Full download test bank at ebook textbookfull.

com

Linear Algebra M. Thamban Nair

CLICK LINK TO DOWLOAD

https://textbookfull.com/product/linear-
algebra-m-thamban-nair/

textbookfull
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Linear Algebra Seymour Lipschutz

https://textbookfull.com/product/linear-algebra-seymour-
lipschutz/

Basic Linear Algebra Jürgen Müller

https://textbookfull.com/product/basic-linear-algebra-jurgen-
muller/

Elementary Linear Algebra Ron Larson

https://textbookfull.com/product/elementary-linear-algebra-ron-
larson/

Linear Algebra G. Shanker Rao

https://textbookfull.com/product/linear-algebra-g-shanker-rao/
Elementary Linear Algebra 12th Edition Howard Anton

https://textbookfull.com/product/elementary-linear-algebra-12th-
edition-howard-anton/

SF1624 Algebra and Geometry Introduction to Linear

Algebra for Science Engineering Daniel Norman

https://textbookfull.com/product/sf1624-algebra-and-geometry-
introduction-to-linear-algebra-for-science-engineering-daniel-
norman/

Linear Algebra Done Right Third Edition Sheldon Axler

https://textbookfull.com/product/linear-algebra-done-right-third-
edition-sheldon-axler/

Fundamentals of linear algebra and optimization Gallier

J.

https://textbookfull.com/product/fundamentals-of-linear-algebra-
and-optimization-gallier-j/

Elementary Linear Algebra 1st Edition James R. Kirkwood

https://textbookfull.com/product/elementary-linear-algebra-1st-
edition-james-r-kirkwood/
M. Thamban Nair · Arindama Singh

Linear
Algebra
Linear Algebra
M. Thamban Nair Arindama Singh
•

Linear Algebra

123
M. Thamban Nair Arindama Singh
Department of Mathematics Department of Mathematics
Indian Institute of Technology Madras Indian Institute of Technology Madras
Chennai, Tamil Nadu, India Chennai, Tamil Nadu, India

ISBN 978-981-13-0925-0 ISBN 978-981-13-0926-7 (eBook)

https://doi.org/10.1007/978-981-13-0926-7
Library of Congress Control Number: 2018944333

© Springer Nature Singapore Pte Ltd. 2018

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 Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721,
Singapore
Preface

Linear Algebra deals with the most fundamental ideas of mathematics in an abstract
but easily understood form. The notions and techniques employed in Linear
Algebra are widely spread across various topics and are found in almost every
branch of mathematics, more prominently, in Differential Equations, Functional
Analysis, and Optimization, which have wide applications in science and engi-
neering. The ideas and techniques from Linear Algebra have a ubiquitous presence
in Statistics, Commerce, and Management where problems of solving systems of
linear equations come naturally. Thus, for anyone who carries out a theoretical or
computational investigation of mathematical problems, it is more than a necessity to
equip oneself with the concepts and results in Linear Algebra, and apply them with
confidence.

Overview and Goals

This book provides background materials which encompass the fundamental

notions, techniques, and results in Linear Algebra that form the basis for analysis
and applied mathematics, and thereby its applications in other branches of study. It
gives an introduction to the concepts that scientists and engineers of our day use to
model, to argue about, and to predict the behaviour of systems that come up often
from applications. It also lays the foundation for the language and framework for
modern analysis. The topics chosen here have shown remarkable persistence over
the years and are very much in current use.
The book realizes the following goals:
• To introduce to the students of mathematics, science, and engineering the
elegant and useful abstractions that have been created over the years for solving
problems in the presence of linearity
• To help the students develop the ability to form abstract notions of their own and
to reason about them

v
vi Preface

• To strengthen the students’ capability of carrying out formal and rigorous

arguments about vector spaces and maps between them
• To make the essential elements of linear transformations and matrices accessible
to not-so-matured students with a little background in a rigorous way
• To lead the students realize that mathematical rigour in arguing about linear
objects can be very attractive
• To provide proper motivation for enlarging the vocabulary, and slowly take the
students to a deeper study of the notions involved
• To let the students use matrices not as static array of numbers but as dynamic
maps that act on vectors
• To acquaint the students with the language and powers of the operator theoretic
methods used in modern mathematics at present

Organization

Chapter 1 lays the foundation of Linear Algebra by introducing the notions of

vector spaces, subspaces, span, linear independence, basis and dimension, and the
quotient space. It is done at a leisurely pace. Sometimes, the steps have been made
elaborate intentionally so that students would not jump to conclusions and develop
the habit of hand waving. Both finite and infinite dimensional vector spaces are
discussed so that a proper foundation is laid for taking up Functional Analysis as a
natural extension, if the student so desires to pursue in his later years of study.
Chapter 2 introduces linear transformations as structure-preserving maps
between vector spaces. It naturally leads to treating the matrices as linear trans-
formations and linear transformations as matrices. Particular attention is given to
bijective linear transformations and how they act on bases. So, change of bases is
considered by looking at the identity linear transformation. The space of linear
transformations is introduced along with the composition of maps. It raises the issue
of equivalence and similarity of matrices leading to the rank theorem.
Chapter 3 deals with elementary operations in detail. Starting from the com-
putation of rank, elementary operations are used to evaluate determinants, compute
the inverse of a matrix, and solve linear systems. The issue of solvability of linear
systems is treated via linear transformations showing to the students how abstrac-
tion helps.
Chapter 4 brings in the so far neglected notion of direction, or angle between
vectors by introducing inner products. The notion of orthogonality and the neces-
sary elegance and ease it ensues are discussed at length. The geometric notion of an
orthogonal projection is given a lead for solving minimization problems such as the
best approximation of a vector from a subspace and least squares solutions of linear
systems. Constructing a linear functional from the inner product via Fourier
expansion leads to Riesz representation theorem. The existence of the adjoint of a
linear transformation is also shown as an application of Riesz representation.
Preface vii

Chapter 5 asks a question of how and when a linear operator on a vector space
may fix a line while acting on the vectors. This naturally leads to the concepts of
eigenvalues and eigenvectors. The notion of fixing a line is further generalized to
invariant subspaces and generalized eigenvectors. It gives rise to polynomials that
annihilate a linear operator, and the ascent of an eigenvalue of a linear operator.
Various estimates involving the ascent, the geometric, and algebraic multiplicities
of an eigenvalue are derived to present a clear view.
Chapter 6 takes up the issue of representing a linear operator as a matrix by using
the information on its eigenvalues. Starting with diagonalization, it goes for Schur
triangularization, block-diagonalization, and Jordan canonical form characterizing
similarity of matrices.
Chapter 7 tackles the spectral representation of linear operators on inner product
spaces. It proves the spectral theorem for normal operators in a finite-dimensional
setting, and once more, that of self-adjoint operators with somewhat a different
flavour. It also discusses the singular value decomposition and polar decomposition
of matrices that have much significance in application.

Special Features

There are places where the approach has become non-conventional. For example, the
rank theorem is proved even before elementary operations are introduced; the relation
between ascent, geometric multiplicity, and algebraic multiplicity are derived in the
main text, and information on the dimensions of generalized eigenspaces is used to
construct the Jordan form. Instead of proving results on matrices, first a result of the
linear transformation is proved, and then it is interpreted for matrices as a particular
case. Some of the other features are:
• Each definition is preceded by a motivating dialogue and succeeded by one or
more examples
• The treatment is fairly elaborate and lively
• Exercises are collected at the end of the section so that a student is not distracted
from the main topic. The sole aim of these exercises is to reinforce the notions
discussed so far
• Each chapter ends with a section listing problems. Unlike the exercises at the
end of each section, these problems are theoretical, and sometimes unusual and
hard requiring the guidance of a teacher
• It puts emphasis on the underlying geometric idea leading to specific results
noted down as theorems
• It lays stress on using the already discussed material by recalling and referring
back to a similar situation or a known result
• It promotes interactive learning building the confidence of the student
viii Preface

• It uses operator theoretic method rather than the elementary row operations. The
latter is primarily used as a computational tool reinforcing and realizing the
conceptual understanding

Target Audience

This is a textbook primarily meant for a one- or two-semester course at the junior
level. At IIT Madras, such a course is offered to master’s students, at the fourth year
after their schooling, and some portions of this are also offered to undergraduate
engineering students at their third semester. Naturally, the problems at the end of
each chapter are tried by such master’s students and sometimes by unusually bright
engineering students.

Notes to the Instructor

The book contains a bit more than that can be worked out (not just covered) in a
semester. The primary reason is: these topics form a prerequisite for undertaking
any meaningful research in analysis and applied mathematics. The secondary rea-
son is the variety of syllabi followed at universities across the globe. Thus different
courses on Linear Algebra can be offered by giving stress on suitable topics and
mentioning others. The authors have taught different courses at different levels from
it sticking to the core topics.
The core topics include vector spaces, up to dimension (Sects. 1.1–1.5), linear
transformation, up to change of basis (Sects. 2.1–2.5), a quick review of determi-
nant (Sect. 3.5), linear equations (Sect. 3.6), inner product space, up to orthogonal
and orthonormal bases (Sects. 4.1–4.5), eigenvalues and eigenvectors, up to
eigenspaces (Sects. 5.1–5.3), the characteristic polynomial in Sect. 5.5, and diag-
onalizability in Sect. 6.1. Depending on the stress in certain aspects, some of the
proofs from these core topics can be omitted and other topics can be added.

Chennai, India M. Thamban Nair

March 2018 Arindama Singh
Contents

1 Vector Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Vector Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Subspaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.3 Linear Span . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.4 Linear Independence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
1.5 Basis and Dimension . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
1.6 Basis of Any Vector Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
1.7 Sums of Subspaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
1.8 Quotient Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
1.9 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
2 Linear Transformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
2.1 Linearity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
2.2 Rank and Nullity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
2.3 Isomorphisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
2.4 Matrix Representation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
2.5 Change of Basis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
2.6 Space of Linear Transformations . . . . . . . . . . . . . . . . . . . . . . . . . 93
2.7 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
3 Elementary Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
3.1 Elementary Row Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
3.2 Row Echelon Form . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
3.3 Row Reduced Echelon Form . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
3.4 Reduction to Rank Echelon Form . . . . . . . . . . . . . . . . . . . . . . . . 128
3.5 Determinant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
3.6 Linear Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
3.7 Gaussian and Gauss–Jordan Elimination . . . . . . . . . . . . . . . . . . . 147
3.8 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156

ix
x Contents

4 Inner Product Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163

4.1 Inner Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
4.2 Norm and Angle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
4.3 Orthogonal and Orthonormal Sets . . . . . . . . . . . . . . . . . . . . . . . . 172
4.4 Gram–Schmidt Orthogonalization . . . . . . . . . . . . . . . . . . . . . . . . 176
4.5 Orthogonal and Orthonormal Bases . . . . . . . . . . . . . . . . . . . . . . . 182
4.6 Orthogonal Complement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
4.7 Best Approximation and Least Squares . . . . . . . . . . . . . . . . . . . . 190
4.8 Riesz Representation and Adjoint . . . . . . . . . . . . . . . . . . . . . . . . 196
4.9 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
5 Eigenvalues and Eigenvectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
5.1 Existence of Eigenvalues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
5.2 Characteristic Polynomial . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
5.3 Eigenspace . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
5.4 Generalized Eigenvectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224
5.5 Two Annihilating Polynomials . . . . . . . . . . . . . . . . . . . . . . . . . . 234
5.6 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238
6 Block-Diagonal Representation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
6.1 Diagonalizability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
6.2 Triangularizability and Block-Diagonalization . . . . . . . . . . . . . . . 248
6.3 Schur Triangularization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
6.4 Jordan Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
6.5 Jordan Normal Form . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264
6.6 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
7 Spectral Representation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
7.1 Playing with the Adjoint . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
7.2 Projections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
7.3 Normal Operators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
7.4 Self-adjoint Operators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
7.5 Singular Value Decomposition . . . . . . . . . . . . . . . . . . . . . . . . . . 308
7.6 Polar Decomposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317
7.7 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
About the Authors

M. Thamban Nair is a professor of mathematics at the Indian Institute of

Technology Madras, Chennai, India. He completed his Ph.D. at the Indian Institute
of Technology Bombay, Mumbai, India, in 1986. His research interests include
functional analysis and operator theory, specifically spectral approximation, the
approximate solution of integral and operator equations, regularization of inverse
and ill-posed problems. He has published three books, including a textbook,
Functional Analysis: A First Course (PHI Learning), and a text-cum-monograph,
Linear Operator Equations: Approximation and Regularization (World Scientific),
and over 90 papers in reputed journals and refereed conference proceedings. He has
guided six Ph.D. students and is an editorial board member of the Journal of
Analysis and Number Theory, and Journal of Mathematical Analysis. He is a life
member of academic bodies such as Indian Mathematical Society and Ramanujan
Mathematical Society.

