Multilinear Algebra PDF

ELEMENTS OF LINEAR AND MULTILINEAR
ALGEBRA
John M. Erdman
Portland State University
Version June 3, 2014
c
2010
John M. Erdman
E-mail address: erdman@pdx.edu
iii
It is not essential for the value of an education that every idea be understood at the
time of its accession. Any person with a genuine intellectual interest and a wealth
of intellectual content acquires much that he only gradually comes to understand
fully in the light of its correlation with other related ideas. . . . Scholarship is a
progressive process, and it is the art of so connecting and recombining individual
items of learning by the force of ones whole character and experience that nothing
is left in isolation, and each idea becomes a commentary on many others.
- NORBERT WIENER
Contents
PREFACE
ix
NOTATION AND TERMINOLOGY

Some Algebraic Objects
Notation for Sets of Numbers
Greek Letters
Fraktur Fonts
1
1
2
3
4
Chapter 1. VECTOR SPACES

1.1. Abelian Groups
1.2. Functions and Diagrams
1.3. Rings
1.4. Vector Spaces
1.5. Subspaces
1.6. Linear Combinations and Linear Independence
1.7. Bases for Vector Spaces
5
5
6
9
10
12
14
16
Chapter 2. LINEAR TRANSFORMATIONS

2.1. Linearity
2.2. Invertible Linear Maps
2.3. Matrix Representations
2.4. Spans, Independence, and Linearity
2.5. Dual Spaces
2.6. Annihilators
21
21
23
24
24
25
27
Chapter 3. THE LANGUAGE OF CATEGORIES

3.1. Objects and Morphisms
3.2. Functors
3.3. Universal Mapping Properties
3.4. Products and Coproducts
3.5. Quotients
3.6. Exact Sequences
3.7. Some Miscellaneous Results
29
29
30
32
33
34
35
37
Chapter 4. The Spectral Theorem for Vector Spaces

4.1. Projections
4.2. Algebras
4.3. Quotients and Unitizations
4.4. The Spectrum
4.5. Polynomials
4.6. Minimal Polynomials
4.7. Invariant Subspaces
4.8. Burnsides Theorem
4.9. Eigenvalues and Eigenvectors
39
39
40
42
42
43
45
47
48
49
vi
CONTENTS
4.10.
4.11.
The Spectral Theorem - Vector Space Version

Two Decomposition Theorems
51
52
Chapter 5. THE SPECTRAL THEOREM FOR INNER PRODUCT SPACES

5.1. Inner Products
5.2. Orthogonality
5.3. Involutions and Adjoints
5.4. Orthogonal Projections
5.5. The Spectral Theorem for Inner Product Spaces
55
55
57
60
63
64
Chapter 6. A BRIEF REVIEW OF DIFFERENTIAL CALCULUS

6.1. Tangency
6.2. The Differential
6.3. The Gradient of a Scalar Field in Rn
67
67
68
70
Chapter 7. MULTILINEAR MAPS AND DETERMINANTS

7.1. Permutations
7.2. Multilinear Maps
7.3. Determinants
7.4. Tensor Products of Vector Spaces
7.5. Tensor Products of Linear Maps
73
73
73
74
76
78
Chapter 8. TENSOR ALGEBRAS

8.1. Grassmann Algebras
8.2. Existence of Grassmann Algebras
8.3. The Hodge -operator
79
79
80
82
Chapter 9. DIFFERENTIAL MANIFOLDS

9.1. Manifolds in R3
9.2. Charts, Atlases, and Manifolds
9.3. Differentiable Functions Between Manifolds
9.4. The Geometric Tangent Space
9.5. The Algebraic Tangent Space
83
83
83
85
86
89
Chapter
10.1.
10.2.
10.3.
10.4.
10.5.
10.6.
10. DIFFERENTIAL FORMS ON MANIFOLDS

Vector Fields
Differential 1-forms
Differential k-forms
Some Classical Vector Analysis
Closed and Exact Forms
Poincares Lemma
97
97
98
98
100
101
101
Chapter
11.1.
11.2.
11.3.
11.4.
11. HOMOLOGY AND COHOMOLOGY

The de Rham Cohomology Group
Cochain Complexes
Simplicial Homology
Simplicial Cohomology
105
105
106
107
110
Chapter 12. STOKES THEOREM

12.1. Integration of Differential Forms
12.2. Generalized Stokes Theorem
113
113
115
Chapter 13. GEOMETRIC ALGEBRA

13.1. Geometric Plane Algebra
13.2. Geometric Algebra in 3-Space
119
119
122
CONTENTS
Chapter
14.1.
14.2.
14.3.
14.4.
14. CLIFFORD ALGEBRAS

Quadratic Forms
Definition of Clifford Algebra
Orthogonality with Respect to Bilinear Forms
Examples of Clifford Algebras
vii
125
125
125
126
126
Bibliography
129
Index
131
PREFACE
This set of notes is an activity-oriented introduction to the study of linear and multilinear
algebra. The great majority of the results in beginning linear and multilinear are straightforward
and can be verified by the thoughtful student. Indeed, that is the main point of these notes
to convince the beginner that the subject is accessible. In the material that follows there are
numerous indicators that suggest activity on the part of the reader: words such as proposition,
example, exercise, and corollary, if not followed by a proof or a reference to a proof, are
invitations to verify the assertions made. When the proof of a theorem appears to me to be too
difficult for the average student to (re)invent and I have no improvements to offer to the standard
proofs, I provide references to standard treatments. These notes were written for a 2-term course in
linear/multilinear algebra for seniors and first year graduate students at Portland State University.
The prerequisites for working through this material are quite modest. Elementary properties
of the real number system, the arithmetic of matrices, ability to solve systems of linear equations,
and the ability to evaluate the determinant of a square matrix are assumed. A few examples and
exercises depend on differentiation and/or integration of real valued functions, but no particular
skill with either is required.
There are of course a number of advantages and disadvantages in consigning a document to
electronic life. One advantage is the rapidity with which links implement cross-references. Hunting
about in a book for lemma 3.14.23 can be time-consuming (especially when an author engages in
the entirely logical but utterly infuriating practice of numbering lemmas, propositions, theorems,
corollaries, and so on, separately). A perhaps more substantial advantage is the ability to correct
errors, add missing bits, clarify opaque arguments, and remedy infelicities of style in a timely
fashion. The correlative disadvantage is that a reader returning to the web page after a short
time may find everything (pages, definitions, theorems, sections) numbered differently. (LATEXis an
amazing tool.) I will change the date on the title page to inform the reader of the date of the last
nontrivial update (that is, one that affects numbers or cross-references).
The most serious disadvantage of electronic life is impermanence. In most cases when a web
page vanishes so, for all practical purposes, does the information it contains. For this reason (and
the fact that I want this material to be freely available to anyone who wants it) I am making use of
a Share Alike license from Creative Commons. It is my hope that anyone who finds this material
useful will correct what is wrong, add what is missing, and improve what is clumsy. For more
information on creative commons licenses see http://creativecommons.org/. Concerning the
text itself, please send corrections, suggestions, complaints, and all other comments to the author
at
erdman@pdx.edu
ix

Some Algebraic Objects
Let S be a nonempty set. Consider the following axioms:
(1) + : S S S.
( + is a binary operation, called addition, on S)
(2) (x + y) + z = x + (y + z) for all x, y, z S.
(associativity of addition)
(3) There exists 0S S such that x + 0S = 0S + x = x for all x S.
(existence of an
additive identity)
(4) For every x S there exists x S such that x + (x) = (x) + x = 0S .
(existence
of additive inverses)
(5) x + y = y + x for all x, y S.
(commutativity of addition)
(6) : S S S : (x, y) 7 x y.
(the map (x, y) 7 x y is a binary operation, called
multiplication, on S. Convention: We will usually write xy instead of x y.)
(7) (xy)z = x(yz) for all x, y, z S.
(associativity of multiplication)
(8) (x+y)z = xz +yz and x(y +z) = xy +xz for all x, y, z S.
(multiplication distributes
over addition)
(9) There exists 1S in S such that x 1S = 1S x = x for all x S.
(existence of a
multiplicative identity or unit)
(10) 1S 6= 0S .
(11) For every x S such that x 6= 0S there exists x1 S such that xx1 = x1 x = 1S .
(existence of multiplicative inverses)
(12) xy = yx for all x, y S.
(commutativity of multiplication)
Definitions.
(S, +) is a semigroup if it satisfies axioms (1)(2).

