Lie Groups 1st Edition Luiz Antonio Barrera San Martin - Download The Entire Ebook Instantly and Explore Every Detail
Lie Groups 1st Edition Luiz Antonio Barrera San Martin - Download The Entire Ebook Instantly and Explore Every Detail
com
https://textbookfull.com/product/lie-groups-1st-edition-
luiz-antonio-barrera-san-martin/
OR CLICK HERE
DOWLOAD EBOOK
https://textbookfull.com/product/theory-of-groups-and-symmetries-
finite-groups-lie-groups-and-lie-algebras-1st-edition-alexey-p-isaev/
textbookfull.com
https://textbookfull.com/product/math-845-notes-on-lie-groups-mark-
reeder/
textbookfull.com
https://textbookfull.com/product/introduction-to-the-theory-of-lie-
groups-1st-edition-roger-godement/
textbookfull.com
https://textbookfull.com/product/theory-of-groups-and-symmetries-
representations-of-groups-and-lie-algebras-applications-1st-edition-
alexey-p-isaev/
textbookfull.com
Moon Austin, San Antonio & the Hill Country 6th Edition
Justin Marler
https://textbookfull.com/product/moon-austin-san-antonio-the-hill-
country-6th-edition-justin-marler/
textbookfull.com
https://textbookfull.com/product/differential-geometry-and-lie-groups-
a-computational-perspective-gallier-j/
textbookfull.com
https://textbookfull.com/product/matrix-inequalities-and-their-
extensions-to-lie-groups-first-edition-liu/
textbookfull.com
https://textbookfull.com/product/representations-of-finite-groups-of-
lie-type-2nd-edition-francois-digne/
textbookfull.com
https://textbookfull.com/product/the-character-theory-of-finite-
groups-of-lie-type-a-guided-tour-1st-edition-meinolf-geck/
textbookfull.com
Latin American Mathematics Series
Lie Groups
Latin American Mathematics Series
More information about this series at http://www.springer.com/series/15993
Luiz A.B. San Martin
Lie Groups
Luiz A.B. San Martin
Department of Mathematics—IMECC
State University of Campinas
Campinas, SP, Brazil
Translation from the Portuguese language edition: Grupos de Lie by Luis Antonio Barrera San Martin,
© Editora da Unicamp 2016. Published by Editora da Unicamp. All Rights Reserved. Translated by
José Emílio Maiorino and Carlos Augusto Bassani Varea with revisions by Simon Chiossi.
Mathematics Subject Classification: 22EXX, 17BXX, 22E25, 22E27, 22E45, 22E46, 22E60
This Springer imprint is published by the registered company Springer Nature Switzerland AG.
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
To the future generations,
represented by my grandchildren,
Pedro and João.
Preface
The purpose of this book is to offer an introductory text on Lie groups, by presenting
the theory from its first principles.
The concept of group is by now one of the cornerstones of contemporary
mathematics and its applications. This is due both to its simplicity as an algebraic
structure and also to the fact that the idea of symmetry, in a broad sense, is
formalized through the notion of invariants under transformation groups.
Lie groups form a special class of groups, which are studied using the methods of
differential and integral calculus. As a mathematical object, a Lie group combines
two structures of different flavor: an algebraic group structure on one side and that
of a differentiable manifold on the other. Lie groups began to be investigated around
1870 as groups of symmetries of differential equations and of the several geometries
that had appeared at the time. Since then, the theory of Lie groups, or what is
more generally referred to as Lie theory, has undergone a substantial development.
Through multiple ramifications and incarnations, it has established itself in most
areas of mathematics, both pure and applied.
The techniques for studying Lie groups are based on the construction of their Lie
algebras, which was started by Sophus Lie in the 1870s. (Incidentally, the name “Lie
theory” itself has its roots in this construction.) Once the idea of Lie algebra of a Lie
group has been set up, the core of the method is to establish local or global properties
for Lie groups from the properties of the Lie algebras. This transfer process, so to
speak, is very effective, for it allows to describe the typical nonlinearities of Lie
groups by means of the linear algebra encapsulated by Lie algebras.
The book presents the body of results that set up the relationship between Lie
groups and Lie algebras. It is divided into four parts, plus a fifth part consisting of
appendices.