Arindama Singh is a professor of mathematics at the Indian Institute of

Technology Madras, Chennai, India. He completed his Ph.D. at the Indian Institute
of Technology Kanpur, India, in 1990. His research interests include knowledge
compilation, singular perturbation, mathematical learning theory, image processing,
and numerical linear algebra. He has published five books, including Elements of
Computation Theory (Springer), and over 47 papers in reputed journals and ref-
ereed conference proceedings. He has guided five Ph.D. students and is a life
member of many academic bodies, including Indian Society for Industrial and
Applied Mathematics, Indian Society of Technical Education, Ramanujan
Mathematical Society, Indian Mathematical Society, and The Association of
Mathematics Teachers of India.

xi
Chapter 1
Vector Spaces

1.1 Vector Space

A vector in the plane is an object with certain length and certain direction. Con-
ventionally it is represented by an arrow with an initial point and an endpoint; the
endpoint being the arrow head. We work with plane vectors by adding them, sub-
tracting one from the other, and by multiplying them with a number. We see that the
plane vectors have a structure, which is revealed through the two operations, namely
addition and multiplication by a number, also called scalar multiplication. These
operations can be seen in an alternate way by identifying the vectors with points in
the plane. The identification goes as follows.
Since only length and direction matter and not exactly the initial or the endpoints,
we may think of each vector having its initial point at the origin. The endpoint
can then be identified with the vector itself. With O as the origin with Cartesian
coordinates (0, 0) and P as the point with Cartesian coordinates (a, b), the vector
−→
O P is identified with the point (a, b) in the plane

R2 = {(α, β) : α ∈ R, β ∈ R}.

Then the familiar parallelogram law for addition of vectors translates to component-
wise addition. If u, v are vectors with initial point (0, 0) and endpoints (a, b) and
(c, d), respectively, then the vector u + v has initial point (0, 0) and endpoint (a +
c, b + d). Similarly, for a real number α, the vector αu has the initial point (0, 0)
and endpoint (αa, αb).

© Springer Nature Singapore Pte Ltd. 2018 1

M. T. Nair and A. Singh, Linear Algebra,
https://doi.org/10.1007/978-981-13-0926-7_1
2 1 Vector Spaces

Thus, (−1) u, which equals (−a, −b), represents the additive inverse −u of the
vector u; the direction of −u is opposite to that of u. Now, the plane is simply viewed
as a set of all plane vectors.
Similarly, in the three-dimensional space, you may identify a vector with a point
by first translating the vector to have its initial point as the origin and its arrow
head as the required point. The sum of two vectors in three dimensions gives rise
to the component-wise sum of two points. A real number α times a vector gives
a vector whose components are multiplied by α. That is, if u = (a1 , b1 , c1 ) and
v = (a2 , b2 , c2 ), then

u + v = (a1 + a2 , b1 + b2 , c1 + c2 ), αu = (αa1 , αb1 , αc1 ).

Notice that the zero vector, written as 0, is identified with the point (0, 0, 0), and the
vector −u = (−a1 , −b1 , −c1 ) satisfies u + (−u) = 0.
The notion of a vector space is an abstraction of the familiar set of vectors in two
or three dimensions. The idea is to keep the familiar properties of addition of vectors
and multiplication of a vector by a scalar. The set of scalars can be any field. For
obtaining interesting geometrical results, we may have to restrict the field of scalars.
In this book, the field F denotes either the field R of real numbers or the field C of
complex numbers.

Definition 1.1 A vector space over F is a nonempty set V along with two operations,
namely
(a) addition, which associates each pair (x, y) of elements x, y ∈ V with a unique
element in V , denoted by x + y, and
(b) scalar multiplication, which associates each pair (α, x), for α ∈ F and x ∈ V ,
with a unique element in V , denoted by αx,
satisfying the following conditions:
(1) For all x, y ∈ V, x + y = y + x.
(2) For all x, y, z ∈ V, (x + y) + z = x + (y + z).
(3) There exists an element in V , called a zero vector, denoted by 0, such that for
all x ∈ V, x + 0 = x.
1.1 Vector Space 3

(4) For each x ∈ V , there exists an element in V , denoted by −x, and called an
additive inverse of x, such that x + (−x) = 0.
(5) For all α ∈ F and for all x, y ∈ V, α(x + y) = αx + αy.
(6) For all α, β ∈ F and for all x ∈ V, (α + β)x = αx + βx.
(7) For all α, β ∈ F and for all x ∈ V, (αβ)x = α(βx).
(8) For all x ∈ V, 1x = x.

Elements of F are called scalars, and elements of a vector space V are called vec-
tors. A vector space V over R is called a real vector space, and a vector space over C
is called a complex vector space. As a convention, we shorten the expression “a vec-
tor space over F” to “a vector space”. We denote vectors by the letters u, v, w, x, y, z
with or without subscripts, and scalars by the letters a, b, c, d, α, β, γ , δ with or
without subscripts.
You have ready-made examples of vector spaces. The plane

R2 = {(a, b) : a, b ∈ R}

and the familiar three-dimensional space

R3 = {(a, b, c) : a, b, c ∈ R}

are real vector spaces. Notice that R is a vector space over R, and C is a vector
space over C as well as over R. Before presenting more examples of vector spaces,
we observe some subtleties about the conditions (3) and (4) in Definition 1.1. It is
unusual to write a particular symbol such as 0 for all zero vectors. It is also unusual to
write −x for all additive inverses of x. The philosophical hurdle will be over once we
prove that a zero vector is unique and an additive inverse of a vector is also unique.

Theorem 1.2 In any vector space the following statements are true:
(1) There exists exactly one zero vector.
(2) Each vector has exactly one additive inverse.

Proof Let V be a vector space.

(1) Suppose 0 and 0̃ in V are zero vectors. Then for all x ∈ V, x + 0 = x and
x + 0̃ = x. Using the condition (1) in Definition 1.1, we have

0̃ = 0̃ + 0 = 0 + 0̃ = 0.

(2) Let x ∈ V . Let x  and 

x be additive inverses of x. Let 0 be the zero vector.
Then x + x  = 0 and x + x = 0. Therefore,


x = x + (x + x  ) = (
x +0 = x + x) + x  = (x + 
x ) + x  = 0 + x  = x  + 0 = x . 

Theorem 1.2 justifies the use of the symbols 0 for the zero vector and −x for the
additive inverse of the vector x. Of course, we could have used any other symbol,
4 1 Vector Spaces

say, θ for the zero vector and x  for the additive inverse of x; but the symbols 0 and
−x follow the custom. Note that −0 = 0. We also write y − x instead of y + (−x)
for all vectors x and y.
Notice the double meanings used in Definition 1.1. The addition of scalars as well
as of vectors is denoted by the same symbol +, and the multiplication of scalars
as well as of a vector with a scalar is written by just concatenating the elements.
Similarly, 0 denotes the zero vector as well as the scalar zero. Even the notation for
the additive inverse of vector x is −x; just the way we write −α for the additive
inverse of a scalar α. You should get acquainted with the double meanings.
It is easy to check that in every vector space V over F,

0 + x = x, x + (y − x) = y for all x, y ∈ V.

Every vector space contains at least one element, the zero vector. On the other
hand, the singleton {0} is a vector space; it is called the zero space or the trivial
vector space. In general, we will be concerned with nonzero vector spaces, which
contain nonzero elements. A nonzero vector space is also called a nontrivial vector
space.
In a vector space, addition of two elements is allowed. This is generalized by
induction to a sum of any finite number of vectors. But an infinite sum of vectors is
altogether a different matter; it requires analytic notions such as convergence.

Example 1.3 In the following, the sets along with the specified addition and scalar
multiplication are vector spaces. (Verify.)
(1) Consider the set Fn of all n-tuples of scalars, that is,

Fn := {(a1 , . . . , an ) : a1 , . . . , an ∈ F}.

We assume that two elements in Fn are equal when their respective components are
equal. For x = (a1 , . . . , an ), y = (b1 , . . . , bn ) ∈ Fn , and α ∈ F, define the addition
and scalar multiplication component-wise, that is,

x + y := (a1 + b1 , . . . , an + bn ), αx := (αa1 , . . . , ααn ).

Then Fn is a vector space with

0 = (0, . . . , 0) and − (a1 , . . . , an ) = (−a1 , . . . , −an ).

(2) We use the notation Fm×n for the set of all m × n matrices with entries from F.
A matrix A ∈ Fm×n is usually written as
⎡ ⎤
a11 · · · a1n
⎢ .. .. ⎥ ,
A=⎣ . . ⎦
am1 · · · amn
1.1 Vector Space 5

or as A = [ai j ] for short, with ai j ∈ F for i = 1, . . . , m; j = 1, . . . , n. The number

ai j which occurs at the entry in ith row and jth column is referred to as the (i, j)th
entry of the matrix [ai j ]. For A = [ai j ] and B = [bi j ] in Fm×n , and α ∈ F, we define

A + B = [ai j + bi j ] ∈ Fm×n , α A = [αai j ] ∈ Fm×n .

We say that two matrices are equal when their respective entries are equal. That is,
for A = [ai j ] and B = [bi j ], we write A = B if and only if ai j = bi j . With these
operations of addition and scalar multiplication, Fm×n becomes a vector space over
F. The zero vector in Fm×n is the zero matrix, i.e. the matrix with all entries 0, and
the additive inverse of A = [ai j ] ∈ Fm×n is the matrix −A := [−ai j ].
(3) For n ∈ {0, 1, 2, . . .}, let Pn (F) denote the set of all polynomials (in the variable
t) of degree at most n, with coefficients in F. That is, x ∈ Pn (F) if and only if x is
of the form
x = a0 + a1 t + · · · + an t n

for some scalars a0 , a1 , . . . , an . Here, we assume that a scalar is a polynomial of

degree 0. Further, two polynomials are considered equal when the coefficients of
respective powers of t are equal. That is,

a0 + a1 t + · · · + an t n = b0 + b1 t + · · · + bn t n if and only if ai = bi for i = 0, . . . , n.

Addition and scalar multiplication on Pn (F) are defined as follows. For

x = a0 + a1 t + · · · + an t n , y = b0 + b1 t + · · · + bn t n in Pn (F), and α ∈ F,

x + y := (a0 + b0 ) + (a1 + b1 )t + · · · + (an + bn )t n ,

αx := αa0 + αa1 t + · · · + αan t n .

The zero vector in Pn (F) is the polynomial with all its coefficients zero, and

−(a0 + a1 t + · · · + an t n ) = −a0 − a1 t − · · · − an t n .

Then Pn (F) is a vector space.

(4) Let P(F) = ∪∞ n=0 Pn (F), the set of all polynomials with coefficients in F. That
is, x ∈ P(F) if and only if x = a0 + a1 t + · · · + an t n for some n ∈ {0, 1, 2, . . .} and
for some scalars a0 , . . . , an ∈ F.
If x, y ∈ P(F), then x ∈ Pm (F) and y ∈ Pn (F) for some m, n. So, x, y ∈ Pk (F),
with k = max {m, n}. The equality relation, addition, and scalar multiplication are
defined as in (3). Then P(F) is a vector space.
(5) Let V be the set of all sequences of scalars. A sequence whose nth term is an is
written as (an ). Two sequences are considered equal if and only if their respective
terms are equal, that is, (an ) = (bn ) if and only if an = bn for each n ∈ N. For
(an ), (bn ) ∈ V , and α ∈ F, define
6 1 Vector Spaces

(an ) + (bn ) := (an + bn ), α(an ) := (αan ).