(S, +) is a monoid if it satisfies axioms (1)(3).
(S, +) is a group if it satisfies axioms (1)(4).
(S, +) is an Abelian group if it satisfies axioms (1)(5).
(S, +, m) is a ring if it satisfies axioms (1)(8).
(S, +, m) is a commutative ring if it satisfies axioms (1)(8) and (12).
(S, +, m) is a unital ring (or ring with identity, or unitary ring) if it satisfies
axioms (1)(9).
(S, +, m) is a division ring (or skew field) if it satisfies axioms (1)(11).
(S, +, m) is a field if it satisfies axioms (1)(12).
Remarks.
A binary operation is often written additively, (x, y) 7 x + y, if it is commutative and
multiplicatively, (x, y) 7 xy, if it is not. This is by no means always the case: in a
commutative ring (the real numbers or the complex numbers, for example), both addition
and multiplication are commutative.
When no confusion is likely to result we often write 0 for 0S and 1 for 1S .
Many authors require a ring to satisfy axioms (1)(9).
1
It is easy to see that axiom (10) holds in any unital ring except the trivial ring S = {0}.
Convention: Unless the contrary is stated we will assume that every unital ring is nontrivial.
Notation for Sets of Numbers
Here is a list of fairly standard notations for some sets of numbers which occur frequently in
these notes:
C is the set of complex numbers
R is the set of real numbers
Rn is the set of all n-tuples (r1 , r2 , . . . , rn ) of real numbers
R+ = {x R : x 0}, the positive real numbers
Q is the set of rational numbers
Q+ = {x Q : x 0}, the positive rational numbers
Z is the set of integers
Z+ = {0, 1, 2, . . . }, the positive integers
N = {1, 2, 3, . . . }, the set of natural numbers
Nn = {1, 2, 3, . . . , n} the first n natural numbers
[a, b] = {x R : a x b}
[a, b) = {x R : a x < b}
(a, b] = {x R : a < x b}
(a, b) = {x R : a < x < b}
[a, ) = {x R : a x}
(a, ) = {x R : a < x}
(, b] = {x R : x b}
(, b) = {x R : x < b}
S1 = {(x, y) R2 : x2 + y 2 = 1}, the unit circle
GREEK LETTERS
Greek Letters
Upper case
Lower case
English name (approximate pronunciation)
A
B
E
Z
H
I
K
M
N
T
Y
or
Alpha (AL-fuh)
Beta (BAY-tuh)
Gamma (GAM-uh)
Delta (DEL-tuh)
Epsilon (EPP-suh-lon)
Zeta (ZAY-tuh)
Eta (AY-tuh)
Theta (THAY-tuh)
Iota (eye-OH-tuh)
Kappa (KAP-uh)
Lambda (LAM-duh)
Mu (MYOO)
Nu (NOO)
Xi (KSEE)
Omicron (OHM-ih-kron)
Pi (PIE)
Rho (ROH)
Sigma (SIG-muh)
Tau (TAU)
Upsilon (OOP-suh-lon)
Phi (FEE or FAHY)
Chi (KHAY)
Psi (PSEE or PSAHY)
Omega (oh-MAY-guh)
Fraktur Fonts
In these notes Fraktur fonts are used (most often for families of sets and families of linear maps).
Below are the Roman equivalents for each letter. When writing longhand or presenting material
on a blackboard it is usually best to substitute script English letters.
Fraktur
Upper case
Fraktur
Lower case
Roman
Lower Case
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Q
R
S
T
U
V
W
X
Y
Z
a
b
c
d
e
f
g
h
i
j
k
l
m
n
o
p
q
r
s
t
u
v
w
x
y
z
a
b
c
d
e
f
g
h
i
j
k
l
m
n
o
p
q
r
s
t
u
v
w
x
y
z
CHAPTER 1
VECTOR SPACES
1.1. Abelian Groups
1.1.1. Convention. In general, Abelian groups are written additively. That is, the usual notation
for the binary operation on an Abelian group is +. Of course, there are special examples where
this notation is inappropriate: most notably for the nonzero real numbers, the strictly positive real
numbers, and the nonzero complex numbers under multiplication. It is conventional, if not entirely
logical to write, let G be an Abelian group, when what is meant is, let (G, +) be an Abelian
group.
1.1.2. Proposition. The identity element in an Abelian group is unique.
e in G such
Hint for proof . Let G be an Abelian group. Suppose that there are elements 0 and 0
e
e
that x + 0 = x and x + 0 = x hold for every x G. Prove that 0 = 0.
1.1.3. Proposition. Each element in an Abelian group has a unique.inverse.
Hint for proof . Let G be an Abelian group and x G. To prove uniqueness of the inverse for
x suppose that there are elements y and z in G such that x + y = 0 and x + z = 0. Prove that
y = z.
1.1.4. Proposition. If x is an element of an Abelian group such that x + x = x, then x = 0.
1.1.5. Proposition. For every element x in an Abelian group (x) = x.
1.1.6. Example. Let S be a nonempty set and F be a field. Denote by F(S, F) the family of all
F-valued functions on S. For f , g F(S, F) define f + g by
(f + g)(x) = f (x) + g(x)
for all x S. Under this operation (called pointwise addition) F(S, F) is an Abelian group.
1.1.7. Example. As a special case of example 1.1.6, we may regard Euclidean n-space Rn as an
Abelian group.
1.1.8. Example. As a special case of example 1.1.6, we may regard the set R of all sequences
of real numbers as an Abelian group.
1.1.9. Example. Let E2 be the Euclidean plane. It contains points (which do not have coordinates)
and lines (which do not have equations). A directed segment is an ordered pair of points. Define
two directed segments to be equivalent if they are congruent (have the same length), lie on parallel
lines, and have the same direction. This is clearly an equivalence relation on the set DS of directed
segments in the plane. We denote by P Q the equivalence class containing the directed segment
(P, Q), going from the point P to the point Q. Define an operation + on these equivalence classes
by

P Q + QR = P R.
This operation is well defined and under it DS is an Abelian group.
1.1.10. Exercise. Suppose that A, B, C, and D are points in the plane such that AB = CD.

Show that AC = BD.
5
1. VECTOR SPACES
1.1.11. Definition. Let G and H be Abelian groups. A map f : G H is a homomorphism if

f (x + y) = f (x) + f (y)
for all x, y G. We will denote by Hom(G, H) the set of all homomorphisms from G into H and
will abbreviate Hom(G, G) to Hom(G).
1.1.12. Proposition. If f : G H is a homomorphism of Abelian groups, then f (0) = 0.
1.1.13. Proposition. If f : G H is a homomorphism of Abelian groups, then f (x) = f (x)
for each x G.
1.1.14. Definition. Let G and H be Abelian groups. For f and g in Hom(G, H) we define
f + g : G H : x 7 f (x) + g(x).
1.1.15. Example. Let G and H be Abelian groups. With addition as defined in 1.1.14 Hom(G, H)
is an Abelian group.
1.1.16. Convention. Let G, H, and J be Abelian groups and f : G H and g : H J be
homomorphisms. Then the composite of g with f is denoted by gf (rather than by g f ). That is,
gf : G J : x 7 g(f (x)).
1.1.17. Proposition. Let G, H, and J be Abelian groups, f Hom(G, H), and g Hom(H, J),
then the composite gf belongs to Hom(G, J).
1.2. Functions and Diagrams
1.2.1. Definition. Let S and T be sets and f : S T . The set S is the domain of f . The set
T is the codomain of f . And {(x, f (x)) : x S} is the graph of f . The domain of f is denoted
by dom f .
1.2.2. Definition. A function f is injective (or one-to-one) if x = y whenever f (x) = f (y).
That is, f is injective if no two distinct elements in its domain have the same image. An injective
map is called an injection.
A function is surjective (or onto) if its range is equal to its codomain. A surjective map is
called a surjection.
A function is bijective (or a one-to-one correspondence) if it is both injective and surjective. A bijective map is called a bijection.
1.2.3. Definition. It is frequently useful to think of functions as arrows in diagrams. For example,
the situation h : R S, j : R T , k : S U , f : T U may be represented by the following
diagram.
R
/T
f

/U
The diagram is said to commute if k h = f j. Diagrams need not be rectangular. For instance,
R
d
S
is a commutative diagram if d = k h.

/U
1.2. FUNCTIONS AND DIAGRAMS
1.2.4. Example. Here is one diagrammatic way of stating the associative law for composition of
functions: If h : R S, g : S T , and f : T U and we define j and k so that the triangles in
the diagram
j
/T
?

/U
commute, then the square also commutes.

1.2.5. Convention. If S, T , and U are sets we will often not distinguish
between
(S T ) U ,

S (T U ), and S T U . That is, the ordered pairs (s, t), u and s, (t, u) and the ordered
triple (s, t, u) will usually be treated as identical.
1.2.6. Notation. Let S be a set. The map
idS : S S : x 7 x
is the identity function on S. When no confusion will result we write id for idS .
1.2.7. Definition.
Let S and T be sets, f : S T , and A S. Then the restriction of f to
A, denoted by f A , is the function f A,S , where A,S : A S : x 7 x is the inclusion map of A
into S. That is, f A is the mapping from A into T whose value at each x in A is f (x).
SO
f
A,S
/T

Suppose that g : A T and A S. A function f : S T is an extension of g to S if f A = g,
that is, if the diagram
SO
f
A,S

g
/T
commutes.
1.2.8. Notation. If S, T , and U are nonempty sets and if f : S T and g : S U , then we
define the function (f, g) : S T U by
(f, g)(s) = (f (s), g(s)).
Suppose, on the other hand, that we are given a function h mapping S into the Cartesian product

T U . Then for each s S the image h(s) is an ordered pair, which we will write as h1 (s), h2 (s) .
(The superscripts have nothing to do with powers.) Notice that we now have functions h1 : S T
and h2 : S U . These are the components of h. In abbreviated notation h = (h1 , h2 ).
1.2.9. Notation. Let f : S U and g : T V be functions between sets. Then f g denotes
the map

f g : S T U V : (s, t) 7 f (s), g(t) .
1. VECTOR SPACES
1.2.10. Exercise. Let S be a set and a : S S S be a function such that the diagram
SSS
aid
//
id a
SS
/S
(D1)
commutes. What is (S, a)? Hint. Interpret a as, for example, addition (or multiplication).
1.2.11. Convention. We will have use for a standard one-element set, which, if we wish, we can
regard as the Cartesian product of an empty family of sets. We will denote it by 1. For each set
S there is exactly one function from S into 1. We will denote it by S . If no confusion is likely to
arise we write for S .
1.2.12. Exercise. Let S be a set and suppose that a : S S S and : 1 S are functions such
that both diagram (D1) above and the diagram (D2) which follows commute.
9 SO e
g
f
a
1S
/ SS o
id
S1
id
(D2)
(Here f and g are the obvious bijections.) What is (S, a, )?

1.2.13. Notation. We denote by the diagonal mapping of a set S into S S. That is,
: S S S : s 7 (s, s).
1.2.14. Exercise. Let S be a set and suppose that a : S S S and : 1 S are functions
such that the diagrams (D1) and (D2) above commute. Suppose further that there is a function
: S S for which the following diagram commutes.
S
id
/ SS
/ SS
id
/: S
(D3)
What is (S, a, , )?
1.2.15. Notation. Let S be a set. We denote by the interchange (or switching) operation
on the S S. That is,
: S S S S : (s, t) 7 (t, s).
1.2.16. Exercise. Let S be a set and suppose that a : S S S, : 1 S, and : S S are
functions such that the diagrams (D1), (D2), and (D3) above commute. Suppose further that the
following diagram commutes.
/ SS
SS
What is (S, a, , , )?
(D4)
1.3. RINGS
1.2.17. Exercise. Let f : G H be a function between Abelian groups. Suppose that the diagram
GG
f f
/ H H

/H
commutes. What can be said about the function f ?