The heart of the theory of Lie groups—and the classification thereof based on Lie
algebras—is developed in Parts II and III. Part II consists of four chapters, where the
Lie algebra of a Lie group is defined, and incorporates the proofs of the fundamental
relationships between the multiplication in the group and the bracket operation of
the Lie algebra. The main ingredient in these formulas is the exponential map, which
is based on solutions to ordinary differential equations (invariant vector fields).
vii
viii Preface
This part also discusses Lie subgroups of a Lie group and their link to Lie
subalgebras, besides other standard concepts of group theory such as homomor-
phisms, normal subgroups, and quotient spaces. The results of Part II build up to
an existence and uniqueness theorem for Lie groups with the given Lie algebra.
Uniqueness is reserved to Lie groups that satisfy the global topological property of
being connected and simply connected.
Part II does not require a deep knowledge of Lie algebras, since the few concepts
used are introduced when needed. In Part III, instead, Lie algebras come in full
force and enable to portray the complete picture of how Lie algebras determine
Lie groups. This includes, in particular, delicate issues such as the classification of
semi-simple Lie algebras. Although the latter class of Lie algebras is not discussed
exhaustively in this book, an effort was made to indicate the main argumentative
steps, as well as references to full proofs in the existing literature.
Regarding the other parts of the book, Part I is dedicated to general topological
groups. Here, only the first chapter, dealing with topological features groups may
enjoy, is required for reading the rest of the book (Chapter 2). These properties
are satisfied by Lie groups and are widely used throughout. The other two chapters
(on Haar measures and representations of compact groups) rely on rather separate
premises, for they require a different background (measure theory and basic
functional analysis), and they appear in the sequel only tangentially.
Part IV is concerned with the actions of Lie group. The orbits of an action are
investigated in detail after proving that they are differentiable submanifolds. Also
introduced here are the notions of principal bundle and associated bundle, whose
respective fibers are Lie groups and spaces acted on by Lie groups. The final chapter
is an invitation to a broad and very active area of modern differential geometry,
namely that of invariant geometric structures on homogeneous spaces.
Every Part unfolds with an initial detailed outline of the contents to follow. These
summaries provide an overview of the main results in the various chapters and show
how they become interconnected.
The introductory chapter opening the book is meant to survey in an informal and
friendly manner the relationship between Lie groups and Lie algebras, including a
review of the classification of Lie groups.
Readers approaching this matter for the first time are recommended to follow this
chapter sequence: Chapters 2, 5, 6, 7, 8 (up to Section 8.1), 9, 10, 11 (up to Sec-
tion 11.2), and 13, to become acquainted with the foundations of Lie group theory.
Finally, I would like to thank the people who variously contributed to the comple-
tion of this book, be it with suggestions, by pointing out several flaws in preliminary
versions, or by expressing their support. I am particularly grateful to all the students
who took my classes on Lie groups at the University of Campinas over the years.
My deepest gratitude goes to my friends and colleagues Alexandre Santana, Adriano
João da Silva, Carlos Braga Barros, Caio Negreiros, Elizabeth Gasparim, Lino
Grama, Lonardo Rabelo, Lucas Seco, Luciana Alves, Mauro Patrão, Osvaldo do
Rocio, Paolo Piccione, Paulo Ruffino, Pedro Catuogno, and Victor Ayala.
Preface ix
Acknowledgement
The support of the grant 2019/10625-4 from the São Paulo Research Foundation
(Fundação de Amparo à Pesquisa do Estado de São Paulo - FAPESP).
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
xi
xii Contents
Part V Appendices
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367
Chapter 1
Introduction
This introductory chapter has an informal character. Its purpose is to provide the
reader with a panoramic view of the theory developed in this book by discussing
some of the main results through examples which are both concrete and illustrative
and therefore central to the theory.
The formal definition of a Lie group will be presented later, in Chapter 5. For all
effects, a Lie group G is a group whose product
(g, h) ∈ G × G −→ gh ∈ G
is a differentiable map. An example rich enough to cover much of the theory, and
which should always be used as a guide, is the general linear group Gl (n, R). The
elements of this group are the invertible n × n matrices with real entries, or, what
is essentially the same, the invertible linear maps of a finite dimension real vector
space.
In what follows, some aspects of the group Gl (n, R) are discussed. The first
observation is that it is an open subset of the vector space of n × n matrices, that is,
2
of Rn . It has two connected components, determined by the sign of the determinant.
One of them is
n
zij = xik ykj ,
k=1
which is a polynomial map of degree two in the variables xij , yij . Therefore, the
product is a differentiable map. For this reason, Gl (n, R) is a Lie group.