That is,

(a1 , a2 , a3 , . . .) + (b1 , b2 , b3 , . . .) = (a1 + b1 , a2 + b2 , a3 + b3 , . . .),

α(a1 , a2 , a3 . . . .) = (αa1 , αa2 , αa3 , . . .).

With this addition and scalar multiplication, V is a vector space, where its zero vector
is the sequence with each term as zero, and −(an ) = (−an ). This space is called the
sequence space and is denoted by F∞ .
(6) Let S be a nonempty set. Let V be a vector space over F. Let F(S, V ) be the set of
all functions from S into V . As usual, x = y for x, y ∈ F(S, V ) when x(s) = y(s)
for each s ∈ S. For x, y ∈ F(S, V ) and α ∈ F, define x + y and αx point-wise; that
is,
(x + y)(s) := x(s) + y(s), (αx)(s) := αx(s) for s ∈ S.

Let the functions 0 and −x in F(S, V ) be defined by

0(s) = 0, (−x)(s) = −x(s) for s ∈ S.

Then F(S, V ) is a vector space over F with the zero vector as 0 and the additive
inverse of x as −x. We sometimes refer to this space as a function space. 

Comments on Notation: Pn (R) denotes the real vector space of all polynomials of
degree at most n with real coefficients. Pn (C) denotes the complex vector space of
all polynomials of degree at most n with complex coefficients. Similarly, P(R) is the
real vector space of all polynomials with real coefficients, and P(C) is the complex
vector space of all polynomials with complex coefficients. Note that C is also a vector
space over R. Similarly, Pn (C) and P(C) are vector spaces over R. More generally,
if V is a complex vector space, then it is also a real vector space. If at all we require
to regard any vector space over C also as a vector space over R, we will specifically
mention it.
As particular cases of Example 1.3(2), (Read: Example 1.3 Part 2) we have the
vector spaces Fm×1 , the set of all column vectors of size m, and F1×n , the set of
all row vectors of size n. To save space, we use the transpose notation in writing
a column vector. That is, a column vector v of size n with its entries a1 , . . . , an is
written as ⎡ ⎤
a1
⎢ .. ⎥
⎣ . ⎦ or as [a1 · · · an ] .
T

an

By putting the superscript T over a row vector v we mean that the column vector is
obtained by taking transpose of the row vector v. When the column vectors are writ-
ten by the lower case letters u, v, w, x, y, z with or without subscripts (sometimes
1.1 Vector Space 7

superscripts), the corresponding row vectors will be written with the transpose nota-
tion, that is, as u T , v T . Further, we will not distinguish between the square brackets
and the parentheses. Usually, we will write a row vector with parentheses. Thus, we
will not distinguish between

[a1 · · · an ] and (a1 , . . . , an ).

Thus we regard Fn same as F1×n . We may recall that by taking transpose of a matrix
in Fm×n , we obtain a matrix in Fn×m . That is, if A = [ai j ] ∈ Fm×n , then

A T := [b ji ] ∈ Fn×m with b ji = ai j for i = 1, . . . , m; j = 1, . . . , n.

Many vector spaces can be viewed as function spaces. For example, with S = N
and V = F, we obtain the sequence space of Example 1.3(5). With S = {1, . . . , n}
and V = F, each function in F(S, F) can be specified by an n-tuple of its function
values. Therefore, the vector space F({1, . . . , n}, F) can be viewed as Fn and also
as Fn×1 . Some more examples of function spaces follow.

Example 1.4 (1) Let I be an interval, and let C(I, R) denote the set of all real-valued
continuous functions defined on I . For x, y ∈ C(I, R) and α ∈ R, define x + y and
αx point-wise as in Example 1.3(6).
The functions x + y and αx are in C(I, R). Then C(I, R) is a real vector space
with the zero element as the zero function and the additive inverse of x ∈ C(I, R) as
the function −x defined by (−x)(t) = −x(t) for all t ∈ I.
(2) Let R([a, b], R) denote the set of all real-valued Riemann integrable functions
on [a, b]. Define addition and scalar multiplication point-wise, as in Example 1.3(6).
From the theory of Riemann integration, it follows that if x, y ∈ R([a, b], R) and
α ∈ R, then x + y, αx ∈ R([a, b], R). It is a real vector space.
(3) For k ∈ N, let C k ([a, b], F) denote the set of all functions x from [a, b] to F such
that the kth derivative x (k) exists and is continuous on [a, b].
Define addition and scalar multiplication point-wise, as in Example 1.3(6). Then
C k ([a, b], F) is a vector space. Notice that

C k ([a, b], R) ⊆ C([a, b], R) ⊆ R([a, b], R) ⊆ F([a, b], R). 

Example 1.5 Let V1 , . . . , Vn be vector spaces over F. Consider the Cartesian product

V = V1 × · · · × Vn = {(x1 , . . . , xn ) : x1 ∈ V1 , . . . , xn ∈ Vn }.

Define addition and scalar multiplication on V by

(x1 , . . . , xn ) + (y1 , . . . , yn ) := (x1 + y1 , . . . , xn + yn ),

α(x1 , . . . , xn ) := (αx1 , . . . , αxn ).

8 1 Vector Spaces

In V , take the zero vector as (0, . . . , 0) and −(x1 , . . . , xn ) = (−x1 , . . . , −xn ).

Here, the addition in the expression xi + yi is the addition operation defined in Vi and
the scalar multiplication in αxi is the scalar multiplication defined in Vi . Similarly,
in (0, . . . , 0), the ith component is the zero vector of Vi and −xi in (−x1 , . . . , −xn )
is the additive inverse of xi in Vi .
With these operations, V is a vector space. We call it the product space of
V1 , . . . , Vn . 

To illustrate Example 1.5, consider V1 = R and V2 = P1 (R). Then V1 × V2 con-

sists of vectors of the form (a, α + βt), and the operations of addition and scalar
multiplication are defined by

(a1 , α1 + β1 t) + (a2 , α2 + β2 t) = (a1 + a2 , (α1 + α2 ) + (β1 + β2 )t),

c(a, α + βt) = (ca, cα + cβt).

Similarly, the space Fn is a product space with each Vi as F.

Some easy consequences of Definition 1.1 are listed in the following theorem.

Theorem 1.6 Let V be a vector space over F. Let x, y, z ∈ V and let α, β ∈ F.

(1) 0x = 0.
(2) α0 = 0.
(3) (−1)x = −x.
(4) −(−x) = x.
(5) If x + z = y + z, then x = y.
(6) If αx = 0, then α = 0 or x = 0.
(7) If α = β and x = 0, then αx = βx.

Proof (1) 0x = (0 + 0)x = 0x + 0x. Adding −0x, we have 0 = 0x.

(2) α0 = α(0 + 0) = α0 + α0. Adding −α0, we obtain 0 = α0.
(3) x + (−1)x = (1 + (−1))x = 0x = 0, by (1). Adding −x we get (−1)x = −x.
(4) From (3), it follows that −(−x) = (−1)(−1)x = x.
(5) Suppose that x + z = y + z. Then (x + z) + (−z) = (y + z) + (−z). But
(x + z) + (−z) = x + (z + (−z)) = x + 0 = x. Similarly, (y + z) + (−z) = y.
Therefore, x = y.
(6) Suppose that αx = 0. If α = 0, then α −1 ∈ F. Multiplying α −1 , we have α −1 αx =
α −1 0. That is, x = 0.
(7) Suppose that α = β and x = 0. If αx = βx, then (α − β)x = 0. By (6), x = 0.
It leads to a contradiction. Hence, αx = βx. 
1.1 Vector Space 9

Theorem 1.6 allows us to do algebraic manipulation of vectors like scalars as long

as addition is concerned. Thus we abbreviate x + (−y) to x − y. However, there is
a big difference: vectors cannot be multiplied or raised to powers, in general. Look
at the proof of Theorem 1.6(6). We used α −1 instead of x −1 .
It follows from Theorem 1.6(7) that every nontrivial vector space contains infinitely
many vectors.
Exercises for Sect. 1.1
In the following a set V , a field F, which is either R or C, and the operations of
addition and scalar multiplication are given. Check whether V is a vector space over
F, with these operations.
1. With addition and scalar multiplication as in R2 ,
(a) V = {(a, b) ∈ R2 : 2a + 3b = 0}, F = R.
(b) V = {(a, b) ∈ R2 : a + b = 1}, F = R.
(c) V = {(a, b) ∈ R2 : ab = 0}, F = R.
2. V = R2 , F = R, with addition as in R2 , and scalar multiplication as given by the
following: for (a, b) ∈ V, α ∈ R,
(a) α(a, b) := (a, 0).
(b) α(a, b) := (b, αa).
(c) α(a, b) := (αa, −αb).
(0, 0) if α = 0
(d) α(a, b) :=
(αa, b/α) if α = 0.
3. V = {x ∈ R : x > 0}, F = R, and for x, y ∈ V, α ∈ R, x + y := x y, αx :=
x α . The operations on the left are defined by the known operations on the right.
4. V = {x ∈ R : x ≥ 0}, F = R, and for x, y ∈ V, α ∈ R, x + y := x y, αx :=
|α|x.
5. V = C2 , F = C, and for x = (a, b), y = (c, d), α ∈ C,

x + y := (a + 2c, b + 3d), αx := (αa, αb).

6. V is the set of all polynomials of degree 5 with real coefficients, F = R, and the
operations are the addition and scalar multiplication of polynomials.
7. S is a nonempty set, s ∈ S, V is the set of all functions f : S → R with f (s) = 0,
F = R, and the operations are the addition and scalar multiplication of functions.
8. V is the set of all functions f : R → C satisfying f (−t) = f (t), F = R, and the
operations are the addition and scalar multiplication of functions.
9. V = {x}, where x is some symbol, and addition and scalar multiplication are
defined as x + x = x, αx = x for all α ∈ F.
10 1 Vector Spaces

1.2 Subspaces

A subset of a vector space may or may not be a vector space. It will be interesting
if a subset forms a vector space over the same underlying field and with the same
operations of addition and scalar multiplication inherited from the given vector space.
If U is a subset of a vector space V (over the field F), then the operations of addition
and scalar multiplication in U inherited from V are defined as follows:
Let x, y ∈ U, α ∈ F. Consider x, y as elements of V. The vector x + y in V
is the result of the inherited addition of x and y in U. Similarly, the vector αx
in V is the result of the inherited scalar multiplication of α with x in U.
In order that the operations of addition (x, y) → x + y and scalar multiplication
(α, x) → αx are well-defined operations on U , we require the vectors x + y and
αx to lie in U . This condition is described by asserting that U is closed under the
inherited operations.

Definition 1.7 Let V be a vector space over F. A nonempty subset U of V is called

a subspace of V if U is a vector space over F with respect to the operations of
addition and scalar multiplication inherited from V.

To show that a nonempty subset U of a vector space V is a subspace, one must

first verify that U is closed under the inherited operations. The closure conditions
can be explicitly stated as follows:
For all x, y ∈ U and for each α ∈ F, x + y ∈ U and αx ∈ U.
Here, of course, the operations are the operations in the given vector space V over
F. Surprisingly, these closure conditions are enough for establishing that a subset is
a subspace, as the following theorem shows.

Theorem 1.8 Let U be a nonempty subset of a vector space V . Then U is a subspace

of V if and only if U is closed under addition and scalar multiplication inherited
from V.

Proof If U is a subspace of V , then x + y ∈ U and αx ∈ U for all x, y ∈ U and for

all α ∈ F.
Conversely, suppose that x + y ∈ U and αx ∈ U for all x, y ∈ U and for all
α ∈ F. Since U is nonempty, let u ∈ U. By Theorem 1.6, 0 = 0 u ∈ U , and for each
x ∈ U, − x = (−1)x ∈ U. Since U ⊆ V , 0 acts as the zero vector of U , and −x acts
as the additive inverse of x in U. Therefore, the conditions (3)–(4) in Definition 1.1
are satisfied for U . The remaining conditions hold since elements of U are elements
of V as well. 