1.2.18. Notation. If S and T are sets we denote by F(S, T ) the family of all functions from S
into T . When F is a field there are several common notations for the family of F-valued functions
on S. We denote by l(S) (or by l(S, F), or by FS , or by F(S, F), or by F(S)) the family of all
functions : S F. For x l(S) we frequently write the value of x at s S as xs rather than
x(s). (Sometimes it seems a good idea to reduce the number of parentheses cluttering a page.)
The support of a function f : S F, denoted by supp(f ), is {x S : f (x) 6= 0}.
Furthermore, we will denote by lc (S) (or by lc (S, F), or by Fc (S)) the family of all functions
: S F with finite support; that is, those functions on S which are nonzero at only finitely many
elements of S.
1.2.19. Exercise. Let S be a set with exactly one element. Discuss the cardinality of (that is, the
number of elements in) the sets F(, ), F(, S),F(S, ), and F(S, S),
1.3. Rings
Recall that an ordered triple (R, +, ) is a ring if (R, +) is an Abelian group, (R, ) is a semigroup,
and the distributive laws (see Some Algebraic Objects 8) hold. The ring is unital if, in addition,
(R \ {0}, ) is a monoid.
1.3.1. Proposition. The additive identity of a ring is an annihilator. That is, for every element
a of a ring 0a = a 0 = 0.
1.3.2. Proposition. If a and b are elements of a ring, then (a)b = a(b) = (ab) and
(a)(b) = ab.
1.3.3. Proposition. Let a and b be elements of a unital ring. Then 1 ab is invertible if and only
if 1 ba is.
Hint for proof . Look at the product of 1 ba and 1 + bca where c is the inverse of 1 ab.
1.3.4. Definition. An element a of a ring is left cancellable if ab = ac implies that b = c. It
is right cancellable if ba = ca implies that b = c. A ring has the cancellation property if
every nonzero element of the ring is both left and right cancellable.
1.3.5. Exercise. Every division ring has the cancellation property.
1.3.6. Definition. A nonzero element a of a ring is a zero divisor (or divisor of zero) if there
exists a nonzero element b of the ring such that (i) ab = 0 or (ii) ba = 0.
Most everyone agrees that a nonzero element a of a ring is a left divisor of zero if it satisfies (i)
for some nonzero b and a right divisor of zero if it satisfies (ii) for some nonzero b. There agreement
on terminology ceases. Some authors ([6], for example) use the definition above for divisor of zero;
others ([18], for example) require a divisor of zero to be both a left and a right divisor of zero; and
yet others ([19], for example) avoid the issue entirely by defining zero divisors only for commutative
rings. Palmer in [27] makes the most systematic distinctions: a zero divisor is defined as above;
an element which is both a left and a right zero divisor is a two-sided zero divisor ; and if the same
nonzero b makes both (i) and (ii) hold a is a joint zero divisor.
10
1. VECTOR SPACES
1.3.7. Proposition. A division ring has no zero divisors. That is, if ab = 0 in a division ring,
then a = 0 or b = 0.
1.3.8. Proposition. A ring has the cancellation property if and only if it has no zero divisors.
1.3.9. Example. Let G be an Abelian group. Then Hom(G) is a unital ring (under the operations
of addition and composition).
1.3.10. Definition. A function f : R S between rings is a (ring) homomorphism if
f (x + y) = f (x) + f (y)
(1.1)
and
f (xy) = f (x)f (y)
for all x and y in R. If in addition R and S are unital rings and
f (1R ) = 1S
(1.2)
(1.3)
then f is a unital (ring) homomorphism.

Obviously a ring homomorphism f : R S is a group homomorphism of R and S regarded as
Abelian groups. The kernel of f as a ring homomorphism is the kernel of f as a homomorphism
of Abelian groups; that is ker f = {x R : f (x) = 0}.
If f 1 exists and is also a ring homomorphism, then f is an isomorphism from R to S. If an
isomorphism from R to S exists, then R and S are isomorphic.
1.4. Vector Spaces
1.4.1. Definition. Let F be a field. An ordered triple (V, +, M ) is a vector space over F if
(V, +) is an Abelian group and M : F Hom(V ) is a unital ring homomorphism. An element of
V is a vector and an element of F is a scalar. A vector space whose scalars are real numbers is
a real vector space and one with complex numbers as scalars is a complex vector space.
The vector space {0} containing a single element is the trivial vector space.
1.4.2. Exercise. The definition of vector space found in many elementary texts is something like
the following: a vector space is a set V together with operations of addition and scalar multiplication
which satisfy the following axioms:
(1) if x, y V , then x + y V ;
(2) (x + y) + z = x + (y + z) for every x, y, z V (associativity);
(3) there exists 0 V such that x + 0 = x for every x V (existence of additive identity);
(4) for every x V there exists x V such that x + (x) = 0 (existence of additive
inverses);
(5) x + y = y + x for every x, y V (commutativity);
(6) if F and x V , then x V ;
(7) (x + y) = x + y for every F and every x, y V ;
(8) ( + )x = x + x for every , F and every x V ;
(9) ()x = (x) for every , F and every x V ; and
(10) 1 x = x for every x V .
Verify that this definition is equivalent to the one given above in 1.4.1.
1.4.3. Proposition. If x is an element of a vector space, then (1)x is the additive inverse of x.
That is, (1)x = x. (Here, of course, 1 is the multiplicative identity of the field F.)
1.4.4. Example. Let F be a field. Then F can be regarded as a vector space over itself.
1.4.5. Example. Let S be a nonempty set and F be a field. In example 1.1.6 we saw that the family
F(S, F) of F-valued functions on S is an Abelian group under pointwise addition. For f F(S, F)
and F define f by
(f )(x) = f (x)
1.4. VECTOR SPACES
11
for all x S. Under this operation (called pointwise scalar multiplication) the Abelian group
F(S, F) becomes a vector space. When F = R we write F(S) for F(S, R).
1.4.6. Example. As a special case of example 1.4.5, we may regard Euclidean n-space Rn as a
vector space.
1.4.7. Example. As another special case of example 1.4.5, we may regard the set R of all
sequences of real numbers as a vector space.
1.4.8. Example. Yet another special case of example 1.4.5, is the vector space Mmn (F) of m n
matrices of members of F (where m, n N). We will use Mn (F) as shorthand for Mnn (F) and
Mn for Mn (R).
1.4.9. Exercise. Let V be the set of all real numbers. Define an operation of addition by
x y = the maximum of x and y
for all x, y V . Define an operation of scalar multiplication by
x = x
for all R and x V . Prove or disprove: under the operations and

space.
the set V is a vector
1.4.10. Exercise. Let V be the set of all real numbers x such that x > 0. Define an operation of
addition by
x y = xy
for all x, y V . Define an operation of scalar multiplication by
x = x
for all R and x V . Prove or disprove: under the operations and

space.
1.4.11. Exercise. Let V be R2 , the set of all ordered pairs (x, y) of real numbers. Define an
operation of addition by
(u, v) (x, y) = (u + x + 1, v + y + 1)
for all (u, v) and (x, y) in V . Define an operation of scalar multiplication by
(x, y) = (x, y)
for all R and (x, y) V . Prove or disprove: under the operations and
space.
1.4.12. Exercise. Let V be the set of all n n matrices of real numbers. Define an operation of
addition by
A B = 21 (AB + BA)
for all A, B V . Define an operation of scalar multiplication by
A=0
for all R and A V . Prove or disprove: under the operations and the set V is a vector
space. (If you have forgotten how to multiply matrices, look in any beginning linear algebra text.)
1.4.13. Proposition. If x is a vector and is a scalar, then x = 0 if and only if = 0 or x = 0.
In example 1.1.9 we saw how to make the family of equivalence classes of directed segments in
the plane into an Abelian group. We may also define scalar multiplication on these equivalence
classes by declaring that

(1) if > 0, then P Q = P R where P , Q, and R are collinear, P does not lie between Q and
R, and the length of the directed segment (P, R) is times the length of (P, Q);
12
1. VECTOR SPACES