The great strength of Lie group theory is based on the existence of Lie algebras
associated with the groups. Lie algebras make it possible to transfer methods of
linear algebra to the study of nonlinear objects such as Lie groups. A Lie algebra is
a quintessential algebraic structure. It is defined as a vector space g endowed with a
product (bracket) [·, ·] : g × g → g which satisfies the following properties:
1. Bilinearity, that is, [·, ·] is linear in both variables or, equivalently, the bracket
product is distributive with respect to vector space operations.
2. Skew-symmetry, that is, [X, Y ] = −[Y, X], for X, Y ∈ g.
3. Jacobi Identity: For X, Y, Z ∈ g,
The elements of the Lie algebra of a Lie group are ordinary differential
equations (vector fields) in the group; they satisfy a symmetry property arising
from the multiplicative structure of the group (translation invariant vector fields,
see Chapter 5). In turn, the group elements are obtained through the solutions of
these equations, given by their flows. Usually the vector space underlying the Lie
algebra of a Lie group is identified with T1 G, the space of tangent vectors at the
identity 1 ∈ G.
In other words, the Lie algebra is a linear object that approximates the group:
A Lie algebra element is given by the derivative of a curve in the group. The inverse
procedure consists in solving differential equations. For this reason, in the first
decades of development of the theory the expression infinitesimal group was used
instead of Lie algebra.
In the case of Gl (n, R), its Lie algebra is the vector space of n × n matrices with
bracket given by the matrix commutator1
[A, B] = BA − AB.
This Lie algebra will be denoted gl (n, R). In order to establish the relation
between the algebra and the group, consider, for each matrix A ∈ gl (n, R) the
vector field
g → Ag
in the space of matrices. This vector field induces the linear differential equation
dg
= Ag. (1.1)
dt
1 The reverse order in the definition of this commutator is due to the choice of right invariant vector
fields which will soon be made.
1 Introduction 3
dx
This equation is just the linear system = Ax, x ∈ Rn , repeated n times, one
dt
time for each column of matrix g. The fundamental solution of the linear system in
Rn is given by
1
etA = (tA)n ,
n!
n≥0
showing that the solution of Equation (1.1) with initial condition g (0) = 1 (where 1
denotes the n × n identity matrix) is g (t) = etA . This solution is entirely contained
in Gl (n, R) as exponentials are invertible matrices. Moreover, the curve
eX eY = ec(X,Y ) ,
where c (X, Y ) is a series (similar to a Taylor series) containing only X, Y and its
successive brackets. The first terms of the series are
1 1 1
c (X, Y ) = X + Y + [X, Y ] + [[X, Y ], Y ] − [[X, Y ], X] + · · · (1.2)
2 12 12
The remaining terms involve brackets with four or more elements. The series
c (X, Y ) converges if X and Y are small enough, showing that, for such values of
X and Y , the product eX eY is completely determined by the Lie algebra, that is, by
bracket combinations of its elements. Hence, the group product is fully determined
locally, around the origin (that is the neutral element), by the Lie algebra bracket.
This kind of relation between bracket and group product can be extended to the
entire group, allowing us to show that, up to global topological properties (such
as, for example, that the underlying topological space of the group is connected
and simply connected), there exists a unique Lie group associated with a given Lie
algebra.
4 1 Introduction
α (t) = 1 + t 2 [A, B] + · · · ,
where the relevant term is [A, B]. This allows us to see the bracket as the
infinitesimal object associated with the group commutator. Derivatives of this kind
can be extended to general vector fields. This Taylor expansion was at the origin of
the concept of a Lie bracket of vector fields, as it is currently known. The concept
was introduced by Sophus Lie and thus the entire theory got his name.
These formulas, although they are useful to illustrate the relation between Lie
groups and Lie algebras, do not constitute the main technical tools of this theory. The
bridge between Lie groups and Lie algebras is built with the adjoint representations
defined in Chapter 5. Such representations provide formulas relating conjugations
Cg (x) = gxg −1 within the group, their differentials Ad (g), which are linear maps
of the Lie algebra, and the differentials of Ad (g) which are given by the bracket
defining the Lie algebra. In applying those formulas to go from Lie groups to Lie
algebras it is necessary to differentiate twice (possibly different functions). The
reverse process, from Lie algebra to Lie group, involves two integrals which are
generally obtained with the help of theorems on the existence and uniqueness of
solutions of ordinary differential equations. The second derivative in the Taylor
expansion of the conjugation in (1.3) provides an heuristic idea that the passage
from group to algebra takes place by means of two derivatives.