Notice that the closure conditions in Theorem 1.8 can be replaced by the following
single condition:
For each scalar α ∈ F and for all x, y ∈ U , x + αy ∈ U.
1.2 Subspaces 11

We may also infer from the proof of Theorem 1.8 that

if U is a subspace of a vector space V , then the zero vector of U is the same as

the zero vector of V , and for each u ∈ U , its additive inverse −u as an element
of U is the same as −u in V.

Therefore, in order that U is a subspace of V , the zero vector of V must be in U.

Example 1.9 (1) Let U = {(a, b) ∈ R2 : b = 0} ⊆ R2 . That is, U = {(a, 0) : a ∈

R}. Addition and scalar multiplication, which are defined component-wise in R2 ,
are also operations on U since for all a, b, c, α ∈ R,

(a, 0) + (b, 0) = (a + b, 0) ∈ U, α(c, 0) = (αc, 0) ∈ U.

By Theorem 1.8, U is a subspace of R2 . Notice that the zero vector of U is the same
(0, 0) and −(a, 0) = (−a, 0) as in R2 .
(2) The set U = {(a, b) ∈ R2 : 2a + 3b = 0} is a subspace of R2 (Verify).
(3) Let Q denote the set of√all rational numbers. Q is not a subspace of the real vector
space R since 1 ∈ Q but 2 · 1 ∈ / Q. Similarly, Q2 is not a subspace of R2 .
(4) Consider C as a complex vector space. Let U = {a + i 0 : a ∈ R}. We see that
1 ∈ U but i · 1 = i ∈
/ U. Therefore, U is not a subspace of C.
However, if we consider C as real vector space, then U is a subspace of C. In this
sense, U = R is a subspace of the real vector space C.
(5) Consider the spaces Pm (F) and Pn (F), where m ≤ n. Each polynomial of degree
at most m is also a polynomial of degree at most n. Thus, Pm (F) ⊆ Pn (F). Further,
Pm (F) is closed under the operations of addition and scalar multiplication inherited
from Pn (F). So, Pm (F) is a subspace of Pn (F) for any m ≤ n.
Also, for each n ∈ N, Pn (F) is a subspace of P(F).
(6) In Examples 1.4(1)–(2), both C([a, b], R) and R([a, b], R) are vector spaces.
Since C([a, b], R) ⊆ R([a, b], R) and the operations of addition and scalar multipli-
cation in C([a, b], R) are inherited from R([a, b], R), we conclude that C([a, b], R)
is a subspace of R([a, b], R).
(7) Consider C k ([a, b], F) of Example 1.4(3). For all α ∈ F, x, y ∈ C k ([a, b], F), we
have x + y ∈ C k ([a, b], F) and αx ∈ C k ([a, b], F). By Theorem 1.8, C k ([a, b], F) is
a subspace of C([a, b], F).
(8) Given α1 , . . . , αn ∈ F, U = {(b1 , . . . , bn ) ∈ Fn : α1 b1 + · · · + αn bn = 0} is a
subspace of Fn . When (α1 , . . . , αn ) is a nonzero n-tuple, the subspace U is a hyper-
plane passing through the origin in n dimensions. This terminology is partially bor-
rowed from the case of F = R and n = 3, when the subspace {(b1 , b2 , b3 ) ∈ R3 :
α1 b1 + α2 b2 + α3 b3 = 0, α1 , α2 , α3 ∈ R} of R3 is a plane passing through the ori-
gin. However,

W = {(b1 , . . . , bn ) ∈ Fn : α1 b1 + · · · + αn bn = 1, α1 , . . . , αn ∈ F}

is not a subspace of Fn , since (0, . . . , 0) ∈

/ W.
12 1 Vector Spaces

(9) Let P([a, b], R) be the vector space P(R) where each polynomial is considered as
a function from [a, b] to R. Then the space P([a, b], R) is a subspace of C k ([a, b], R)
for each k ≥ 1. 

Consider two planes passing through the origin in R3 . Their intersection is a

straight line passing through the origin. In fact, intersection of two subspaces is a
subspace. We prove a more general result.

Theorem 1.10 Let C be any collection of subspaces of a vector space V. Let U be

the intersection of all subspaces in C. Then U is a subspace of V.

Proof Let x, y ∈ U and let α ∈ F. Then x, y ∈ W for each W ∈ C. Since W is a

subspace of V , αx ∈ W and x + y ∈ W for each W ∈ C. Then αx ∈ U and x + y ∈
U. By Theorem 1.8, U is a subspace of V. 

In contrast, union of two subspaces need not be a subspace. For, consider

U := {(a, b) ∈ R2 : b = a}, V := {(a, b) ∈ R2 : b = 2a}

as subspaces of R2 . Here, (1, 1), (1, 2) ∈ U ∪ V but (1, 1) + (1, 2) = (2, 3) ∈

/U∪
V . Hence U ∪ V is not a subspace of R2 .
Union of two subspaces can fail to be a subspace since addition of a vector from
one with a vector from the other may not be in the union. What happens to the set of
all vectors of the form x + y, where x is a vector from one subspace and y is from
another subspace? To answer this question, we introduce the notion of a (finite) sum
of subsets of a vector space.

Definition 1.11 Let S1 , . . . , Sn be nonempty subsets of a vector space V. Their sum

is defined by

S1 + · · · + Sn := {x1 + · · · + xn ∈ V : xi ∈ Si , i = 1, . . . , n}.

As expected, sum of two subspaces is a subspace. And, the proof can be general-
ized easily to any finite sum of subspaces.

Theorem 1.12 Let V1 , . . . , Vn be subspaces of a vector space V. Then V1 + · · · + Vn

is a subspace of V.
n n
Proof Let x, y ∈ V1 + · · · + Vn and let α ∈ F. Then, x = i=1 xi and y = i=1 yi
for some xi , yi ∈ Vi , i = 1, . . . , n. Since each Vi is a subspace of V , xi + yi ∈ Vi
and αxi ∈ Vi . Then
n n n
x+y= i=1 xi + i=1 yi = i=1 (x i + yi ) ∈ V1 + · · · + Vn ,
n n
αx = α i=1 xi = i=1 (αx i ) ∈ V1 + · · · + Vn .

Therefore, V1 + · · · + Vn is a subspace of V. 
1.2 Subspaces 13

Example 1.13 Consider the planes:

V1 = {(a, b, c) ∈ R3 : a + b + c = 0}, V2 = {(a, b, c) ∈ R3 : a + 2b + 3c = 0}.

Both V1 and V2 are subspaces of R3 . Their intersection is the subspace

V1 ∩ V2 = {(a, b, c) ∈ R3 : a + b + c = 0 = a + 2b + 3c}.

The condition a + b + c = 0 = a + 2b + 3c is equivalent to b = −2a, c = a.

Hence
V1 ∩ V2 = {(a, −2a, a) : a ∈ R}

which is a straight line through the origin. Both V1 ∩ V2 and V1 + V2 are subspaces
of R3 . In this case, we show that V1 + V2 = R3 . For this, it is enough to show that
R3 ⊆ V1 + V2 . It requires to express any (a, b, c) ∈ R3 as (a1 + a2 , b1 + b2 , c1 + c2 )
for some (a1 , b1 , c1 ) ∈ V1 and (a2 , b2 , c2 ) ∈ V2 . This demands determining the six
unknowns a1 , b1 , c1 , a2 , b2 , c2 from the five linear equations

a1 + a2 = a, b1 + b2 = b, c1 + c2 = c, a1 + b1 + c1 = 0, a2 + 2b2 + 3c2 = 0.

It may be verified that with

a1 = −a − 2b − 2c, b1 = a + 2b + c, c1 = c
a2 = 2a + 2b + 2c, b2 = −a − b − c, c2 = 0,

the five equations above are satisfied. Thus, (a1 , b1 , c1 ) ∈ V1 , (a2 , b2 , c2 ) ∈ V2 , and
(a, b, c) = (a1 + b1 + c1 ) + (a2 + b2 + c2 ) as desired. 

Exercises for Sect. 1.2

1. In the following, check whether the given subset U is a subspace of V. Assume

the usual operations of addition and scalar multiplication along with a suitable
field R or C.
(a) V = R2 , U is any straight line passing through the origin.
(b) V = R2 , U = {(a, b) : b = 2a − 1}.
(c) V = R3 , U = {(a, b, c) : 2a − b − c = 0}.
(d) V = P3 (R), U = {a + bt + ct 2 + dt 3 : c = 0}.
(e) V = P3 (C), U = {a + bt + ct 2 + dt 3 : a = 0}.
(f) V = P3 (C), U = {a + bt + ct 2 + dt 3 : b + c + d = 0}.
(g) V = P3 (R), U = { p ∈ V : p(0) = 0}.
(h) V = P3 (C), U = { p ∈ V : p(1) = 0}.
(i) V = P3 (C), U = {a + bt + ct 2 + dt 3 : a, b, c, d integers }.
(j) V = C([−1, 1], R), U = { f ∈ V : f is an odd function}.
(k) V = C([0, 1], R), U = { f ∈ V : f (t) ≥ 0 for all t ∈ [0, 1]}.
14 1 Vector Spaces

(l) V = C k [a, b], for k ∈ N, U = P[a, b], the set of all polynomials considered
as functions on [a, b].
(m) V = C([0, 1], R), U = { f ∈ V : f is differentiable}.
2. For α ∈ F, let Vα = {(a, b, c) ∈ F3 : a + b + c = α}. Show that Vα is a subspace
of F3 if and only if α = 0.
3. Give an example of a nonempty subset of R2 which is closed under addition
and under additive inverse (i.e. if u is in the subset, then so is −u), but is not a
subspace of R2 .
4. Give an example of a nonempty subset of R2 which is closed under scalar mul-
tiplication but is not a subspace of R2 .
5. Suppose U is a subspace of V and V is a subspace of W. Show that U is a
subspace of W.
6. Give an example of subspaces of C3 whose union is not a subspace of C3 .
7. Show by a counter-example that if U + W = U + X for subspaces U, W, X of
V , then W need not be equal to X.
8. Let m ∈ N. Does the set {0} ∪ {x ∈ P(R) : degree of x is equal to m} form a
subspace of P(R)?
9. Prove that the only nontrivial proper subspaces of R2 are straight lines passing
through the origin.
10. Let U = {(a, b) ∈ R2 : a = b}. Find a subspace V of R2 such that U + V = R2
and U ∩ V = {(0, 0)}. Is such a V unique?
11. Let U be the subspace of P(F) consisting of all polynomials of the form at 3 + bt 7
for a, b ∈ F. Find a subspace V of P(F) such that U + V = P(F) and U ∩ V =
{0}.
12. Let U and W be subspaces of a vector space V. Prove the following:
(a) U ∪ W = V if and only if U = V or W = V.
(b) U ∪ W is a subspace of V if and only if U ⊆ W or W ⊆ U.
13. Let U = {A ∈ Fn×n : A T = A} and let W = {A ∈ Fn×n : A T = −A}. Matrices
in U are called symmetric matrices, and matrices in W are called skew-symmetric
matrices. Show that U and W are subspaces of Fn×n , Fn×n = U + W , and U ∩
W = {0}.

1.3 Linear Span

The sum of vectors x1 , . . . , xn can be written as x1 + · · · + xn , due to Property (2)

of vector addition in Definition 1.1. We may also multiply the vectors with scalars
and then take their sum. That is, we write the sum
n
(α1 x1 ) + · · · + (αn xn ) as α1 x1 + · · · + αn xn , and also as αi xi .
i=1
1.3 Linear Span 15

If the jth term in α1 x1 + · · · + αn xn is absent, that is, if we want to consider

the sum (α1 x1 + · · · + αn xn ) − α j x j for some j ∈ {1, . . . , n}, then we write it as
n
i=1, i= j αi x i and also as

α1 x1 + · · · + α j−1 x j−1 + α j+1 x j+1 + · · · αn xn ,

with the understanding that if j = 1, then the above sum is α2 x2 + · · · + αn xn , and

if j = n, then the sum is equal to α1 x1 + · · · + αn−1 xn−1 .