(2) if = 0, then P Q = P P ; and

(3) if < 0, then P Q = P R where P , Q, and R are collinear, P does lie between Q and R,
and the length of the directed segment (R, P ) is times the length of (P, Q).
1.4.14. Exercise. Show that the scalar multiplication presented above is well-defined and that
it makes the Abelian group of equivalence classes of directed segments in the plane into a vector
space.
1.4.15. Remark. Among the methods for proving elementary facts about Euclidean geometry of
the plane three of the most common are synthetic geometry, analytic geometry, and vector geometry.
In synthetic geometry points do not have coordinates, lines do not have equations, and vectors
are not mentioned; but standard theorems from Euclids Elements are used. Analytic geometry
makes use of a coordinate system in terms of which points are assigned coordinates and lines are
described by equations; little or no use is made of vectors or major theorems of Euclidean geometry.
Vector geometry uses vectors as defined in the preceding exercise, but does not rely on Euclidean
theorems or coordinate systems. Although there is nothing illogical about mixing these methods in
establishing a result, it is interesting to try to construct separate proofs of some elementary results
using each method in turn. That is what the next four exercises are about.
1.4.16. Exercise. Use each of the three geometric methods described above to show that the
diagonals of a parallelogram bisect each other.
1.4.17. Exercise. Use each of the three geometric methods described above to show that if the
diagonals of a quadrilateral bisect each other then the quadrilateral is a parallelogram.
1.4.18. Exercise. Use each of the three geometric methods described above to show that the line
joining the midpoints of the non-parallel sides of a trapezoid is parallel to the bases and its length
is half the sum of the lengths of the bases.
1.4.19. Exercise. Use each of the three geometric methods described above to show that the line
segments joining the midpoints of adjacent sides of an arbitrary quadrilateral form a parallelogram.
1.4.20. Exercise. Three vertices of a parallelogram P QRS in 3-space are P = (1, 3, 2), Q =
(4, 5, 3), and R = (2, 1, 0). What are the coordinates of the point S, opposite Q?
1.5. Subspaces
1.5.1. Definition. A subset M of a vector space V is a subspace of V if it is a vector space under
the operations it inherits from V .
1.5.2. Notation. For a vector space V we will write M V to indicate that M is a subspace
of V . To distinguish this concept from other uses of the word subspace (topological subspace, for
example) writers frequently use the expressions linear subspace, vector subspace, or linear manifold.
1.5.3. Proposition. A nonempty subset of M of a vector space V is a subspace of V if and only
if it is closed under addition and scalar multiplication. (That is: if x and y belong to M , so does
x + y; and if x belongs to M and F, then x belongs to M .
1.5.4. Example. In each of the following cases prove or disprove that the set of points (x, y, z) in
R3 satisfying the indicated condition is a subspace of R3 .
(a) x + 2y 3z = 4.
x1
y+2
z
(b)
=
= .
2
3
4
(c) x + y + z = 0 and x y + z = 1.
(d) x = z and x = z.
(e) x2 + y 2 = z.
x
y3
(f)
=
.
2
5
1.5. SUBSPACES
13
1.5.5.
Proposition. Let M be a family of subspaces of a vector space V . Then the intersection
T
M of this family is itself a subspace of V .
1.5.6. Exercise. Let A be a nonempty set of vectors in a vector space V . Explain carefully why it
makes sense to say that the intersection of the family of all subspaces containing A is the smallest
subspace of V which contains A.
1.5.7. Exercise. Find and describe geometrically the smallest subspace of R3 containing the
vectors (0, 3, 6) and (0, 1, 2).
1.5.8. Exercise. Find and describe geometrically the smallest subspace of R3 containing the
vectors (2, 3, 3) and (0, 3, 2).
1.5.9. Example. Let R denote the vector space of all sequences of real numbers. (See example 1.4.5.) In each of the following a subset of R is described. Prove or disprove that it is a
subspace of R .
(a) Sequences that have infinitely many zeros (for example, (1, 1, 0, 1, 1, 0, 1, 1, 0, . . . )).
(b) Sequences which are eventually zero. (A sequence (xk ) is eventually zero if there is an
index n0 such that xn = 0 whenever n n0 .)
(c) P
Sequences that are absolutely summable. (A sequence (xk ) is absolutely summable if
k=1 |xk | < .)

(d) Bounded sequences. (A sequence (xk ) is bounded if there is a positive number M such
that |xk | M for every k.)
(e) Decreasing sequences. (A sequence (xk ) is decreasing if xn+1 xn for each n.)
(f) Convergent sequences. (A sequence (xk ) is convergent if there is a number ` such that the
sequence is eventually in every neighborhood of `; that is, if there is a number ` such that
for every > 0 there exists n0 N such that |xn ell| < whenever n n0 .)
(g) Arithmetic progressions. (A sequence (xk ) is arithmetic if it is of the form (a, a + k,
a + 2k, a + 3k, . . . ) for some constant k.)
(h) Geometric progressions. (A sequence (xk ) is geometric if it is of the form (a, ka, k 2 a, k 3 a, . . . )
for some constant k.)
1.5.10. Notation. Here are some frequently encountered families of functions:
F = F[a, b] = {f : f is a real valued function on the interval [a, b]}
(1.4)
P = P[a, b] = {p : p is a polynomial function on [a, b]}
(1.5)
P4 = P4 [a, b] = {p P : the degree of p is less than 4}
(1.6)
Q4 = Q4 [a, b] = {p P : the degree of p is equal to 4}
(1.7)
C = C[a, b] = {f F : f is continuous}
(1.8)
D = D[a, b] = {f F : f is differentiable}
(1.9)
K = K[a, b] = {f F : f is a constant function}
(1.10)
B = B[a, b] = {f F : f is bounded}
(1.11)
J = J [a, b] = {f F : f is integrable}
(1.12)
(A function f F is bounded if there exists a number M 0 such that |f (x)| M for all x in
Rb
[a, b]. It is (Riemann) integrable if it is bounded and a f (x) dx exists.)
1.5.11. Exercise. For a fixed interval [a, b], which sets of functions in the list 1.5.10 are vector
subspaces of which?
1.5.12. Notation. If A and B are subsets of a vector space then the sum of A and B, denoted by
A + B, is defined by
A + B := {a + b : a A and b B}.
14
1. VECTOR SPACES
The set A B is defined similarly. For a set {a} containing a single element we write a + B instead
of {a} + B.
1.5.13. Exercise. Let M and N be subspaces of a vector space V . Consider the following subsets
of V .
(1) M N . (A vector v belongs to M N if it belongs to either M or N .)
(2) M + N .
(3) M \ N (A vector v belongs to M \ N if it belongs to M but not to N .)
(4) M N .
For each of the sets (a)(d) above, either prove that it is a subspace of V or give a counterexample
to show that it need not be a subspace of V .
1.5.14. Definition. Let M and N be subspaces of a vector space V . If M N = {0} and
M + N = V , then V is the (internal) direct sum of M and N . In this case we write
V = M N.
In this case the subspaces M and N are complementary and each is the complement of the
other.
1.5.15. Example. In R3 let M be the line x = y = z, N be the line x = 12 y = 13 z, and L = M + N .
Then L = M N .
1.5.16. Example. Let M be the plane x + y + z = 0 and N be the line x = y = z in R3 . Then
R3 = M N .
1.5.17. Example. Let C = C[1, 1] be the vector space of all continuous real valued functions
on the interval [1, 1]. A function f in C is even if f (x) = f (x) for all x [1, 1]; it is odd
if f (x) = f (x) for all x [1, 1]. Let Co = {f C : f is odd } and Ce = {f C : f is even }.
Then C = Co Ce .
1.5.18. Example. Let C = C[0, 1] be the family of continuous real valued functions on the interval [0, 1]. Define
f1 (t) = t
and
f2 (t) = t4
for 0 t 1. Let M be the set of all functions of the form f1 + f2 where , R. And let N
be the set of all functions g in C which satisfy
Z 1
Z 1
tg(t) dt = 0
and
t4 g(t) dt = 0.
0
Then C = M N .
1.5.19. Exercise. In the preceding example let g(t) = t2 for 0 t 1. Find polynomials f M
and h N such that f = g + h.
1.5.20. Theorem (Vector Decomposition Theorem). Let V be a vector space such that V = M N .
Then for every vector v V there exist unique vectors m M and n N such that v = m + n.
1.5.21. Exercise. Define what it means for a vector space V to be the direct sum of subspaces
M1 , . . . , Mn . Show (using your definition) that if V is the direct sum of these subspaces, then for
every v V there exist unique vectors mk Mk (for k = 1, . . . , n) such that v = m1 + + mn .
1.6. Linear Combinations and Linear Independence
Some authors of linear algebra texts make it appear as if the terms linear dependence and linear
independence, span, and basis pertain only to finite sets of vectors. This is misleading. The terms
should make sense for arbitrary sets. In particular, do not be misled into believing that a basis for
a vector space must be a finite set of vectors (or a sequence of vectors).
1.6. LINEAR COMBINATIONS AND LINEAR INDEPENDENCE
15
1.6.1. Definition. A vector y P