The following are other examples of Lie groups and their respective Lie
algebras:
1. If G is an abelian Lie group, then its Lie algebra is abelian, that is, the bracket
[·, ·] is identically zero (and vice versa, in the case of connected groups, as can
be seen from the Campbell–Hausdorff formula). Connected abelian Lie groups
are described in Chapter 7, Section 7.3.
2. Let
G = O (n) = {g ∈ Gl (n, R) : gg T = g T g = 1}
1 Introduction 5
be the group of orthogonal matrices. Its Lie algebra is the subalgebra of skew-
symmetric matrices:
The bracket in so (n) is the matrix commutator. The reason for this is that A is a
skew-symmetric matrix if and only if etA is an orthogonal matrix for every t ∈ R.
This means that the group O (n) is a submanifold of the space of matrices whose
tangent space at the identity 1 coincides with the subspace of skew-symmetric
matrices.
3. The group Gl (n, C) of invertible n × n complex matrices is a Lie group for the
same reason as Gl (n, R) is. The Lie algebra of Gl (n, C) is gl (n, C), which is
the Lie algebra n × n complex matrices.
The program of Lie theory consists in studying Lie groups through their Lie
algebras, making an effort to describe the geometric and algebraic properties of
the group by means of the corresponding properties of its Lie algebra. Such a
description should ultimately approach structural properties which would allow the
classification of Lie groups in terms of Lie algebras. In this program, two concepts
from abstract group theory play a central role, the concepts of subgroup and of group
homomorphism. Such concepts are thoroughly worked with the help of Lie algebras
and the results are the best possible:
1. Subgroups: If G is a Lie group with Lie algebra g, then the subgroups of G are
in bijective correspondence with the Lie subalgebras of g, with two important
restrictions about the subgroups entering such a bijection. The first one is that
only Lie subgroups which are both subgroups and differentiable submanifolds
are considered, so that their underlying structures make them Lie groups. The
second restriction is that the bijection involves only connected Lie groups. This
must be so because the exponential map sees only the connected component of
the Lie group that contains the identity.
In this bijection, a Lie subalgebra is associated with a Lie subgroup by taking
derivatives (twice, as mentioned above). In the reverse process a Lie subgroup is
obtained from a Lie subalgebra as an integral manifold of a distribution. (For the
bijection, see Chapter 6. The theory of distributions is presented in Appendix B.)
With regard to Lie subgroups, it must be mentioned the famous closed
subgroup theorem by Cartan, which states that if a subgroup of G is a closed
set, then it is automatically a Lie subgroup.
2. Homomorphisms: If φ : G → H is a differentiable homomorphism between
Lie groups, then its differential dφ1 : T1 G → T1 H is a linear map between the
tangent spaces at their identities, which are the Lie algebras g and h of G and
H , respectively. This linear map turns out to be a Lie algebra homomorphism
between g and h. The reverse construction does not work in all generality due to
global topological restrictions on G. It happens that a homomorphism θ : g → h
gives rise to a local homomorphism between groups G and H , defined around the
identity of G and with values in a neighborhood of the identity of H . The only
6 1 Introduction
obstacle for this application to extend to the whole group G is its fundamental
group, so that if G is connected and simply connected, then θ : g → h is the
differential of a homomorphism φ : G → H (see Chapter 7).
A good example of this phenomenon is given by groups (R, +) and S 1 =
{z ∈ C : |z| = 1}. Their Lie algebras are isomorphic (both have dimension 1),
there are homomorphisms R → S 1 (t → eait ), but there are no homomorphisms
S 1 → R. Around their identities, the groups are isomorphic.
These comments on homomorphisms are in line with the above discussion
about the Baker–Campbell–Hausdorff formula, in which it was concluded that
the bracket in the Lie algebra determines the product in the Lie group around its
identity.
This analysis of homomorphisms, especially the theorem on the extension to
simply connected groups, gives rise to the description of all connected Lie groups
on the basis of some classification of Lie algebras. This description comes down to
two points (see Chapter 7):
1. Given a (real, finite dimensional) Lie algebra g there exists a unique connected
and simply connected Lie group G with Lie algebra g. Uniqueness stems from the
extension theorem mentioned above: An isomorphism between the Lie algebras
defines an isomorphism between connected and simply connected Lie groups.