Definition 1.14 Let V be a vector space.

(a) A vector v ∈ V is called a linear combination of vectors u 1 , . . . , u n in V if
v = α1 u 1 + · · · + αn u n for some scalars α1 , . . . , αn .
(b) The linear span or the span of any nonempty subset S of V is the set of all
linear combinations of finite number of vectors from S; it is denoted by span(S).
We define span(∅) as {0}.

In view of the above definition, for any nonempty subset S of V , we have

x ∈ span(S) if and only if there exist n ∈ N, vectors u 1 , . . . , u n in S, and scalars

α1 , . . . , αn ∈ F such that x = α1 u 1 + · · · + αn u n .

It follows that

span(S) = {α1 u 1 + · · · + αn u n : n ∈ N, α1 , . . . , αn ∈ F, u 1 , . . . , u n ∈ S}.

Moreover, for u 1 , . . . , u n in V ,

span({u 1 , . . . , u n }) = {α1 u 1 + · · · + αn u n : α1 , . . . , αn ∈ F}.

We also write span({u 1 , . . . , u n }) as span{u 1 , . . . , u n }. Informally, we say that

span(S) is the set of all linear combinations of elements of S, remembering the
special case that span(∅) = 0.
By a linear combination, we always mean a linear combination of a finite number
of vectors. As we know an expression of the form α1 v1 + α2 v2 + · · · for vectors
v1 , v2 , . . . and scalars α1 , α2 , . . . has no meaning in a vector space, unless there is
some additional structure which may allow infinite sums.
In what follows, we will be freely using Kronecker’s delta defined as follows:

1 if i = j
δi j = for i, j ∈ N.
0 if i = j

Example 1.15 (1) In R3 , consider the set S = {(1, 0, 0), (0, 2, 0), (0, 0, 3), (2, 1, 3)}.
A linear combination of elements of S is a vector in the form

α1 (1, 0, 0) + α2 (0, 2, 0) + α3 (0, 0, 3) + α4 (2, 1, 3)

16 1 Vector Spaces

for some scalars α1 , α2 , α3 , α4 . Since span(S) is the set of all linear combinations
of elements of S, it contains all vectors that can be expressed in the above form. For
instance, (1, 2, 3), (4, 2, 9) ∈ span(S) since

(1, 2, 3) = 1(1, 0, 0) + 1(0, 2, 0) + 1(0, 0, 3),

(4, 2, 9) = 4(1, 0, 0) + 1(0, 2, 0) + 3(0, 0, 3).

Also, (4, 2, 9) = 1(0, 0, 3) + 2(2, 1, 3). In fact, span(S) = R3 since (α, β, γ ) =

α(1, 0, 0) + (β/2)(0, 2, 0) + (γ /3)(0, 0, 3).
(2) In R3 , span{(3, 0, 0)} is the set of all scalar multiples of (3, 0, 0). It is the straight
line L joining the origin and the point (3, 0, 0). Notice that
√
L = span{(1, 0, 0)} = span{( 2, 0, 0), (π, 0, 0), (3, 0, 0)} = span(L).

It can be seen that span of any two vectors not in a straight line containing the
origin is the plane containing those two vectors and the origin.
(3) For each j ∈ {1, . . . , n}, let e j be the vector in Fn whose jth coordinate is 1 and all
other coordinates are 0, that is, e j = (δ1 j , . . . , δn j ). Then for any (α1 , . . . , αn ) ∈ Fn ,
we have
(α1 , . . . , αn ) = α1 e1 + · · · + αn en .

Thus, span{e1 , . . . , en } = Fn . Also, for any k with 1 ≤ k < n,

span{e1 , . . . , ek } = {(α1 , . . . , αn ) ∈ Fn : α j = 0 for j > k}.

(4) Consider the vector spaces P(F) and Pn (F). Define the polynomials u j := t j−1
for j ∈ N. Then Pn (F) is the span of {u 1 , . . . , u n+1 }, and P(F) is the span of
{u 1 , u 2 , . . .}.
(5) Let V = F∞ , the set of all sequences with scalar entries. For each n ∈ N, let en
be the sequence whose nth term is 1 and all other terms are 0, that is,

en = (δn1 , δn2 , . . .).

Then span{e1 , e2 , . . .} is the space of all scalar sequences having only a finite number
of nonzero terms. This space is usually denoted by c00 (N, F), also as c00 . Notice that
c00 (N, F) = F∞ . 

Clearly, if S is a nonempty subset of a vector space V , then S ⊆ span(S). For,

if x ∈ S then x = 1 · x ∈ span(S). However, span(S) need not be equal to S. For
instance, S = {(1, 0)} ⊆ R2 and span(S) = {(α, 0) : α ∈ R} = S.
1.3 Linear Span 17

Theorem 1.16 Let S be a subset of a vector space V. Then the following statements
are true:
(1) span(S) is a subspace of V , and it is the smallest subspace containing S.
(2) span(S) is the intersection of all subspaces of V that contain S.

Proof (1) First, we show that span(S) is a subspace of V. If S = ∅, then span(∅) =

{0} is a subspace of V that contains ∅. So, suppose that S = ∅. Trivially, S ⊆
span(S). To show that span(S) is a subspace of V , let x, y ∈ span(S) and let α ∈ F.
Then
x = a1 x1 + · · · + an xn , y = b1 y1 + · · · + bm ym

for some vectors x1 , . . . , xn , y1 , . . . , ym in S and scalars a1 , . . . , an , b1 , . . . , bm in

F. Then

x + y = a1 x1 + · · · + an xn + b1 y1 + · · · + bm ym ∈ span(S),
αx = αa1 x1 + · · · + αan xn ∈ span(S).

By Theorem 1.8, span(S) is a subspace of V.

For the second part of the statement in (1), let U be a subspace of V containing
S. Let v ∈ span(S). There exist vectors v1 , . . . , vn ∈ S and scalars α1 , . . . , αn such
that v = α1 v1 + · · · + αn vn . Since v1 , . . . , vn ∈ U and U is a subspace of V , v ∈ U.
Hence, span(S) ⊆ U. That is, span(S) is a subset of every subspace that contains S.
Therefore, span(S) is the smallest subspace of V containing S.
(2) Let W be the intersection of all subspaces of V that contain S. Since span(S) is
a subspace of V , W ⊆ span(S). Also, W is a subspace of V containing S; thus, due
to (1), span(S) ⊆ W. 

Theorem 1.16 implies that taking span of a subset amounts to extending the subset
to a subspace in a minimalistic way.
Some useful consequences of the notion of span are contained in the following
theorem.

Theorem 1.17 Let S, S1 and S2 be subsets of a vector space V. Then the following
are true:
(1) S = span(S) if and only if S is a subspace of V.
(2) span(span(S)) = span(S).
(3) If S1 ⊆ S2 , then span(S1 ) ⊆ span(S2 ).
(4) span(S1 ) + span(S2 ) = span(S1 ∪ S2 ).
(5) If x ∈ S, then span(S) = span{x} + span(S\{x}).

Proof (1) Since span(S) is a subspace of V , the condition S = span(S) implies that
S is a subspace of V. Conversely, if S is a subspace of V , then the minimal subspace
containing S is S. By Theorem 1.16, span(S) = S.
18 1 Vector Spaces

(2) As span(S) is a subspace of V , by (1), span(span(S)) = span(S).

(3) Suppose S1 ⊆ S2 . If S1 = ∅, the conclusion is obvious. Suppose S1 = ∅. Any
linear combination of vectors from S1 is an element of span(S2 ). Hence span(S1 ) ⊆
span(S2 ).
(4) If S1 = ∅ or S2 = ∅, then the statement is true trivially. Assume that S1 = ∅
and S2 = ∅. Let x ∈ span(S1 ) + span(S2 ). Then x is a linear combination of vectors
from S1 plus a linear combination of vectors from S2 . That is, x is a linear combination
of vectors from S1 ∪ S2 . Therefore,

span(S1 ) + span(S2 ) ⊆ span(S1 ∪ S2 ).

Conversely, suppose x ∈ span(S1 ∪ S2 ). Then x is a linear combination of vectors

from S1 ∪ S2 . If such a linear combination uses vectors only from S1 , then x ∈
span(S1 ) ⊆ span(S1 ) + span(S2 ). Similarly, if such a linear combination uses vectors
from only S2 , then x ∈ span(S2 ) ⊆ span(S1 ) + span(S2 ). Otherwise, x is equal to a
linear combinations of vectors from S1 plus a linear combination of vectors from S2 .
In that case, x ∈ span(S1 ) + span(S2 ). Therefore,

span(S1 ∪ S2 ) ⊆ span(S1 ) + span(S2 ).

(5) This follows from (4) by taking S1 = {x} and S2 = S\{x}. 

As a corollary of Theorem 1.17(4), we obtain the following.

Theorem 1.18 If V1 and V2 are subspaces of a vector space V , then V1 + V2 =

span(V1 ∪ V2 ).

Theorems 1.16 and 1.18 show that in extending the union of two subspaces to a
subspace could not have been better; the best way is to take their sum.

Definition 1.19 Let S be a subset of a vector space V. If span(S) = V , then we say

that S spans V , S is a spanning set of V , and also V is spanned by S.
In case, S = {u 1 , . . . , u n } is a finite spanning set of V , we also say that the vectors
u 1 , . . . , u n span V , and that V is spanned by the vectors u 1 , . . . , u n .