is a linear combination of vectors x1 , . . . , xn if there exist
scalars 1 , . . P
. n such that y = nk=1 k xk . Note: linear combinations are finite sums. The linear
combination nk=1 k xk is trivial if all the coefficients 1 , . . . n are zero. If at least one k is
different from zero, the linear combination is nontrivial.
1.6.2. Example. In R2 the vector (8, 2) is a linear combination of the vectors (1, 1) and (1, 1).
1.6.3. Example. In R3 the vector (1, 2, 3) is not a linear combination of the vectors (1, 1, 0) and
(1, 1, 0).
1.6.4. Definition. Let A be a subset of a vector space V . The span of A is the intersection of the
family of all subspaces of V which contain A. It is denoted by span(A) (or by spanF (A) if we wish
to emphasize the role of the scalar field F). The subset A spans the space V if V = span(A). In
this case we also say that A is a spanning set for V .
1.6.5. Proposition. If A is a nonempty subset of a vector space V , then span A is the set of all
linear combinations of elements of A.
1.6.6. Remark. Occasionally one must consider the not-too-interesting question of what is meant
by the span of the empty set. According to the abstract definition 1.6.4 above it is the intersection
of all the subspaces which contain the empty set. That is, span = {0}. (Had we preferred
proposition 1.6.5 as a more constructive definition of spanthe set of all linear combinations of
elements in then the span of the empty set would have been just itself.)
1.6.7. Example. For each n = 0, 1, 2, . . . define a function pn on R by pn (x) = xn . Let P be the
set of polynomial functions on R. It is a subspace of the vector space of continuous functions on R.
Then P = span{p0 , p1 , p2 , . . . }. The exponential function exp, whose value at x is ex , is not in the
span of the set {p0 , p1 , p2 . . . }.
1.6.8. Definition. A subset A (finite or not) of a vector space is linearly dependent if the zero
vector 0 can be written as a nontrivial linear combination of elements of A; that
P is, if there exist
distinct vectors x1 , . . . , xn A and scalars 1 , . . . , n , not all zero, such that nk=1 k xk = 0. A
subset of a vector space is linearly independent if it is not linearly dependent.
Technically, it is a set of vectors that is linearly dependent or independent. Nevertheless, these
terms are frequently used as if they were properties of the vectors themselves. For instance, if
S = {x1 , . . . , xn } is a finite set of vectors in a vector space, you may see the assertions the set S is
linearly independent and the vectors x1 , . . . xn are linearly independent used interchangeably.
Supersets of linearly dependent sets are linearly dependent and subsets of linearly independent
sets linearly independent.
1.6.9. Proposition. Suppose that V is a vector space and A B V . If A is linearly dependent,
then so is B. Equivalently, if B is linearly independent, then so is A.
1.6.10. Exercise. Let w = (1, 1, 0, 0), x = (1, 0, 1, 0), y = (0, 0, 1, 1), and z = (0, 1, 0, 1).
(a) Show that {w, x, y, z} does not span R4 by finding a vector u in R4 such that u
/
span(w, x, y, z).
(b) Show that {w, x, y, z} is a linearly dependent set of vectors by finding scalars , , , and
not all zerosuch that w + x + y + z = 0.
(c) Show that {w, x, y, z} is a linearly dependent set by writing z as a linear combination of
w, x, and y.
1.6.11. Example. The (vectors going from the origin to) points on the unit circle in R2 are linearly
dependent.
1.6.12. Example. For each n = 0, 1, 2, . . . define a function pn on R by pn (x) = xn . Then the set
{p0 , p1 , p2 , . . . } is a linearly independent subset of the vector space of continuous functions on R.
16
1. VECTOR SPACES
1.6.13. Example. In the vector space C[0, ] of continuous functions on the interval [0, ] define
the vectors f , g, and h by
f (x) = x
g(x) = sin x
h(x) = cos x
for 0 x . Then f , g, and h are linearly independent.
1.6.14. Example. In the vector space C[0, ] of continuous functions on [0, ] let f , g, h, and j
be the vectors defined by
f (x) = 1
g(x) = x
h(x) = cos x
x
j(x) = cos2
2
for 0 x . Then f , g, h, and j are linearly dependent.
1.6.15. Exercise. Let a, b, and c be distinct real numbers. Show that the vectors (1, 1, 1), (a, b, c),
and (a2 , b2 , c2 ) form a linearly independent subset of R3 .
1.6.16. Exercise. In the vector space C[0, 1] define the vectors f , g, and h by
f (x) = x
g(x) = ex
h(x) = ex
for 0 x 1. Are f , g, and h linearly independent?
1.6.17. Exercise. Let u = (, 1, 0), v = (1, , 1), and w = (0, 1, ). Find all values of which
make {u, v, w} a linearly dependent subset of R3 .
1.6.18. Exercise. Suppose that {u, v, w} is a linearly independent set in a vector space V . Show
that the set {u + v, u + w, v + w} is linearly independent in V .
1.7. Bases for Vector Spaces
1.7.1. Definition. A set B (finite or not) of vectors in a vector space V is a basis for V if it is
linearly independent and spans B.
1.7.2. Example. The vectors e1 = (1, 0, 0), e2 = (0, 1, 0), and e3 = (0, 0, 1) constitute an ordered
basis for the vector space R3 . This is the standard basis for R3 . In elementary calculus texts
these vectors are usually called i, j, and k, respectively.
More generally, in Rn for 1 k n let ek be the n-tuple which is zero in every coordinate
except the k th coordinate where is value is 1. Then {e1 , e2 , . . . , en } is the standard basis for Rn .
1.7.3. Example. The space Pn (J) of polynomial functions of degree strictly less than n N on
some interval J R with nonempty interior is a vector space of dimension n. For each positive
integer n = 0, 1, 2, . . . let pn (t) = tn for all t J. Then {p0 , p1 , p2 , . . . , pn1 } is a basis for Pn (J).
We take this to be the standard basis for Pn (J).
1.7.4. Example. The space P(J) of polynomial functions on some interval J R with nonempty
interior is an infinite dimensional vector space. For each n = 0, 1, 2, . . . define a function pn on
J by pn (x) = xn . Then the set {p0 , p1 , p2 , . . . } is a basis for the vector space P(J) of polynomial
functions on J. We take this to be the standard basis for P(J).
1.7. BASES FOR VECTOR SPACES
17
1.7.5. Example. Let Mmn be the vector space of all m n matrices of real numbers. For
1 i m and 1 j n let E ij be the m n matrix whose entry in the ith row and j th column
is 1 and all of whose other entries are 0. Then E ij : 1 i m and 1 j n is a basis for Mmn .

a b
1.7.6. Exercise. A 2 2 matrix
has zero trace if a + d = 0. Show that the set of all such
c d
matrices is a subspace of M22 and find a basis for it.

u u x
1.7.7. Exercise. Let U be the set of all matrices of real numbers of the form
and V
0
x

v 0
be the set of all real matrices of the form
. Find bases for U, V, U + V, and U V.
w v
To show that every nontrivial vector space has a basis.we need to invoke Zorns lemma, a set
theoretic assumption which is equivalent to the axiom of choice. To this end we need to know
about such things as partial orderings, chains, and maximal elements.
1.7.8. Definition. A relation on a set S is a subset of the Cartesian product S S. If the
relation is denoted by , then it is conventional to write x y (or equivalently, y x) rather than
(x, y) .
1.7.9. Definition. A relation on a set S is reflexive if x x for all x S. It is transitive
if x z whenever x y and y z. It is antisymmetric if x = y whenever x y and y x.
A relation which is reflexive, transitive, and antisymmetric is a partial ordering. A partially
ordered set is a set on which a partial ordering has been defined.
1.7.10. Example. The set R of real numbers is a partially ordered set under the usual relation .
1.7.11. Example. A family A of subsets of a set S is a partially ordered set under the relation .
When A is ordered in this fashion it is said to be ordered by inclusion.
1.7.12. Example. Let F(S) be the family of real valued functions defined on a set S. For f ,
g F(S) write f g if f (x) g(x) for every x S. This is a partial ordering on F(S). It is
known as pointwise ordering.
1.7.13. Definition. Let A be a subset of a partially ordered set S. An element u S is an upper
bound for A if a u for every a A. An element m in the partially ordered set S is maximal
if there is no element of the set which is strictly greater than m; that is, m is maximal if c = m
whenever c S and c m. An element m in S is the largest element of S if m s for every
s S.
Similarly an element l S is a lower bound for A if l a for every a A. An element
m in the partially ordered set S is minimal if there is no element of the set which is strictly less
than m; that is, m is minimal if c = m whenever c S and c m. An element m in S is the
smallest element of S if m s for every s S.
1.7.14. Example. Let S = {a, b, c} be a three-element set. The family P(S) of all subsets of S is
partially ordered by inclusion. Then S is the largest element of P(S)and, of course, it is also a
maximal element of P(S). The family Q(S) of all proper subsets of S has no largest element; but
it has three maximal elements {b, c}, {a, c}, and {a, b}.
1.7.15. Proposition. A linearly independent subset of a vector space V is a basis for V if and
only if it is a maximal linearly independent subset.
1.7.16. Proposition. A spanning subset for a nontrivial vector space V is a basis for V if and
only if it is a minimal spanning set for V .
1.7.17. Definition. Let S be a partially ordered set with partial ordering .
18
1. VECTOR SPACES
(1) Elements x and y in S are comparable if either x y or y x.

(2) If is a partial ordering with respect to which any two elements of S are comparable, it
is a linear ordering (or a total ordering) and S is a linearly ordered set.
(3) A linearly ordered subset of S is a chain in S.
1.7.18. Axiom (Zorns lemma). A partially ordered set in which every chain has an upper bound
has a maximal element.
1.7.19. Theorem. Let A be a linearly independent subset of a vector space V . Then there exists
a basis for V which contains A.
Hint for proof . Order the set of linearly independent subsets of V which contain A by inclusion.
Apply Zorns lemma.
1.7.20. Corollary. Every vector space has a basis.
Note that the empty set is a basis for the trivial vector space. (See remark 1.6.6.)
1.7.21. Proposition. Let B be a basis for a vector space V . Every element of V can be written
in a unique way as a linear combination of elements of B.
1.7.22. Notation. Let B be a basis for a vector space V over a field F and x V . By the
preceding proposition there exists a unique finite set
P S of vectors in B and for each element e in
S there exists a unique scalar xe such that
P x = eS xe e. If we agree to let xe = 0 whenever
e B \ S, we can just as well write x = eB xe e. Although this notation may make it appear
as if we are summing over an arbitrary, perhaps uncountable, set, the fact of the matter is that all
but finitely many of the terms are zero.
P The function x : B F : e 7 xe has finite support, so no
convergence problems arise. Treat eB xe e as a finite sum. Associativity and commutativity
of addition in V make the expression unambiguous.
Notice in the preceding that the symbol x ends up denoting two different things: a vector in
V and a function in lc (B, F). We show in proposition 2.2.10 that this identification is harmless. It
is a good idea to teach yourself to feel comfortable with identifying these two objects whenever you
are dealing with a vector space with a basis.
1.7.23. Notation. In finite dimensional vector spaces it is usual to adopt some special notational
conventions. Let V be an n-dimensional vector space with an ordered basis {e1 , e2 , . . . , en }. If
x V , then by proposition 1.7.21 we know that there are unique scalars xe1 , xe2 , . . . xen such that
n
X
x=
xek ek .
k=1
The notation can be unambiguously simplified by writing

n
X
x=
xk ek .
k=1
Since the scalars x1 , x2 , . . . xn uniquely determine the vector x it has become standard to write

x1
x2

x = (x1 , x2 , . . . , xn )
or
x = .. .
.
xn
That is, a vector x in a finite dimensional space with an ordered basis may be represented as an
n-tuple or as an n 1-matrix. The first of these is frequently referred to as a row vector and the
second as a column vector.
Next we verify that every subspace has a complementary subspace.
1.7. BASES FOR VECTOR SPACES
19
1.7.24. Proposition. Let M be a subspace of a vector space V . Then there exists a subspace N
of V such that V = M N .
P
1.7.25. Lemma. Let V be a vector space with a finite basis {e1 , . . . , en } and let v = nk=1 k ek be
a vector in V . If p Nn and p 6= 0, then {e1 , . . . , ep1 , v, ep+1 , . . . , en } is a basis for V .
1.7.26. Proposition. If some basis for a vector space V contains n elements, then every linearly
independent subset of V with n elements is also a basis.
Hint for proof . Suppose {e1 , . . . , en } is a basis for V and {v 1 , . . . , v n } is linearly independent
in V . Start by using lemma 1.7.25 to show that (after perhaps renumbering the ek s) the set
{v 1 , e2 , . . . , en } is a basis for V .
1.7.27. Corollary. If a vector space V has a finite basis B, then every basis for V is finite and
contains the same number of elements as B.
1.7.28. Definition. A vector space is finite dimensional if it has a finite basis and the dimension of the space is the number of elements in this (hence any) basis for the space. The dimension
of a finite dimensional vector space V is denoted by dim V . If the space does not have a finite basis,
it is infinite dimensional.
Corollary 1.7.27 can be generalized to arbitrary vector spaces.
1.7.29. Theorem. If B and C are bases for a vector space V , then B and C are cardinally
equivalent; that is, there exists a bijection from B onto C.
Proof. See [29], page 45, Theorem 1.12.
1.7.30. Proposition. Let V be a vector space and suppose that V = U W . Prove that if B
is a basis for U and C is a basis for W , then B C is a basis for V . From this conclude that
dim V = dim U + dim W .