Existence is proved in two steps: (1) the construction of a Lie group G with Lie
algebra isomorphic to g (in Chapter 7 this is done with the help of the theorem
of Ado, which ensures that every Lie algebra is isomorphic to a matrix Lie
subalgebra); (2) the formal construction of a Lie group structure in the universal
covering space G of a Lie group G.
2. Every connected Lie group is equal to the quotient of a simply connected Lie
group G by a discrete subgroup ⊂ G contained in the center of G.
This description works well for connected groups, as these are the groups that
can be accessed by Lie algebras through the solution of differential equations.
An example are the groups of dimension 1. The additive group (R, +) is simply
connected and its Lie algebra is the unique (up to isomorphisms) Lie algebra of
dimension 1. Hence, any connected and simply connected Lie group of dimension
1 is isomorphic to (R, +). A discrete subgroup of R has the form ωZ, with ω > 0.
It thus follows that any group of dimension 1 is isomorphic to R or to R/ωZ ≈ S 1 .
The classification of connected Lie groups is generally done in three steps: 1)
classifying real Lie algebras; 2) determining, for each real Lie algebra g (or rather
for its Lie algebra isomorphism class),
a simply connected Lie group G whose Lie
of G
algebra is g; 3) finding the center Z G and the discrete subgroups ⊂ Z(G.
From this point on, a deeper development of the theory of Lie algebras is
required. Lie algebras are divided into two great classes, solvable algebras and
semi-simple algebras. The Levi decomposition theorem combines these two types of
algebra by means of the semi-direct product, in order to build all finite dimensional
Lie algebras (see Chapter 9). This decomposition of Lie algebras extends to simply
connected Lie groups, in such a way that all that is necessary to do is to determine
1 Introduction 7
separately the simply connected groups for semi-simple and solvable Lie algebras,
respectively.
In the case of solvable algebras, it is possible to prove that the manifolds
underlying connected and simply connected groups are diffeomorphic to Euclidean
spaces Rn . As usual when one deals with solvable algebras, the proof is done
by induction, starting from the group (R, +), of dimension 1, (see Chapter 10).
A typical example is the group of upper triangular matrices
⎛ ⎞
a1 · · · ∗
⎜ .. . . .. ⎟
⎝ . . . ⎠ a1 , . . . , an > 0,
0 · · · an
1. SO (n) = {g ∈ O (n) : det g = 1}, with Lie algebra so (n). These Lie algebras
are simple if n = 2 and n = 4. The Lie algebra so (2) is abelian, while so (4) is
semi-simple and decomposes into two simple components isomorphic to so (3).
The groups SO (n) are not simply connected.
2. SU (n) = {g ∈ Gl (n, C) : gg T = g T g = 1, det g = 1}, with Lie algebra su (n),
which is simple for n ≥ 2. The groups SU (n), n ≥ 2, are simply connected.
3. The group U (n) is defined as SU (n), but without the requirement that the
determinant must be 1. Its Lie algebra is u (n) (defined as su (n), without the
restriction on the trace). Lie algebras u (n) are not semi-simple.
4. Sp (n), with Lie algebra sp (n). These groups are simply connected. Their
elements are given by quaternionic unitary matrices, that is, matrices with entries
in H which satisfy gg T = id.
Simple noncompact algebras are also classified (see the tables in Chapter 12).
The corresponding Lie groups have the property that their underlying manifolds are
diffeomorphic to the cartesian product of a compact group by a Euclidian space
RN . A product of this kind is given by a Cartan decomposition or an Iwasawa
decomposition (Chapter 12). Using this fact, the Levi decomposition theorem and
the information about solvable simply connected groups, the conclusion is that every
connected and simply connected Lie group is diffeomorphic to the cartesian product
of a compact Lie group by a Euclidian space.
The following are examples of noncompact semi-simple groups:
1. Sl (n, R) = {g ∈ Gl (n, R) : det g = 1}.
2. Sl (n, C) = {g ∈ Gl (n, C) : det g = 1}.
3. Sp (n, R) = {g ∈ Gl (2n, R) : gJ g T = 1, det g = 1}, where
0 −1n×n
J = .
1n×n 0
1p×p 0
Ip,q = .
0 −1q×q
1.1 Exercises
1. Find the first three terms of the Baker–Campbell–Hausdorff formula (1.2) for the
linear group Gl (n, R), expanding the product etA etB and grouping terms in t k ,
k = 0, 1, 2, 3.