Any vector space spans itself. But there can be much smaller subsets that also
span the space as Example 1.15 shows. Note that both ∅ and {0} are spanning sets
of the vector space {0}.
If S is a spanning set of V and x ∈ span(S\{x}), then S\{x} is also a spanning
set of V. For, in this case, the vector x is a linear combination of some vectors from
S\{x}, and if any vector in V is a linear combination, where x appears, we can replace
this x by its linear combination to obtain v as a linear combination of vectors where
x does not appear.
Another random document with
no related content on Scribd:
and clear into an earthen basin. Our bread and vegetables, in an
open basket of palm-blades, are suspended beside it, and the roof of
the cabin supports our poultry-yard and pigeon-house. Sometimes
(but not often) a leg of mutton may be seen hanging from the ridge-
pole, which extends over the deck as a support to the awning.
The cabin, or Mansion of the Executive Powers, is about twenty-
five feet long. Its floor is two feet below the deck, and its ceiling five
feet above, so that we are not cramped or crowded in any particular.
Before the entrance is a sort of portico, with a broad, cushioned seat
on each side, and side-awnings to shut out the sun. This place is
devoted to pipes and meditation. We throw up the awnings, let the
light pour in on all sides, and look out on the desert mountains while
we inhale the incense of the East. Our own main cabin is about ten
feet long, and newly painted of a brilliant blue color. A broad divan,
with cushions, extends along each side, serving as a sofa by day,
and a bed by night. There are windows, blinds, and a canvas cover
at the sides, so that we can regulate our light and air as we choose.
In the middle of the cabin is our table and two camp stools, while
shawls, capotes, pistols, sabre and gun are suspended from the
walls. A little door at the further end opens into a wash-room, beyond
which is a smaller cabin with beds, which we have alloted to
Achmet’s use. Our cook sleeps on deck, with his head against the
provision chest. The raïs and pilot sleep on the roof of our cabin,
where the latter sits all day, holding the long arm of the rudder, which
projects forward over the cabin from the high end of the stern.
Our manner of life is simple, and might even be called
monotonous, but we have never found the greatest variety of
landscape and incident so thoroughly enjoyable. The scenery of the
Nile, thus far, scarcely changes from day to day, in its forms and
colors, but only in their disposition with regard to each other. The
shores are either palm-groves, fields of cane and dourra, young
wheat, or patches of bare sand, blown out from the desert. The
villages are all the same agglomerations of mud-walls, the tombs of
the Moslem saints are the same white ovens, and every individual
camel and buffalo resembles its neighbor in picturesque ugliness.
The Arabian and Libyan Mountains, now sweeping so far into the
foreground that their yellow cliffs overhang the Nile, now receding
into the violet haze of the horizon, exhibit little difference of height,
hue, or geological formation. Every new scene is the turn of a
kaleidoscope, in which the same objects are grouped in other
relations, yet always characterized by the most perfect harmony.
These slight, yet ever-renewing changes, are to us a source of
endless delight. Either from the pure atmosphere, the healthy life we
lead, or the accordant tone of our spirits, we find ourselves unusually
sensitive to all the slightest touches, the most minute rays of that
grace and harmony which bathes every landscape in cloudless
sunshine. The various groupings of the palms, the shifting of the blue
evening shadows on the rose-hued mountain walls, the green of the
wheat and sugar-cane, the windings of the great river, the
alternations of wind and calm—each of these is enough to content
us, and to give every day a different charm from that which went
before. We meet contrary winds, calms and sand-banks without
losing our patience, and even our excitement in the swiftness and
grace with which our vessel scuds before the north-wind is mingled
with a regret that our journey is drawing so much the more swiftly to
its close. A portion of the old Egyptian repose seems to be infused
into our natures, and lately, when I saw my face in a mirror, I thought
I perceived in its features something of the patience and resignation
of the Sphinx.
Although, in order to enjoy this life as much as possible, we
subject ourselves to no arbitrary rules, there is sufficient regularity in
our manner of living. We rise before the sun, and after breathing the
cool morning air half an hour, drink a cup of coffee and go ashore for
a walk, unless the wind is very strong in our favor. My friend, who is
an enthusiastic sportsman and an admirable shot, takes his fowling-
piece, and I my sketch-book and pistols. We wander inland among
the fields of wheat and dourra, course among the palms and acacias
for game, or visit the villages of the Fellahs. The temperature, which
is about 60° in the morning, rarely rises above 75°, so that we have
every day three or four hours exercise in the mild and pure air. My
friend always brings back from one to two dozen pigeons, while I,
who practise with my pistol on such ignoble game as hawks and
vultures, which are here hardly shy enough to shoot, can at the best
but furnish a few wing feathers to clean our pipes.
It is advisable to go armed on these excursions, though there is no
danger of open hostility on the part of the people. Certain
neighborhoods, as that of Beni Hassan, are in bad repute, but the
depredations of the inhabitants, who have been disarmed by the
Government, are principally confined to thieving and other petty
offences. On one occasion I fell in with a company of these people,
who demanded my tarboosh, shoes and shawl, and would have
taken them had I not been armed. In general, we have found the
Fellahs very friendly and well disposed. They greet us on our
morning walks with “Salamàt!” and “Sàbah el Kheyr!” and frequently
accompany us for miles. My friend’s fowling-piece often brings
around him all the men and boys of a village, who follow him as long
as a pigeon is to be found on the palm-trees. The certainty of his
shot excites their wonder. “Wallah!” they cry; “every time the Howadji
fires, the bird drops.” The fact of my wearing a tarboosh and white
turban brings upon me much Arabic conversation, which is
somewhat embarrassing, with my imperfect knowledge of the
language; but a few words go a great way. The first day I adopted
this head-dress (which is convenient and agreeable in every
respect), the people saluted me with “good morning, O Sidi!” (Sir, or
Lord) instead of the usual “good morning, O Howadji!” (i. e.
merchant, as the Franks are rather contemptuously designated by
the Arabs).
For this climate and this way of life, the Egyptian costume is
undoubtedly much better than the European. It is light, cool, and
does not impede the motion of the limbs. The turban thoroughly
protects the head against the sun, and shades the eyes, while it
obstructs the vision much less than a hat-brim. The broad silk shawl
which holds up the baggy trowsers, shields the abdomen against
changes of temperature and tends to prevent diarrhœa, which,
besides ophthalmia, is the only ailment the traveller need fear. The
latter disease may be avoided by bathing the face in cold water after
walking or any exercise which induces perspiration. I have followed
this plan, and though my eyes are exposed daily to the full blaze of
the sun, find them growing stronger and clearer. In fact, since leaving
the invigorating camp-life of California, I have not felt the sensation
of health so purely as now. The other day, to the great delight of our
sailors and the inexhaustible merriment of my friend, I donned one of
Achmet’s dresses. Though the short Theban’s flowing trowsers and
embroidered jacket gave me the appearance of a strapping Turk,
who had grown too fast for his garments, they were so easy and
convenient in every respect, that I have decided to un-Frank myself
for the remainder of the journey.
But our day is not yet at an end. We come on board about eleven
o’clock, and find our breakfast ready for the table. The dishes are
few, but well cooked, and just what a hungry man would desire—
fowls, pigeons, eggs, rice, vegetables, fruit, the coarse but
nourishing bread of the country, and the sweet water of the Nile,
brought to a blush by an infusion of claret. After breakfast we seat
ourselves on the airy divans in front of the cabin, and quietly indulge
in the luxury of a shebook, filled by Achmet’s experienced hand, and
a finjan of Turkish coffee. Then comes an hour’s exercise in Arabic,
after which we read guide-books, consult our maps, write letters, and
occupy ourselves with various mysteries of our household, till the
noonday heat is over. Dinner, which is served between four and five
o’clock, is of the same materials as our breakfast, but differently
arranged, and with the addition of soup. My friend avers that he no
longer wonders why Esau sold his birthright, now that he has tasted
our pottage of Egyptian lentils. Coffee and pipes follow dinner, which
is over with the first flush of sunset and the first premonition of the
coolness and quiet of evening.
We seat ourselves on deck, and drink to its fulness the balm of
this indescribable repose. The sun goes down behind the Libyan
Desert in a broad glory of purple and rosy lights; the Nile is calm and
unruffled, the palms stand as if sculptured in jasper and malachite,
and the torn and ragged sides of the Arabian Mountains, pouring
through a hundred fissures the sand of the plains above, burn with a
deep crimson lustre, as if smouldering from some inward fire. The
splendor soon passes off and they stand for some minutes in dead,
ashy paleness. The sunset has now deepened into orange, in the
midst of which a large planet shines whiter than the moon. A second
glow falls upon the mountains, and this time of a pale, but intense
yellow hue, which gives them the effect of a transparent painting.
The palm-groves are dark below and the sky dark behind them; they
alone, the symbols of perpetual desolation, are transfigured by the
magical illumination. Scarcely a sound disturbs the solemn
magnificence of the hour. Even our full-throated Arabs are silent, and
if a wave gurgles against the prow, it slides softly back into the river,
as if rebuked for the venture. We speak but little, and then mostly in
echoes of each other’s thoughts. “This is more than mere enjoyment
of Nature,” said my friend, on such an evening: “it is worship.”
Speaking of my friend, it is no more than just that I should confess
how much of the luck of this Nile voyage is owing to him, and therein
may be the secret of my complete satisfaction and the secret of the
disappointment of others. It is more easy and yet more difficult for
persons to harmonize while travelling, than when at home. By this I
mean, that men of kindred natures and aims find each other more
readily and confide in each other more freely, while the least jarring
element rapidly drives others further and further apart. No
confessional so completely reveals the whole man as the
companionship of travel. It is not possible to wear the conventional
masks of Society, and one repulsive feature is often enough to
neutralize many really good qualities. On the other hand, a
congeniality of soul and temperament speedily ripens into the firmest
friendship and doubles every pleasure which is mutually enjoyed. My
companion widely differs from me in age, in station, and in his
experiences of life; but to one of those open, honest and loving
natures which are often found in his native Saxony, he unites a most
warm and thorough appreciation of Beauty in Nature or Art. We
harmonize to a miracle, and the parting with him at Assouan will be
the sorest pang of my journey.
My friend, the Howadji, in whose “Nile-Notes” the Egyptian
atmosphere is so perfectly reproduced, says that “Conscience falls
asleep on the Nile.” If by this he means that artificial quality which
bigots and sectarians call Conscience, I quite agree with him, and do
not blame the Nile for its soporific powers. But that simple faculty of
the soul, native to all men, which acts best when it acts
unconsciously, and leads our passions and desires into right paths
without seeming to lead them, is vastly strengthened by this quiet
and healthy life. There is a cathedral-like solemnity in the air of
Egypt; one feels the presence of the altar, and is a better man
without his will. To those rendered misanthropic by disappointed
ambition—mistrustful by betrayed confidence—despairing by
unassuageable sorrow—let me repeat the motto which heads this
chapter.
I have endeavored to picture our mode of life as faithfully and
minutely as possible, because it bears no resemblance to travel in
any other part of the world. Into the heart of a barbarous continent
and a barbarous land, we carry with us every desirable comfort and
luxury. In no part of Europe or America could we be so thoroughly
independent, without undergoing considerable privations, and wholly
losing that sense of rest which is the greatest enjoyment of this
journey. We are cut off from all communication with the great world
of politics, merchandise and usury, and remember it only through the
heart, not through the brain. We go ashore in the delicious mornings,
breathe the elastic air, and wander through the palm-groves, as
happy and care-free as two Adams in a Paradise without Eves. It is
an episode which will flow forward in the under-currents of our
natures through the rest of our lives, soothing and refreshing us
whenever it rises to the surface. I do not reproach myself for this
passive and sensuous existence. I give myself up to it unreservedly,
and if some angular-souled utilitarian should come along and
recommend me to shake off my laziness, and learn the conjugations
of Coptic verbs or the hieroglyphs of Kneph and Thoth, I should not
take the pipe from my mouth to answer him. My friend sometimes
laughingly addresses me with two lines of Hebel’s quaint Allemanic
poetry:

“Ei solch a Leben, junges Bluat,

Desh ish wohl für a Thierle guat.”

(such a life, young blood, best befits an animal), but I tell him that the
wisdom of the Black Forest won’t answer for the Nile. If any one
persists in forcing the application, I prefer being called an animal to
changing my present habits. An entire life so spent would be
wretchedly aimless, but a few months are in truth “sore labor’s bath”
to every wrung heart and overworked brain.
I could say much more, but it requires no little effort to write three
hours in a cabin, when the palms are rustling their tops outside, the
larks singing in the meadows, and the odor of mimosa flowers
breathing through the windows. To travel and write, is like inhaling
and exhaling one’s breath at the same moment. You take in
impressions at every pore of the mind, and the process is so
pleasant, that you sweat them out again most reluctantly. Lest I
should overtake the remedy with the disease, and make to-day
Labor, which should be Rest, I shall throw down the pen, and mount
yonder donkey which stands patiently on the bank, waiting to carry
me to Siout once more, before starting for Thebes.
 CHAPTER VIII.
UPPER EGYPT.

Calm—Mountains and Tombs—A Night Adventure in Ekhmin—Character

of the Boatmen—Fair Wind—Pilgrims—Egyptian Agriculture—Sugar
and Cotton—Grain—Sheep—Arrival at Kenneh—A Landscape—The
Temple of Dendera—First Impressions of Egyptian Art—Portrait of
Cleopatra—A Happy Meeting—We approach Thebes.