1.7.31. Definition. The transpose of an nn matrix A = aij is the matrix At = aji obtained
by interchanging the rows and columns of A. The matrix A is symmetric if At = A.
1.7.32. Exercise. Let S3 be the vector space of all symmetric 3 3 matrices of real numbers.
(a) What is the dimension of S3 ?
(b) Find a basis for S3 .
CHAPTER 2
LINEAR TRANSFORMATIONS
2.1. Linearity
2.1.1. Definition. Let V and W be vector spaces over the same field F. A function T : V W is
linear if T (x + y) = T x + T y and T (x) = T x for all x, y V and F. For linear functions
it is a matter of convention to write T x instead of T (x) whenever it does not cause confusion. (Of
course, we would not write T x + y when we intend T (x + y).) Linear functions are frequently called
linear transformations or linear maps.
2.1.2. Notation. If V and W are vector spaces (over the same field F) the family of all linear
functions from V into W is denoted by L(V, W ). Linear functions are frequently called linear
transformations, linear maps, or linear mappings. When V = W we condense the notation L(V, V )
to L(V ) and we call the members of L(V ) operators.
2.1.3. Example. Let V and W be vector spaces over a field F. For S, T L(V, W ) define S + T
by
(S + T )(x) := Sx + T x
for all x V . For T L(V, W ) and F define T by
(T )(x) := (T x)
for all x V . Under these operations L(V, W ) is a vector space.
2.1.4. Proposition. If S : V W and T : W X are linear maps between vector spaces, then
the composite of these two functions, nearly always written as T S rather than T S, is a linear
map from V into X.
2.1.5. Convention. If S : V W and T : W X are linear maps between vector spaces, then
the composite linear map of these two functions is nearly always written as T S rather than T S.
In the same vein, T 2 = T T , T 3 = T T T , and so on.
2.1.6. Exercise. Use the notation of definition 1.4.1 and suppose that (V, +, M ) and (W, +, M )
are vector spaces over a common field F and that T Hom(V, W ) is such that the diagram
V
/W

/W
commutes for every F. What can be said about the homomorphism T ?

2.1.7. Example. Let a < b and C = C([a, b]) be the vector space of all continuous real valued
functions on the interval [a, b]. Then integration
Z b
T : C R : f 7
f (t) dt
a
is a linear map.
21
22
2. LINEAR TRANSFORMATIONS
2.1.8. Example. Let a < b and C 1 = C 1 ([a, b]) be the set of all continuously differentiable real
valued functions on the interval [a, b]. (Recall that a function is continuously differentiable
if it has a derivative and the derivative is continuous.) Then differentiation
D : C 1 C : f 7 f 0
is linear.
2.1.9. Example. Let R be the vector space of all sequences of real numbers and define
S : R R : (x1 , x2 , x3 , . . . ) 7 (0, x1 , x2 , . . . ).
Then S is a linear operator. It is called the unilateral shift operator.
2.1.10. Definition. Let T : V W be a linear transformation between vector spaces. Then ker T ,
the kernel (or nullspace) of T is defined to be the set of all x in V such that T x = 0. Also,
ran T , the range of T (or the image of T ), is the set of all y in W such that y = T x for some x
in V . The rank of T is the dimension of its range and the nullity of T is the dimension of its
kernel.
2.1.11. Definition. Let T : V W be a linear transformation between vector spaces and let A
be a subset of V . Define T (A) := {T x : x A}. This is the (direct) image of A under T .
2.1.12. Proposition. Let T : V W be a linear map between vector spaces and M V . Then
T (M ) is a subspace of W . In particular, the range of a linear map is a subspace of the codomain
of the map.
2.1.13. Definition. Let T : V W be a linear transformation between vector spaces and let B be
a subset of W . Define T (B) := {x V : T x B}. This is the inverse image of B under T .
2.1.14. Proposition. Let T : V W be a linear map between vector spaces and M W . Then
T (M ) is a subspace of V . In particular, the kernel of a linear map is a subspace of the domain
of the map.
2.1.15. Exercise. Let T : R3 R3 : x = (x1 , x2 , x3 ) 7 (x1 + 3x2 2x3 , x1 4x3 , x1 + 6x2 ).
(a) Identify the kernel of T by describing it geometrically and by giving its equation(s).
(b) Identify the range of T by describing it geometrically and by giving its equation(s).
2.1.16. Exercise. Let T be the linear map from R3 to R3 defined by
T (x, y, z) = (2x + 6y 4z, 3x + 9y 6z, 4x + 12y 8z).
Describe the kernel of T geometrically and give its equation(s). Describe the range of T geometrically and give its equation(s).
2.1.17. Exercise. Let C = C[a, b] be the vector space of all continuous real valued functions on
the interval [a, b] and C 1 = C 1 [a, b] be the vector space of all continuously differentiable real valued
functions on [a, b]. Let D : C 1 C be the linear transformation defined by
Df = f 0
and let T : C C 1 be the linear transformation defined by
Z x
f (t) dt
(T f )(x) =
a
for all f
(a)
(b)
(c)
(d)
C and x [a, b].

Compute (and simplify) (DT f )(x).
Compute (and simplify) (T Df )(x).
Find the kernel of T .
Find the range of T .
2.1.18. Proposition. A linear map T : V W between vector spaces is injective if and only if
ker T = {0}.
2.2. INVERTIBLE LINEAR MAPS
23
2.2. Invertible Linear Maps

2.2.1. Notation. In the sequel we will usually denote the identity operator x 7 x on a vector
space V by IV , or just I, rather than by idV .
2.2.2. Definition. A linear map T : V W between vector spaces is left invertible (or has a
left inverse, or is a section) if there exists a linear map L : W V such that LT = IV .
The map T is right invertible (or has a right inverse, or is a retraction) if there exists
a linear map R : W V such that T R = IW . We say that T is invertible (or has an inverse,
or is an isomorphism) if there exists a linear map T 1 : W V which is both a left and a right
inverse for T . If there exists an isomorphism between two vector spaces V and W , we say that the
spaces are isomorphic and we write V
= W.
2.2.3. Exercise. Show that an operator T L(V ) is invertible if it satisfies the equation
T 2 T + IV = 0.
2.2.4. Example. The unilateral shift operator S on the vector space R of all sequences of real
numbers (see 2.1.9) is injective but not surjective. It is left invertible but not right invertible.
2.2.5. Proposition. A linear map T : V W between vector spaces is invertible if and only if
has both a left inverse and a right inverse.
2.2.6. Proposition. A linear map between vector spaces is invertible if and only if it is bijective.
2.2.7. Proposition. Let T : V W and S : W X be linear maps between vector spaces. If T
and S are invertible, then so is ST and (ST )1 = T 1 S 1 .
2.2.8. Proposition. An operator T on a vector space V is invertible if and only if it has a unique
right inverse.
Hint for proof . Consider ST + S IV , where S is the unique right inverse for T .
2.2.9. Example. Every n-dimensional real vector space is isomorphic to Rn .
Hint for proof . Recall the notational conventions made in 1.7.23.
2.2.10. Example. Let B be a basis for a vector space V over a field F. Then V
= lc (B, F).
Hint for proof . Recall the notational conventions made in 1.7.22.
2.2.11. Proposition. Let S, T L(V, W ) where V and W are vector spaces over a field F; and
let B be a basis for V . If S(e) = T (e) for every e B, then S = T .
2.2.12. Proposition. Let V and W be a vector spaces. If V =M N and T : M W is a linear
map, then there exists Tb L(V, W ) such that Tb M = T and Tb N = 0.
We can now make a considerable improvement on proposition 2.2.6.
2.2.13. Proposition. A linear transformation has a left inverse if and only if it is injective.
2.2.14. Proposition. A linear transformation has a right inverse if and only if it is surjective.
2.2.15. Proposition. Let V and W be vector spaces over a field F and B be a basis for V . If
f : B W , then there exists a unique linear map Tf : V W which is an extension of f .
2.2.16. Exercise. Let S be a set, V be a vector space over a field F, and f : S V be a bijection.
Explain how to use f to make S into a vector space isomorphic to V .
24
2.3. Matrix Representations

2.3.1. Proposition. Let T L(V, W ) where V is an n-dimensional vector space and W is an mdimensional vector space and let {e1 , e2 , . . . , en } be an ordered basis for V . Define an m n-matrix
[T ] whose k th column (1 k n) is the column vector T ek (see 1.7.23). Then for each x V we
have
T x = [T ]x.
The displayed equation above requires a little interpretation. The left side is T evaluated at x;
the right side is an m n matrix multiplied by an n 1 matrix (that is, a column vector). Then
the asserted equality is of two m 1 matrices (column vectors).
2.3.2. Definition. If V and W are finite dimensional vector spaces with ordered bases and T
L(V, W ), then the matrix [T ] in the preceding proposition is the matrix representation of T .
2.3.3. Exercise. Let T : R4 R3 be defined by
T x = (x1 3x3 + x4 , 2x1 + x2 + x3 + x4 , 3x2 4x3 + 7x4 )
for every x = (x1 , x2 , x3 , x4 ) R4 .