1.1 Exercises 9
2. Let g be a Lie algebra satisfying [X, [Y, Z]] = 0 for every X, Y, Z ∈ g, so that
the Baker–Campbell–Hausdorff series reduces to
1
c (X, Y ) = X + Y + [X, Y ].
2
Show that the product X ∗ Y = c (X, Y ) defines a group structure on g.
3. Let A be an n×n matrix. If exp A = k≥0 k!1 Ak , show that A is skew-symmetric
(A+AT = 0) if and only if exp tA is an orthogonal matrix for every t ∈ R. (Hint:
consider the curve α (t) = exp tA (exp tA)T .)
4. Let Sl (n, R) = {g ∈ Gl (n, R) : det g = 1} be the group of unimodular matrices.
Assume that Sl (n, R) is a Lie subgroup and verify, using exponentials, that its
Lie algebra is
Assume that SU (2) is a Lie subgroup of invertible matrices and verify, using
exponentials, that its Lie algebra is the space of skew-Hermitian matrices
T
su (2) = {A ∈ M2×2 (C) : A + A = 0, trA = 0}.
Verify that su (2) is a real Lie algebra with dim su (2) = 3 (where the Lie bracket
is given by the matrix commutator). Verify also that su (2) is isomorphic to the
following Lie algebras: (1) so (3) = {A ∈ M3×3 (R) : A + AT = 0} (with the
commutator); (2) R3 endowed with the vector product ∧.
6. Let H = {a + bi + cj + dk : a, b, c, d ∈ R} be the algebra of quaternions. Write
ξ = a + ib + j c + kd as ξ = (a + ib) + j (c − id), that is, ξ = z + j w with
z, w ∈ C. Left multiplication by ξ can be seen as a linear map on C2 . Write the
matrix of this map in basis {1, j } and show that the map
z −w
φ : a + bi + cj + dk = z + j w −→ ∈ M2×2 (C)
w z
Overview
This part consists of three chapters on topological groups. The only chapter whose
reading is essential to the rest of the book is Chapter 2, which deals with the concepts
of group theory from a topological point of view. The purpose of this chapter
is to set up the language to be used throughout the entire theory of Lie groups.
We present the concepts of subgroups (open, closed, etc.), connected components
of topological groups, quotient spaces (which inherit the quotient topology in a
natural way), and quotient topological groups. As quotient spaces, together with
their respective topologies, are closely related to the orbits of continuous actions
of topological groups, we include a discussion about homeomorphisms between
orbits and quotient spaces. The topological concepts presented in this chapter will
be revisited later, within the differentiable universe of Lie groups.
Chapter 3 constructs the Haar measures on locally compact Hausdorff topo-
logical groups. Their uniqueness, up to a scale factor, is demonstrated. The Haar
measure is a central object in the theory of topological groups, particularly Lie
groups, as it allows the use of methods of integral calculus in the study of those
groups. The chapter is entirely developed within the framework of measure theory.
It is independent of the rest of the book, particularly because the construction of the
Haar measure on Lie groups can be done in a technically simpler way by means of
invariant volume forms (as described in Section 5.6). The study of Chapter 3 can
be postponed without loss, except for the information contained in the statement of
Theorem 3.1.
Chapter 4 provides an introduction to the beautiful theory of representations
of compact groups, which generalizes the theory of Fourier series for periodic
functions and which was initially developed by Schur and Weyl at the beginning
of twentieth century. The main results of this chapter are the Schur orthogonality
relations and the Peter–Weyl theorem, which provide a clear description of the space
L2 of a compact group endowed with its Haar measure. In this chapter, we use some
results of functional analysis. Like the chapter on Haar measures, it is not essential
12 I Topological Groups
for the rest of the book except for some initial results on the decomposition of
representations, which will be applied in Chapter 11, a chapter dealing with compact
Lie groups.
Chapter 2
Topological Groups
Several properties of Lie groups depend only on their topology and not on their
differentiable manifold structure. Some of these properties, which are also valid for
more general topological groups, are studied in this chapter. The aim here is not to
present an exhaustive development of the theory of topological groups, but only to
establish a language and to prove some results which are useful for the theory of Lie
groups.
The identity element of a group G will be denoted by 1. Given a subset A ⊂ X
of a topological space, the interior, the closure, and the boundary of A are denoted
by A◦ , A, and ∂A, respectively.