Our men were ready at the appointed time, and precisely twenty-
four hours after reaching the port of Siout we spread our sails for
Kenneh, and exchanged a parting salute with the boat of a New York
physician, which arrived some hours after us. The north wind, which
had been blowing freshly during the whole of our stay, failed us
almost within sight of the port, and was followed by three days of
breathless calm, during which time we made about twelve miles a
day, by towing. My friend and I spent half the time on shore,
wandering inland through the fields and making acquaintances in the
villages. We found such tours highly interesting and refreshing, but
nevertheless always returned to our floating Castle of Indolence,
doubly delighted with its home-like cabin and lazy divans. Many of
the villages in this region are built among the mounds of ancient
cities, the names whereof are faithfully enumerated in the guide-
book, but as the cities themselves have wholly disappeared, we
were spared the necessity of seeking for their ruins.
On the third night after leaving Siout, we passed the village of Gow
el-Kebir, the ancient Antæopolis, whose beautiful temple has been
entirely destroyed during the last twenty-five years, partly washed
away by the Nile and partly pulled down to furnish materials for the
Pasha’s palace at Siout. Near this the famous battle between
Hercules and Antæus is reported to have taken place. The fable of
Antæus drawing strength from the earth appears quite natural, after
one has seen the fatness of the soil of Upper Egypt. We ran the
gauntlet of Djebel Shekh Hereedee, a mountain similar to Aboufayda
in form, but much more lofty and imposing. It has also its legend: A
miraculous serpent, say the Arabs, has lived for centuries in its
caverns, and possesses the power of healing diseases. All these
mountains, on the eastern bank of the Nile, are pierced with tombs,
and the openings are sometimes so frequent and so near to each
other as to resemble a colonnade along the rocky crests. They rarely
contain inscriptions, and many of them were inhabited by hermits
and holy men, during the early ages of Christianity. At the most
accessible points the Egyptians have commenced limestone
quarries, and as they are more concerned in preserving piastres
than tombs, their venerable ancestors are dislodged without scruple.
Whoever is interested in Egyptian antiquities, should not postpone
his visit longer. Not only Turks, but Europeans are engaged in the
work of demolition, and the very antiquarians who profess the
greatest enthusiasm for these monuments, are ruthless Vandals
towards them when they have the power.
We dashed past the mountain of Shekh Hereedee in gallant style,
and the same night, after dusk, reached Ekhmin, the ancient
Panopolis. This was one of the oldest cities in Egypt, and dedicated
to the Phallic worship, whose first symbol, the obelisk, has now a
purely monumental significance. A few remnants of this singular
ancient faith appear to be retained among the modern inhabitants of
Ekhmin, but only in the grossest superstitions, and without reference
to the abstract creative principle typified by the Phallic emblems. The
early Egyptians surrounded with mystery and honored with all
religious solemnity what they regarded as the highest human miracle
wrought by the power of their gods, and in a philosophical point of
view, there is no branch of their complex faith more interesting than
this.
As we sat on the bank in the moonlight, quietly smoking our pipes,
the howling of a company of dervishes sounded from the town,
whose walls are a few hundred paces distant from the river. We
inquired of the guard whether a Frank dare visit them. He could not
tell, but offered to accompany me and try to procure an entrance. I
took Achmet and two of our sailors, donned a Bedouin capote, and
set out in search of the dervishes. The principal gate of the town was
closed, and my men battered it vainly with their clubs, to rouse the
guard. We wandered for some time among the mounds of Panopolis,
stumbling over blocks of marble and granite, under palms eighty feet
high, standing clear and silvery in the moonlight. At last, the clamor
of the wolfish dogs we waked up on the road, brought us one of the
watchers outside of the walls, whom we requested to admit us into
the city. He replied that this could not be done. “But,” said Achmet,
“here is on Effendi who has just arrived, and must visit the mollahs
to-night; admit him and fear nothing.” The men thereupon conducted
us to another gate and threw a few pebbles against the window
above it. A woman’s voice replied, and presently the bolts were
undrawn and we entered. By this time the dervishes had ceased
their howlings, and every thing was as still as death. We walked for
half an hour through the deserted streets, visited the mosques and
public buildings, and heard no sound but our own steps. It was a
strangely interesting promenade. The Arabs, armed with clubs,
carried a paper lantern, which flickered redly on the arches and
courts we passed through. My trusty Theban walked by my side, and
took all possible trouble to find the retreat of the dervishes—but in
vain. We passed out through the gate, which was instantly locked
behind us, and had barely reached our vessel, when the unearthly
song of the Moslem priests, louder and wilder than ever, came to our
ears.
The prejudice of the Mohammedans against the Christians is
wearing away with their familiarity with the Frank dress and their
adoption of Frankish vices. The Prophet’s injunction against wine is
heeded by few of his followers, or avoided by drinking arakee, a
liquor distilled from dates and often flavored with hemp. Their
conscience is generally satisfied with a pilgrimage to Mecca and the
daily performance of the prescribed prayers, though the latter is
often neglected. All of my sailors were very punctual in this respect,
spreading their carpets on the forward deck, and occupying an hour
or two every day with genuflexions, prostrations, and salutations
toward Mecca, the direction of which they never lost, notwithstanding
the windings of the Nile. In the cathedrals of Christian Europe I have
often seen pantomimes quite as unnecessary, performed with less
apparent reverence. The people of Egypt are fully as honest and
well-disposed as the greater part of the Italian peasantry. They
sometimes deceive in small things, and are inclined to take trifling
advantages, but that is the natural result of living under a
government whose only rule is force, and which does not even
hesitate to use fraud. Their good humor is inexhaustible. A single
friendly word wins them, and even a little severity awakes no lasting
feeling of revenge. I should much rather trust myself alone among
the Egyptian Fellahs, than among the peasants of the Campagna, or
the boors of Carinthia. Notwithstanding our men had daily
opportunities of plundering us, we never missed a single article. We
frequently went ashore with our dragoman, leaving every thing in the
cabin exposed, and especially such articles as tobacco, shot, dates,
&c., which would most tempt an Arab, yet our confidence was never
betrayed. We often heard complaints from travellers in other boats,
but I am satisfied that any one who will enforce obedience at the
start, and thereafter give none but just and reasonable commands,
need have no difficulty with his crew.
The next morning, the wind being light, we walked forward to El
Menschieh, a town about nine miles distant from Ekhmin. It was
market-day, and the bazaar was crowded with the countrymen, who
had brought their stock of grain, sugar-cane and vegetables. The
men were taller and more muscular than in Lower Egypt, and were
evidently descended from a more intelligent and energetic stock.
They looked at us curiously, but with a sort of friendly interest, and
courteously made way for us as we passed through the narrow
bazaar. In the afternoon the wind increased to a small gale, and bore
us rapidly past Gebel Tookh to the city of Girgeh, so named in Coptic
times from the Christian saint, George. Like Manfalout, it has been
half washed away by the Nile, and two lofty minarets were hanging
on the brink of the slippery bank, awaiting their turn to fall. About
twelve miles from Girgeh, in the Libyan Desert, are the ruins of
Abydus, now covered by the sand, except the top of the portico and
roof of the temple-palace of Sesostris, and part of the temple of
Osiris. We held a council whether we should waste the favorable
wind or miss Abydus, and the testimony of Achmet, who had visited
the ruins, having been taken, we chose the latter alternative. By this
time Girgeh was nearly out of sight, and we comforted ourselves with
the hope of soon seeing Dendera.
The pilgrims to Mecca, by the Kenneh and Kosseir route, were on
their return, and we met a number of boats, crowded with them, on
their way to Cairo from the former place. Most of the boats carried
the red flag, with the star and crescent. On the morning after leaving
Girgeh, we took a long stroll through the fields of Farshoot, which is,
after Siout, the richest agricultural district of Upper Egypt. An
excellent system of irrigation, by means of canals, is kept up, and the
result shows what might be made of Egypt, were its great natural
resources rightly employed. The Nile offers a perpetual fountain of
plenty and prosperity, and its long valley, from Nubia to the sea,
would become, in other hands, the garden of the world. So rich and
pregnant a soil I have never seen. Here, side by side, flourish wheat,
maize, cotton, sugar-cane, indigo, hemp, rice, dourra, tobacco,
olives, dates, oranges, and the vegetables and fruits of nearly every
climate. The wheat, which, in November, we found young and green,
would in March be ripe for the sickle, and the people were cutting
and threshing fields of dourra, which they had planted towards the
end of summer. Except where the broad meadows are first reclaimed
from the rank, tufted grass which has taken possession of them, the
wheat is sowed upon the ground, and then ploughed in by a sort of
crooked wooden beam, shod with iron, and drawn by two camels or
buffaloes. I saw no instance in which the soil was manured. The
yearly deposit made by the bountiful river seems to be sufficient. The
natives, it is true, possess immense numbers of pigeons, and every
village is adorned with towers, rising above the mud huts like the
pylons of temples, and inhabited by these birds. The manure
collected from them is said to be used, but probably only in the
culture of melons, cucumbers, and other like vegetables with which
the gardens are stocked.
The fields of sugar-cane about Farshoot were the richest I saw in
Egypt. Near the village, which is three miles from the Nile, there is a
steam sugar-refinery, established by Ibrahim Pasha, who seems to
have devoted much attention to the culture of cane, with a view to
his own profit. There are several of these manufactories along the
Nile, and the most of them were in full operation, as we passed. At
Radamoon, between Minyeh and Siout, there is a large manufactory,
where the common coarse sugar made in the Fellah villages is
refined and sent to Cairo. We made use of this sugar in our
household, and found it to be of excellent quality, though coarser
than that of the American manufactories. The culture of cotton has
not been so successful. The large and handsome manufactory built
at Kenneh, is no longer in operation, and the fields which we saw
there, had a forlorn, neglected appearance. The plants grow
luxuriantly, and the cotton is of fine quality, but the pods are small
and not very abundant. About Siout, and in Middle and Lower Egypt,
we saw many fields of indigo, which is said to thrive well. Peas,
beans and lentils are cultivated to a great extent, and form an
important item of the food of the inhabitants. The only vegetables we
could procure for our kitchen, were onions, radishes, lettuce and
spinage. The Arabs are very fond of the tops of radishes, and eat
them with as much relish as their donkeys.
One of the principal staples of Egypt is the dourra (holcus
sorghum), which resembles the zea (maize) in many respects. In
appearance, it is very like broom-corn, but instead of the long, loose
panicle of red seeds, is topped by a compact cone of grains, smaller
than those of maize, but resembling them in form and taste. The
stalks are from ten to fifteen feet high, and the heads frequently
contain as much substance as two ears of maize. It is planted in
close rows, and when ripe is cut by the hand with a short sickle, after
which the heads are taken off and threshed separately. The grain is
fed to horses, donkeys and fowls, and in Upper Egypt is used almost
universally for bread. It is of course very imperfectly ground, and
unbolted, and the bread is coarse and dark, though nourishing. In
the Middle and Southern States of America this grain would thrive
well and might be introduced with advantage.
The plains of coarse, wiry grass (halfeh), which in many points on
the Nile show plainly the neglect of the inhabitants, who by a year’s
labor might convert them into blooming fields, are devoted to the
pasturage of large herds of sheep, and goats, and sometimes droves
of buffaloes. The sheep are all black or dark-brown, and their bushy
heads remind one of terriers. The wool is rather coarse, and when
roughly spun and woven by the Arabs, in its natural color, forms the
mantle, something like a Spanish poncho, which is usually the
Fellah’s only garment. The mutton, almost the only meat to be found,
is generally lean, and brings a high price, considering the abundance
of sheep. The flesh of buffaloes is eaten by the Arabs, but is too
tough, and has too rank a flavor, for Christian stomachs. The goats
are beautiful animals, with heads as slender and delicate as those of
gazelles. They have short, black horns, curving downward—long,
silky ears, and a peculiarly mild and friendly expression of
countenance. We had no difficulty in procuring milk in the villages,
and sometimes fresh butter, which was more agreeable to the taste
than the sight. The mode of churning is not calculated to excite one’s
appetite. The milk is tied up in a goat’s skin, and suspended by a
rope to the branch of a tree. One of the Arab housewives (who are
all astonishingly ugly and filthy) then stations herself on one side,
and propels it backward and forward till the process is completed.
The cheese of the country resembles a mixture of sand and slacked
lime, and has an abominable flavor.
Leaving Farshoot, we swept rapidly past Haou, the ancient
Diospolis parva, or Little Thebes, of which nothing is left but some
heaps of dirt, sculptured fragments, and the tomb of a certain
Dionysius, son of a certain Ptolemy. The course of the mountains,
which follow the Nile, is here nearly east and west, as the river
makes a long curve to the eastward on approaching Kenneh. The
valley is inclosed within narrower bounds, and the Arabian
Mountains on the north, shooting out into bold promontories from the
main chain, sometimes rise from the water’s edge in bluffs many
hundred feet in height. The good wind, which had so befriended us
for three days, followed us all night, and when we awoke on the
morning of December 4th, our vessel lay at anchor in the port of
Kenneh, having beaten by four hours the boat of our American
friend, which was reputed to be one of the swiftest on the river.
Kenneh, which lies about a mile east of the river, is celebrated for
the manufacture of porous water-jars, and is an inferior mart of trade
with Persia and India, by means of Kosseir, on the Red Sea, one
hundred and twenty miles distant. The town is large, but mean in
aspect, and does not offer a single object of interest. It lies in the
centre of a broad plain. We rode through the bazaars, which were
tolerably well stocked and crowded with hadji, or pilgrims of Mecca.
My friend, who wished to make a flag of the Saxe-Coburg colors, for
his return voyage, tried in vain to procure a piece of green cotton
cloth. Every other color was to be had but green, which, as the
sacred hue, worn only by the descendants of Mohammed, was
nowhere to be found. He was finally obliged to buy a piece of white
stuff and have it specially dyed. It came back the same evening,
precisely the color of the Shereef of Mecca’s turban.
On the western bank of the Nile, opposite Kenneh, is the site of
the city of Tentyra, famed for its temple of Athor. It is now called
Dendera, from the modern Arab village. After breakfast, we shipped
ourselves and our donkeys across the Nile, and rode off in high
excitement, to make our first acquaintance with Egyptian temples.
The path led through a palm grove, which in richness and beauty
rivalled those of the Mexican tierra caliente. The lofty shafts of the
date and the vaulted foliage of the doum-palm, blended in the most
picturesque groupage, contrasted with the lace-like texture of the
flowering mimosa, and the cloudy boughs of a kind of gray cypress.
The turf under the trees was soft and green, and between the slim
trunks we looked over the plain, to the Libyan Mountains—a long
train of rosy lights and violet shadows. Out of this lovely wood we
passed between magnificent fields of dourra and the castor-oil bean,
fifteen feet in height, to a dyke which crossed the meadows to
Dendera. The leagues of rank grass on our right rolled away to the
Desert in shining billows, and the fresh west-wind wrapped us in a
bath of intoxicating odors. In the midst of this green and peaceful
plain rose the earthy mounds of Tentyra, and the portico of the
temple, almost buried beneath them, stood like a beacon, marking
the boundary of the Desert.
We galloped our little animals along the dyke, over heaps of dirt
and broken bricks, among which a number of Arabs were burrowing
for nitrous earth, and dismounted at a small pylon, which stands two
or three hundred paces in front of the temple. The huge jambs of
sandstone, covered with sharply cut hieroglyphics and figures of the
Egyptian gods, and surmounted by a single block, bearing the
mysterious winged globe and serpent, detained us but a moment,
and we hurried down what was once the dromos of the temple, now
represented by a double wall of unburnt bricks. The portico, more
than a hundred feet in length, and supported by six columns, united
by screens of masonry, no stone of which, or of the columns
themselves, is unsculptured, is massive and imposing, but struck me
as being too depressed to produce a very grand effect. What was my
astonishment, on arriving at the entrance, to find that I had
approached the temple on a level with half its height, and that the
pavement of the portico was as far below as the scrolls of its cornice
were above me. The six columns I had seen covered three other
rows, of six each, all adorned with the most elaborate sculpture and
exhibiting traces of the brilliant coloring which they once possessed.
The entire temple, which is in an excellent state of preservation,
except where the hand of the Coptic Christian has defaced its
sculptures, was cleaned out by order of Mohammed Ali, and as all its
chambers, as well as the roof of enormous sandstone blocks, are
entire, it is considered one of the most complete relics of Egyptian
art.
I find my pen at fault, when I attempt to describe the impression
produced by the splendid portico. The twenty-four columns, each of
which is sixty feet in height, and eight feet in diameter, crowded upon
a surface of one hundred feet by seventy, are oppressive in their
grandeur. The dim light, admitted through the half closed front, which
faces the north, spreads a mysterious gloom around these mighty
shafts, crowned with the fourfold visage of Athor, still rebuking the
impious hands that have marred her solemn beauty. On the walls,
between columns of hieroglyphics, and the cartouches of the
Cæsars and the Ptolemies, appear the principal Egyptian deities—
the rigid Osiris, the stately Isis and the hawk-headed Orus. Around
the bases of the columns spring the leaves of the sacred lotus, and
the dark-blue ceiling is spangled with stars, between the wings of the
divine emblem. The sculptures are all in raised relief, and there is no
stone in the temple without them. I cannot explain to myself the
unusual emotion I felt while contemplating this wonderful
combination of a simple and sublime architectural style with the
utmost elaboration of ornament. My blood pulsed fast and warm on
my first view of the Roman Forum, but in Dendera I was so
saddened and oppressed, that I scarcely dared speak for fear of
betraying an unmanly weakness. My friend walked silently between
the columns, with a face as rigidly sad as if he had just looked on the
coffin of his nearest relative. Though such a mood was more painful
than agreeable, it required some effort to leave the place, and after a
stay of two hours, we still lingered in the portico and walked through
the inner halls, under the spell of a fascination which we had hardly
power to break.
The portico opens into a hall, supported by six beautiful columns,
of smaller proportions, and lighted by a square aperture in the solid
roof. On either side are chambers connected with dim and lofty
passages, and beyond is the sanctuary and various other
apartments, which receive no light from without. We examined their
sculptures by the aid of torches, and our Arab attendants kindled
large fires of dry corn stalks, which cast a strong red light on the
walls. The temple is devoted to Athor, the Egyptian Venus, and her
image is everywhere seen, receiving the homage of her worshippers.
Even the dark stair case, leading to the roof—up which we climbed
over heaps of sand and rubbish—is decorated throughout with
processions of symbolical figures. The drawing has little of that
grotesque stiffness which I expected to find in Egyptian sculptures,
and the execution is so admirable in its gradations of light and
shade, as to resemble, at a little distance, a monochromatic painting.
The antiquarians view these remains with little interest, as they date
from the comparatively recent era of the Ptolemies, at which time
sculpture and architecture were on the decline. We, who had seen
nothing else of the kind, were charmed with the grace and elegance
of this sumptuous mode of decoration. Part of the temple was built
by Cleopatra, whose portrait, with that of her son Cæsarion, may still
be seen on the exterior wall. The face of the colossal figure has been
nearly destroyed, but there is a smaller one, whose soft, voluptuous
outline is still sufficient evidence of the justness of her renown. The
profile is exquisitely beautiful. The forehead and nose approach the
Greek standard, but the mouth is more roundly and delicately
curved, and the chin and cheek are fuller. Were such an outline
made plastic, were the blank face colored with a pale olive hue,
through which should blush a faint rosy tinge, lighted with bold black
eyes and irradiated with the lightning of a passionate nature, it would
even now “move the mighty hearts of captains and of kings.”
Around the temple and over the mounds of the ancient city are
scattered the ruins of an Arab village which the inhabitants suddenly
deserted, without any apparent reason, two or three years previous
to our visit. Behind it, stretches the yellow sand of the Desert. The
silence and aspect of desertion harmonize well with the spirit of the
place, which would be much disturbed were one beset, as is usual in
the Arab towns, by a gang of naked beggars and barking wolf-dogs.
Besides the temple, there are also the remains of a chapel of Isis,
with a pylon, erected by Augustus Cæsar, and a small temple, nearly
whelmed in the sand, supposed to be one of the mammeisi, or lying-
in houses of the goddess Athor, who was honored in this form, on
account of having given birth to the third member of the divine Triad.
At sunset, we rode back from Dendera and set sail for Thebes. In
the evening, as we were sweeping along by moonlight, with a full
wind, a large dahabiyeh came floating down the stream. Achmet,
who was on the look-out, saw the American flag, and we hailed her.
My delight was unbounded, to hear in reply the voice of my friend,
Mr. Degen, of New York, who, with his lady and two American and
English gentlemen, were returning from a voyage to Assouan. Both
boats instantly made for the shore, and for the first time since leaving
Germany I had the pleasure of seeing familiar faces. For the space
of three hours I forgot Thebes and the north wind, but towards
midnight we exchanged a parting salute of four guns and shook out
the broad sails of the Cleopatra, who leaned her cheek to the waves
and shot off like a sea-gull. I am sure she must have looked beautiful
to my friends, as they stood on deck in the moonlight.
 CHAPTER IX.
THEBES—THE WESTERN BANK.