(a) Find [T ].
1
(b) Find T (1, 2, 1, 3).
2
(c) Independently of part (b) calculate the matrix product [T ]

1 .
(d) Find ker T .
3
(e) Find ran T .
2.3.4. Exercise. Let P4 (R) be the vector space of polynomial functions of degree strictly less
than 4 on R. Consider the linear transformation D2 : P4 P4 : f 7 f 00 .
(a) Find the matrix representation of D2 (with respect to the standard basis for P 4 (R).
(b) Find the kernel of D2 .
(c) Find the range of D2 .
2.3.5. Exercise. Let T : P4 (R) P5 (R) be the linear transformation defined by (T p)(t) = (2 +
3t)p(t) for every p P4 (R) and t R. Find the matrix representation of T with respect to the
standard bases for P4 (R) and P5 (R).
2.3.6. Exercise. Let Pn (R) be the vector space of all polynomial functions on R with degree
strictly less than n. Define T : P3 (R) P5 (R) by
Z xZ u
T f (x) =
p(t) dt du
0
for all x, u R.
(a) Find the matrix representation of the linear map T (with respect to the standard bases
for P3 (R) and P5 (R)).
(b) Find the kernel of T .
(c) Find the range of T .
2.4. Spans, Independence, and Linearity
2.4.1. Proposition. Let T : V W be an injective linear map between vector spaces. If A is a
subset of V such that T (A) is linearly independent, then A is linearly independent.
2.4.2. Proposition. Let T : V W be an injective linear map between vector spaces. If A is a
linearly independent subset of V , then T (A) is a linearly independent subset of W .
2.4.3. Proposition. Let T : V W be a linear map between vector spaces and A V . Then
T (span A) = span T (A).
2.5. DUAL SPACES
25
2.4.4. Proposition. Let T : V W be an injective linear map between vector spaces. If B is a

basis for a subspace U of V , then T (B) is a basis for T (U ).
2.4.5. Proposition. Let T : V W be an injective linear map between vector spaces. If V is
spanned by a set B of vectors and T (B) is a basis for W , then B is a basis for V and T is an
isomorphism.
2.4.6. Exercise. Prove that a linear transformation T : R3 R2 cannot be one-to-one and that
a linear transformation S : R2 R3 cannot be onto. What is the most general version of these
assertions that you can invent (and prove)?
2.4.7. Proposition. Suppose that V and W are finite dimensional vector spaces of the same finite
dimension and that T : V W is a linear map. Then the following are equivalent;
(a) T is injective;
(b) T is surjective; and
(c) T is invertible.
2.5. Dual Spaces
2.5.1. Definition. Let V be a vector space over a field F. A linear map f : V F is a linear
functional on V . The set of linear functionals on V is denoted by V ; that is, V = L(V, F).
The vector space V is the dual space of V .
2.5.2. Convention. Let B be a basis for a vector space V over a field F. Recall that in 1.7.22 we
adopted the notation
X
x=
xe e
eB
where x denotes both an element of V and a scalar valued function on B with finite support. In
example 2.2.10 we justified this identification by establishing that the vector spaces V and lc (B, F)
are isomorphic. Notice that this is an extension of the usual notation in Rn where we write a vector
v in terms of its components:
n
X
x=
xk ek .
k=1
2.5.3. Exercise. According to convention 2.5.2 above, what is the value of fe when e and f are
elements of the basis B?
2.5.4. Proposition. Let V be a vector space with basis B. For every v V define a function v
on V by
X
v (x) =
xe ve
for all x V .
eB
Then
is a linear functional on V .
2.5.5. Notation. In the preceding proposition 2.5.4 the value v (x) of v at x is often written as
hx, vi.
2.5.6. Exercise. Consider the notation 2.5.5 above in the special case that the scalar field F = R.
Then h , i is an inner product on the vector space V . (For a definition of inner product see 5.1.1,)
2.5.7. Exercise. In the special case that the scalar field F = C, things above are usually done a
bit differently. For v V the function v is defined by
X
v (x) = hx, vi =
xe ve .
eB
Why do you think things are done this way?
26
2.5.8. Proposition. Let v be a nonzero vector in a vector space V and B be a basis for V which
contains the vector v. Then there exists a linear functional f V such that f (v) = 1 and f (e) = 0
for every e B \ {v}.
2.5.9. Corollary. Let M be a subspace of a vector space V and v a vector in V which does not
belong to M . Then there exists f V such that f (v) = 1 and f (M ) = {0}.
2.5.10. Corollary. If v is a vector in a vector space V and f (v) = 0 for every f V , then v = 0.
2.5.11. Definition. Let F be a field. A family F of F-valued functions on a set S containing at
least two points separates points of S if for every x, y S such that x 6= y there exists f F
such that f (x) 6= f (y).
2.5.12. Corollary. For every nontrivial vector space V , the dual space V separates points of V .
2.5.13. Proposition. Let V be a vector space with basis B. The map : V V : v 7 v (see
proposition 2.5.4) is linear and injective.
The next result is the Riesz-Frechet theorem for finite dimensional vector spaces with basis. It
is important to keep in mind that the result does not hold for infinite dimensional vector spaces
(see proposition 2.5.18) and that the mapping depends on the basis which has been chosen for
the vector space.
2.5.14. Theorem. Let V be a finite dimensional vector space with basis B. Then the map
defined in the preceding proposition 2.5.13 is an isomorphism. Thus for every f V there exists
a unique vector a V such that a = f .
2.5.15. Definition. Let V be a vector space with basis {e : }. A basis { : } for V
is the dual basis for V if it satisfies

1, if = ;

(e ) =
0, if 6= .
2.5.16. Theorem. Every finite dimensional vector space V with a basis has a unique dual basis
for its dual space. In fact, if {e1 , . . . , en } is a basis for V , then {(e1 ) , . . . , (en ) } is the dual basis
for V .
2.5.17. Corollary. If a vector space V is finite dimensional, then so is its dual space and dim V =
dim V .
In proposition 2.5.13 we showed that the map
: V V : v 7 v
is always an injective linear map. In corollary 2.5.17 we showed that if V is finite dimensional, then
so is V and is an isomorphism between V and V . This is never true in infinite dimensional
spaces.
2.5.18. Proposition. If V is infinite dimensional, then is not an isomorphism.
Hint for proof . Let B be a basis for V . Is there a functional g V such that g(e) = 1 for
every e B? Could such a functional be (x) for some x V ?
2.5.19. Proposition. Let V be a vector space over a field F. For every x in V define
x
b : V F : 7 (x) .
(a) The vector x
b belongs to V for each x V .
(b) Let V be the map from V to V which takes x to x
b. (When no confusion is likely we
write for V , so that (x) = x
b for each x V .) The function is linear.
(c) The function is injective.
2.6. ANNIHILATORS
27
2.5.20. Proposition. If V is a finite dimensional vector space, then the map : V V defined
in the preceding proposition 2.5.19 is an isomorphism.
2.5.21. Proposition. If V is infinite dimensional, then the mapping (defined in 2.5.19) is not
an isomorphism.
Hint for proof . Let B be a basis for V and V be as in proposition 2.5.18. Show that if we
let C0 be {e : e B}, then the set C0 {} is linearly independent and can therefore be extended
to a basis C for V . Find an element in V such that () = 1 and () = 0 for every other
C. Can be x for some x V ?
2.6. Annihilators
2.6.1. Notation. Let V be a vector space and M V . Then
M := {f V : f (x) = 0 for all x M }
We say that M is the annihilator of M . (The reasons for using the familiar orthogonal
complement notation M (usually read M perp) will become apparent when we study inner
product spaces, where orthogonality actually makes sense.)

2.6.2. Exercise. Find the annihilator in R2 of the vector (1, 1) in R2 . (Express your answer in
terms of the standard dual basis for R2 .)

2.6.3. Proposition. Let M and N be subsets of a vector space V . Then
(a) M is a subspace of V .
(b) If M N , then N M .
(c) (span M ) = M .
(d) (M N ) = M N .
2.6.4. Proposition. If M and N be subspaces of a vector space V , then
(M + N ) = M N .
2.6.5. Exercise. Explain why it is necessary in the preceding proposition to assume that M and
N are subspaces of V and not just subsets of V .
2.6.6. Notation. Let V be a vector space and F V . Then
F := {x V : f (x) = 0 for all f F }
We say that F is the pre-annihilator of F .
2.6.7. Proposition. If M is a subspace of a vector space V , then M
= M.
2.6.8. Exercise. Propositions 2.6.3 and 2.6.4 asserted some properties of the annihilator mapping
M 7 M . See to what extent you can prove similar results about the pre-annihilator mapping

F 7 F . What can you say about the set F ?
2.6.9. Proposition. Let V be a finite dimensional vector space and F be a subspace of V . If
F = {0}, then F = V .
CHAPTER 3
THE LANGUAGE OF CATEGORIES