2.1 Introduction
7. Let (C \ {0}, ·) be endowed with the topology generated by the basis of open
sets that is formed by the open intervals of the vertical straight lines ra = {a +
ix ∈ C : x ∈ R}. This group is not a topological group with respect to this
topology. Indeed, the left translation Leiθ is a rotation by an angle θ ∈ R. The
image of the open set r1 = {1 + ix ∈ C : x ∈ R} is not an open set if, for
instance, θ = −π/2.
8. Let G be a topological group and X a topological space. Denote by A (X, G)
the set of continuous maps f : X → G. This set has a group structure with
the product (f g) (x) = f (x) g (x), whose inverse is ι (f ) (x) = f (x)−1 .
Introduce in A (X, G) the compact-open topology, which has as subbasis of
open sets the open sets of the type
gh = (g1 h1 , . . . , gn hn )
by the basis formed by the intervals [a, b), a < b. The product is a continuous
map. To see that, take x, y, such that x + y ∈ [a, b). Then, for some ε > 0,
x + y + ε < b, ensuring that [a, b) contains [x, x + ε/2) + [y, y + ε/2) (=
{z + w : z ∈ [x, x + ε/2) and w ∈ [x, x + ε/2)}). This means that the open
set [x, x + ε/2) × [y, y + ε/2) is contained in p−1 [a, b), showing that p is
continuous. On the other hand, ι (x) = −x is not continuous. For instance,
(−2, −1] = ι−1 [1, 2) is not open.
11. This example illustrates the case of a group G in which the inverse ι (g) = g −1
is continuous and p is partially continuous (that is, G is semitopological),
but not continuous. Take the additive group R2 , + with R2 endowed with
the topology generated by the Siamese balls, which are defined as follows:
Take two balls of equal radius, with centers lying on the same vertical straight
line and which are tangent to each other. The corresponding Siamese ball is
equal to the union of the interior of the two balls together with their point of
tangency. The set of Siamese balls constitutes a basis for the topology. Endowed
with this topology, the inverse in R2 is continuous (by symmetry with respect
to the origin), as well as translations. However, the product p = + is not
continuous. Indeed, (1, 0) + (−1, 0) = (0, 0). Take a Siamese ball B with point
of tangency at (0, 0) and let B1 and B2 be Siamese balls with points of tangency
respectively at (1, 0) and (−1, 0). Then, the set B1 + B2 is not contained in B,
as can be geometrically verified. This means that B1 × B2 is not contained in
p−1 (B). As B, B1 , and B2 are arbitrary elements of the basis of the topology,
it follows that p is not continuous on ((1, 0) , (−1, 0)).
If A is a subset of G and g ∈ G, then the translation Lg (A) is denoted simply
by gA = {gx : x ∈ A}. The fact that translations are homeomorphisms implies that
gA is open or closed if A is open or closed, respectively. The same observation is
valid for right translations Ag. More generally, let B ⊂ G and write
A · B = AB = {xy ∈ G : x ∈ A, y ∈ B}.
By definition, AB = x∈B Ax = x∈A xB. In this way, if A (or B) is open,
then AB is open, as it is a union of open sets. It must be noted
2 that the same
statement is
not valid for closed
sets.
For instance, take
in R , + the closed
1 1
sets A = x, :x>0 ,B = −x, : x > 0 . Then, the sum A + B
x x
is contained in the half-plane y > 0 and, nevertheless, (0, 0) is in the closure of
A + B. In the case in which one of the sets is closed, the following statement is
valid.
Proposition 2.1 If K ⊂ G is compact and F ⊂ G is closed, then KF and F K are
closed.
Proof If x ∈ KF , then x is the limit of a net kα fα with kα ∈ K, fα ∈ F , and
α ∈ D, where D is a directed set. As K is compact, there exists a subnet kαj
2.2 Neighborhoods of Identity 17
such that k = limαj kαj ∈ K, and this implies that k −1 = limαj kα−1 j
by continuity
of the inverse.
Using
now the continuity of the product, it is clear that the subnet
fαj = kα−1j
k f
αj αj converges to f = k −1 x, which belongs to the closed set F , as
— Mitä turhia, tytöt, ei minua tarvitse kehua. Tein vain niin kuin
olisin muiden toivonut tekevän minulle. Teitä naurattaa kun sanon,
että tahdon olla lady, mutta minä tarkoitan, että on oltava hieno
sekä tavoiltaan että mieleltään, ja minä yritän parhaani mukaan. En
osaa oikein selittää, mutta tahtoisin päästä eroon pikkusieluisuudesta
ja typeryydestä, jotka usein ovat naisten vikana. En ole vielä
kovinkaan pitkällä, mutta koetan parhaani ja toivon joskus tulevani
äidin kaltaiseksi.