Arrival at Thebes—Ground-Plan of the Remains—We Cross to the

Western Bank—Guides—The Temple of Goorneh—Valley of the
Kings’ Tombs—Belzoni’s Tomb—The Races of Men—Vandalism of
Antiquarians—Bruce’s Tomb—Memnon—The Grandfather of
Sesostris—The Head of Amunoph—The Colossi of the Plain—
Memnonian Music—The Statue of Remeses—The Memnonium—
Beauty of Egyptian Art—More Scrambles among the Tombs—The
Bats of the Assasseef—Medeenet Abou—Sculptured Histories—The
Great Court of the Temple—We return to Luxor.

On the following evening, about nine o’clock, as my friend and I

were taking our customary evening pipe in the cabin, our vessel
suddenly stopped. The wind was still blowing, and I called to Achmet
to know what was the matter. “We have reached Luxor,” answered
the Theban. We dropped the shebooks, dashed out, up the bank,
and saw, facing us in the brilliant moonlight, the grand colonnade of
the temple, the solid wedges of the pylon, and the brother-obelisk of
that which stands in the Place de la Concorde, in Paris. The wide
plain of Thebes stretched away on either hand, and the beautiful
outlines of the three mountain ranges which inclose it, rose in the
distance against the stars. We looked on the landscape a few
moments, in silence. “Come,” said my friend, at length, “this is
enough for to-night. Let us not be too hasty to exhaust what is in
store for us.” So we returned to our cabin, closed the blinds, and
arranged our plans for best seeing, and best enjoying the wonders of
the great Diospolis.
Before commencing my recital, let me attempt to give an outline of
the typography of Thebes. The course of the Nile is here nearly
north, dividing the site of the ancient city into two almost equal parts.
On approaching it from Kenneh, the mountain of Goorneh, which
abuts on the river, marks the commencement of the western division.
This mountain, a range of naked limestone crags, terminating in a
pyramidal peak, gradually recedes to the distance of three miles
from the Nile, which it again approaches further south. Nearly the
whole of the curve, which might be called the western wall of the city,
is pierced with tombs, among which are those of the queens, and the
grand priestly vaults of the Assasseef. The Valley of the Kings’
Tombs lies deep in the heart of the range seven or eight miles from
the river. After passing the corner of the mountain, the first ruin on
the western bank is that of the temple-palace of Goorneh. More than
a mile further, at the base of the mountain, is the Memnonium, or
temple of Remeses the Great, between which and the Nile the two
Memnonian colossi are seated on the plain. Nearly two miles to the
south of this is the great temple of Medeenet Abou, and the
fragments of other edifices are met with, still further beyond. On the
eastern bank, nearly opposite Goorneh, stands the temple of
Karnak, about half a mile from the river. Eight miles eastward, at the
foot of the Arabian Mountains, is the small temple of Medamot,
which, however, does not appear to have been included in the limits
of Thebes. Luxor is directly on the bank of the Nile, a mile and a half
south of Karnak, and the plain extends several miles beyond it,
before Beaching the isolated range, whose three conical peaks are
the landmarks of Thebes to voyagers on the river.
These distances convey an idea of the extent of the ancient city,
but fail to represent the grand proportions of the landscape, so well
fitted, in its simple and majestic outlines, to inclose the most
wonderful structures the world has ever seen. The green expanse of
the plain; the airy coloring of the mountains; the mild, solemn blue of
the cloudless Egyptian sky;—these are a part of Thebes, and
inseparable from the remembrance of its ruins.
At sunrise we crossed to the western bank and moored our boat
opposite Goorneh. It is advisable to commence with the Tombs, and
close the inspection of that side with Medeenet Abou, reserving
Karnak, the grandest of all, for the last. The most unimportant
objects in Thebes are full of interest when seen first, whereas
Karnak, once seen, fills one’s thoughts to the exclusion of every
thing else. There are Arab guides for each bank, who are quite