C(J, τ )
is the set of all α + γ, where α = (α1 , . . . , αn ) ∈ Zn0 is such that
αj = 0 for j ∈ J. Thus, |C(J, τ )s | is the number of monomials
in the variables Xj (j ∈ J) such that, when multiplied with X γ ,
the total degree is at most s. This is the same as the number of
monomials in d := n − |J| variables of total degree at most s − |γ|
(note that |τ | = |γ|). Hence, as in Example 1.2.12(b), we find that
d + s − |τ | 1 d
|C(J, τ )s | = = (s + combination of lower
d d!
powers of s).
Step 3. If A1 , . . . , Am are finite subsets of some set, then
we have
m
|A1 ∪ · · · ∪ Am | = (−1)r−1 |Ai1 ∩ · · · ∩ Air |;
r=1 1i1 <···<ir m
Using the expression for FJ,τ (t) in Step 2, we now see that the terms
of degree d0 in the above sum do not cancel. Hence a HPI (t) has
degree d0 , and the leading coefficient is of the form a/d0 ! for an inte-
ger a > 0. All the remaining coefficients are also seen to be integer
multiples of d0 !.
Step 4. It remains to prove the characterization of the degree
of a HPI (t) in (a). By the discussion in Step 3, deg a HPI (t) is the
largest d such that there exists a subset J ⊆ {1, . . . , n} of size n − d
and a function τ : J → Z0 such that C(J, τ ) ⊆ C(I). First we show
that I ∩ k[Xj | j ∈ J ] = {0} for any such pair (J, τ ). Assume, if
possible, that this is not the case and let 0 = f ∈ I ∩ k[Xj | j ∈ J ].
But then we also have 0 = LT(f) ∈ (LT(I)) ∩ k[Xj | j ∈ J ]. Writing
LT(f) = aX α , where 0 = a ∈ k and α = (α1 , . . . , αn ) ∈ Zn0 , we have
αj = 0 for j ∈ J , and so γ + α ∈ C(J, τ ) ⊆ C(I) for any γ ∈ C(J, τ ).
On the other hand, we have aX γ+α = X γ LT(f) ∈ (LT(I)), and so
X γ+α is divisible by X β for some β ∈ M , contradicting γ +α ∈ C(I).
Thus, we have
there exist 1 i1 < · · · < id n
deg HPI (t) max 0 d n
a
.
such that I ∩ k[Xi1 , . . . , Xid ] = {0}
Note that the above proof provides in fact an algorithm for com-
puting the polynomial a HPI (t), once we know a Groebner basis for
I with respect to a graded lexicographic order . See Chapter 9 of
Cox et al. (1992) for more details.
20 Algebraic sets and algebraic groups
have I(C) = (X2 − X12 , X3 − X13 ) and Exercise 1.8.4 shows that
k[X1 , X2 , X3 ]/I(C) ∼ = k[Y1 ], where Y1 is an indeterminate. Thus, we
conclude that dim C = 1.
An irreducible algebraic set of dimension 1 is called an affine
curve; we have just seen that the twisted cubic is an affine curve. In
Exercise 1.8.10(b) it is shown that the affine curves in k2 (where k is
algebraically closed) are precisely the algebraic sets defined by one
non-constant irreducible polynomial.
is a closed set and, hence, f also vanishes on ϕ(V ); that is, we have
f ∈ I(ϕ(V )).
ϕ∗ : A[W ] → A[V ], ḡ → ḡ ◦ ϕ.
ˆ ⊆ kn × km = kn+m and
Then consider the algebraic set V̂ = V(I)
the projection πm : k ×k → k . We claim that, if |k| = ∞, then
n m m
= V(S ∪ T ) ⊆ kn+m ,
where we use the identities f(v) = f(v, w) for all f ∈ S and g(w) =
g(v, w) = 0 for all g ∈ T . Thus, the direct product V × W is an
algebraic set in kn+m . Since (I(V ), I(W )) ⊆ I(V × W ), we have a
well-defined k-bilinear map
A[V ] × A[W ] → A[V × W ], (f, ¯ ḡ) → f¯ × ḡ := fg + I(V × W ).
So we get an induced k-linear map A[V ] ⊗k A[W ] → A[V × W ].
see Exercise 1.8.19 for the case n = 3. Now assume that we are in
the case (Alt) where k may have any characteristic. Then n = 2m
must be even, and
⎡ ⎤
0 Qm
Sp2m (k) := Γ2m (Q− −
2m , k), where Q2m :=
⎣ ⎦∈ M2m (k),
−Qm 0
34 Algebraic sets and algebraic groups
2m
fP := pij Xi Xj ∈ k[X1 , . . . , Xn ]
i,j=1
2m
fP (x + y) − fP (x) − fP (y) = (pij + pji )xi yj = βQ2m (x, y), (∗)
i,j=1
[x, [y, z]] + [y, [z, x]] + [z, [x, y]] = 0 for all x, y, z ∈ L.
n
n
Dn (ab) = Di (a)Dn−i (b) for all a, b ∈ A.
i
i=0
n
D= D(Xi ) Di for any D ∈ Derk (R, R).
i=1
« You may carve it on his tombstone, you may cut it on his card,
That a young man married is a young man marred [21] . »
[21]
La Barbe-Bleue.