— Voi äiti kulta, hän on liian nuori; ensin on minun vuoroni. Olen
toivonut sitä jo kauan — se olisi kovin hyödyllistä ja samalla
äärettömän ihanaa — minun täytyy päästä.
— Aina käy näin. Amyn osaksi tulee kaikki huvi ja minun täytyy
vain raataa. Se ei ole reilua, huudahti Jo kuohuksissaan.
— Hm, huokasi Jo, jos sinä toivot sitä, niin saat sen, sillä sinun
toiveesi toteutuvat aina — minun eivät milloinkaan.
— Totta kai!
ULKOMAANKIRJEENVAIHTAJAMME
Lontoossa
Rakas kotiväki.
Keskiyöllä
Amynanne
Pariisissa
Rakkaat tytöt!
Viime kirjeessäni kerroin teille Lontoossa olostamme —
miten ystävällisiä Vaughanit meille olivat ja miten hauskoja
kutsuja he järjestivät. Minä nautin eniten käynnistä Hampton
Courtissa ja Kensington-museossa. Hamptonissa näin Rafaelin
luonnoksia ja museossa Turnerin, Lawrencen, Reynoldsin,
Hogarthin ja muiden suuruuksien maalauksia.
Heidelbergissä
Rakas äiti.
Oma Amysi.
9
HELLIÄ HUOLIA
— Hän istuu usein yksikseen eikä enää juttele isän kanssa niin
paljon kuin ennen. Eräänä päivänä näin hänen itkevän pienokaisten
vuoteen ääressä. Hän laulaa vain surullisia lauluja ja joskus hänen
kasvoillaan on outo ilme, jota en ymmärrä. Se ei ole Bethin tapaista
ja se huolestuttaa minua.
— Sitä ei voi auttaa, äiti kulta, sinun täytyy vain alistua kantamaan
huolia ja antaa poikasten lentää pesästä. Minä lupaan, etten
milloinkaan lennä kauas, jos se sinua yhtään lohduttaa.
Vaikka Laurie liehitteli Amya ja laski leikkiä Jon kanssa, hän oli
aina käyttäytynyt Bethiä kohtaan erikoisen hellästi ja ystävällisesti.
Mutta kaikkihan olivat Bethille kilttejä, eikä kukaan siksi ollut tullut
ajatelleeksi, että Laurie pitäisi Bethistä enemmän kuin muista.
Päinvastoin perheessä oli viime aikoina vakiintunut se käsitys, että
'poikakulta' alkoi yhä enemmän pitää Josta, vaikka tämä ei sietänyt
kuulla puhuttavankaan asiasta, vaan suuttui silmittömästi, jos joku
uskalsi vihjata siihen. Jos he olisivat olleet selvillä kuluneen vuoden
hellistä kohtauksista tai pikemminkin niiden yrityksistä — ne oli näet
tukahdutettu alkuunsa — he olisivat suureksi tyydytyksekseen
voineet huomauttaa: — Enkös minä sanonut. Mutta Jo vihasi
hakkailua ja hänellä oli aina valmiina leikkipuhe tai hymy vaaran
uhatessa.
— Kas niin, Jo, älä nyt piikittele. Kun mies on koko viikon
ahertanut, niin että on kuin luuranko, häntä voisi kyllä pitää hyvänä
näin lauantai-iltana.
— Mitä minä voin tehdä, kun järkevät tytöt, joista minä välitän
monen dollarin verran, eivät anna minun lähettää kukkia eivätkä
muutakaan? Täytyyhän minun tunteitteni jollain tavoin purkautua.
— Ei, tämä on uusi; kyllä minä kestän sen, ja Beth koetti hillitä
itkuaan.
— Kerro minulle, niin minä parannan sinut niin kuin niin monesti
ennenkin.
— Ei, ei, älä hae, älä kerro hänelle mitään. Kohta tuntuu
paremmalta. Tule tänne makaamaan ja silitä minun päätäni. Sitten
minä rauhoitun ja nukahdan, ihan varmasti.
— Ei nyt, ei vielä.
— Tuntuuko jo paremmalta?
Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.
textbookfull.com