3.1. Objects and Morphisms
3.1.1. Definition. Let A be a class, whose members we call objects. For every pair (S, T ) of
objects we associate a set Mor(S, T ), whose members we call morphisms from S to T . We assume
that Mor(S, T ) and Mor(U, V ) are disjoint unless S = U and T = V .
We suppose further that there is an operation (called composition) that associates with
every Mor(S, T ) and every Mor(T, U ) a morphism Mor(S, U ) in such a way that:
(1) ( ) = ( ) whenever Mor(S, T ), Mor(T, U ), and Mor(U, V );
(2) for every object S there is a morphism IS Mor(S, S) satisfying IS = whenever
Mor(S, T ) and IS = whenever Mor(R, S).
Under these circumstances the class A, together with the associated families of morphisms, is a
category.
/
T for a situation in which S and T are objects in some
We will reserve the notation S
category and is a morphism belonging to Mor(S, T ). As is the case with groups and vector spaces
we usually omit the composition symbol and write for .
3.1.2. Example. The category SET has sets for objects and functions (maps) as morphisms.
3.1.3. Example. The category AbGp has Abelian groups for objects and group homomorphisms
as morphisms. (See proposition 1.1.17.)
3.1.4. Example. The category VEC has vector spaces for objects and linear transformations as
morphisms. (See proposition 2.1.4.)
3.1.5. Example. Let S and T be partially ordered sets. A function f : S T is order preserving if f (x) f (y) in T whenever x y in S. The category POSET has partially ordered sets for
objects and order preserving maps as morphisms.
The preceding examples are examples of concrete categoriesthat is, categories in which the
objects are sets (together, usually, with additional structure) and the morphism are functions
(usually preserving, in some sense, this extra structure). In these notes the categories of interest
are concrete ones. Even so, it may be of interest to see an example of a category that is not
concrete.
3.1.6. Example. Let G be a monoid. Consider a category CG having exactly one object, which
we call ?. Since there is only one object there is only one family of morphisms Mor(?, ?), which
we take to be G. Composition of morphisms is defined to be the monoid multiplication. That is,
a b := ab for all a, b G. Clearly composition is associative and the identity element of G is the
identity morphism. So CG is a category.
3.1.7. Definition. In any concrete category we will call an injective morphism a monomorphism
and a surjective morphism an epimorphism.
3.1.8. CAUTION. The definitions above reflect the original Bourbaki use of the term and are
the ones most commonly adopted by mathematicians outside of category theory where monomorphism means left cancellable and epimorphism means right cancellable. (Notice that the
29
30
3. THE LANGUAGE OF CATEGORIES
terms injective and surjective may not make sense when applied to morphisms in a category that
is not concrete.)
g
f1
f2
/ C is left cancellable if whenever morphisms A

/ B and A
/B
A morphism B
satisfy gf1 = gf2 , then f1 = f2 . Saunders Mac Lane suggested calling left cancellable morphisms
monic morphisms. The distinction between monic morphisms and monomorphisms turns out to
be slight. In these notes almost all of the morphisms we encounter are monic if and only if they are
monomorphisms. As an easy exercise prove that any injective morphism in a (concrete) category
is monic. The converse sometimes fails.
In the same vein Mac Lane suggested calling a right cancellable morphism (that is, a morphism
f
g1
g2
/ C and B
/ C satisfy g1 f = g2 f , then g1 = g2 ) an
/ B such that whenever morphisms B
A
epic morphism. Again it is an easy exercise to show that in a (concrete) category any epimorphism
is epic. The converse, however, fails in some rather common categories.
3.1.9. Definition. The terminology for inverses of morphisms in categories is essentially the same
/
/ S be morphisms in a category. If = IS , then is a
as for functions. Let S
T and T
left inverse of and, equivalently, is a right inverse of . We say that the morphism is
/ S which is both a left and

an isomorphism (or is invertible) if there exists a morphism T
1
a right inverse for . Such a function is denoted by and is called the inverse of .
3.1.10. Proposition. If a morphism in some category has both a left and a right inverse, then it
is invertible.
In any concrete category one can inquire whether every bijective morphism (that is, every map
which is both a monomorphism and an epimorphism) is an isomorphism. We saw in proposition 2.2.6 that in the category VEC the answer is yes. In the next example the answer is no.
3.1.11. Example. In the category POSET of partially ordered sets and order preserving maps
not every bijective morphism is an isomorphism.
3.1.12. Example. If in the category CG of example 3.1.6 the monoid G is a group, then every
morphism in CG is an isomorphism.
3.2. Functors
3.2.1. Definition. If A and B are categories a (covariant) functor F from A to B (written
/ B) is a pair of maps: an object map F which associates with each object S in A an
A
object F (S) in B and a morphism map (also denoted by F ) which associates with each morphism
f Mor(S, T ) in A a morphism F (f ) Mor(F (S), F (T )) in B, in such a way that
(1) F (g f ) = F (g) F (f ) whenever g f is defined in A; and
(2) F (idS ) = idF (S) for every object S in A.
F
/ B differs from the preceding definition

The definition of a contravariant functor A
only in that, first, the morphism map associates with each morphism f Mor(S, T ) in A a
morphism F (f ) Mor(F (T ), F (S)) in B and, second, condition (1) above is replaced by
(10 ) F (g f ) = F (f ) F (g) whenever g f is defined in A.
F
3.2.2. Example. A forgetful functor is a functor that maps objects and morphisms from a
category C to a category C0 with less structure or fewer properties. For example, if V is a vector
space, the functor F which forgets about the operation of scalar multiplication on vector spaces
would map V into the category of Abelian groups. (The Abelian group F (V ) would have the same
set of elements as the vector space V and the same operation of addition, but it would have no
scalar multiplication.) A linear map T : V W between vector spaces would be taken by the
functor F to a group homomorphism F (T ) between the Abelian groups F (V ) and F (W ).
3.2. FUNCTORS
31
Forgetful functor can forget about properties as well. If G is an object in the category of
Abelian groups, the functor which forgets about commutativity in Abelian groups would take G
into the category of groups.
It was mentioned in the preceding section that all the categories that are of interest in these
notes are concrete categories (ones in which the objects are sets with additional structure and the
morphisms are maps which preserve, in some sense, this additional structure). We will have several
occasions to use a special type of forgetful functorone which forgets about all the structure of
the objects except the underlying set and which forgets any structure preserving properties of the
morphisms. If A is an object in some concrete category C, we denote by |A| its underlying set.
f
/ B is a morphism in C we denote by |f | the map from |A| to |B| regarded simply

And if A
as a function between sets. It is easy to see that | | , which takes objects in C to objects in SET
(the category of sets and maps) and morphisms in C to morphisms in SET, is a covariant functor.
In the category VEC of vector spaces and linear maps, for example, | | causes a vector space V
to forget about both its addition and scalar multiplication (|V | is just a set). And if T : V W
is a linear transformation, then |T | : |V | |W | is just a map between setsit has forgotten
about preserving the operations.
3.2.3. Notation. Let f : S T be a function between sets. Then we define f (A) = {f (x) : x
A} and f (B) = {x S : f (x) B}. We say that f (A) is the image of A under f and that
f (B) is the preimage of B under f .
3.2.4. Definition. A partially ordered set is order complete if every nonempty subset has a
supremum (that is, a least upper bound) and an infimum (a greatest lower bound).
3.2.5. Definition. Let S be a set. Then the power set of S, denoted by P(S), is the family of
all subsets of S.
3.2.6. Example (The power set functors). Let S be a nonempty set.
(a) The power set P(S) of S partially ordered by is order complete.
(b) The class of order complete partially ordered sets and order preserving maps is a category.
(c) For each function f between sets let P(f ) = f . Then P is a covariant functor from the
category of sets and functions to the category of order complete partially ordered sets and
order preserving maps.
(d) For each function f between sets let P(f ) = f . Then P is a contravariant functor from
the category of sets and functions to the category of order complete partially ordered sets
and order preserving maps.
3.2.7. Definition. Let T : V W be a linear map between vector spaces. For every g W
let T (g) = g T . Notice that T (g) V . The map T from the vector space W into the vector
space V is the (vector space) adjoint map of T .
3.2.8. CAUTION. In inner product spaces we will use the same notation T for a different map.
If T : V W is a linear map between inner product spaces, then the (inner product space) adjoint
transformation T maps W to V (not W to V ).
3.2.9. Example (The vector space duality functor). Let T L(V, W ) where V and W are vector
spaces over a field F. Then the pair of maps V 7 V and T 7 T is a contravariant functor from
the category of vector spaces and linear maps into itself. Show that (the morphism map of) this
functor is linear. (That is, show that (S + T ) = S + T and (T ) = T for all S, T L(V, W )
and F.)
There are several quite different results that in various texts are labeled as the fundamental
theorem of linear algebra. Many of them seem to me not to be particularly fundamental because
32
they apply only to finite dimensional inner product spaces or, what amounts to the same thing,
matrices. I feel the following result deserves the name because it holds for arbitrary linear maps
between arbitrary vector spaces.
3.2.10. Theorem (Fundamental Theorem of Linear Algebra). For every linear map T : V W
between vector spaces the following hold.
(1) ker T = (ran T ) ;
(2) ran T = (ker T ) ;
(3) ker T = (ran T ) ; and
(4) ran T = (ker T ) .
Hint for proof . Showing in (2) that (ker T ) ran T takes some thought. Let f (ker T ) .
We wish to find a g W such that f = T g. By propositions 2.1.14 and 1.7.24 there exists a
subspace M of V such that V = ker T M . Let : M V be the inclusion mapping of M into V
and T |M be the restriction of T to M . Use proposition 2.2.13 to show that T |M has a left inverse
S : W M . Let g := f S. Use theorem 1.5.20.
3.2.11. Exercise. What is the relationship between a linear map T being injective and its adjoint
T being surjective? between T being surjective and T being injective?
3.3. Universal Mapping Properties
Much of mathematics involves the construction of new objects from old onesthings such
as products, coproducts, quotients, completions, compactifications, and unitizations. Often it is
possibleand highly desirableto characterize such a construction by means of a diagram which
describes what the constructed object does rather than telling what it is or how it is constructed. Such a diagram is a universal mapping diagram and it describes the universal
mapping property of the object being constructed.
Here is a first example of such a property.
3.3.1. Definition. Let F be an object in a concrete category C and : S |F | be an inclusion
map whose domain is a nonempty set S. We say that the object F is free on the set S (or that
F is the free object generated by S) if for every object A in C and every map f : S |A|
there exists a unique morphism fe : F A in C such that |fe | = f .
/ |F |
F
fe
|fe |

|A|
We will be interested in free vector spaces; that is, free objects in the category VEC of vector
spaces and linear maps. Naturally, merely defining a concept does not guarantee its existence. It
turns out, in fact, that free vector spaces exist on arbitrary sets. (See exercise 3.3.5.)
3.3.2. Exercise. In the preceding definition reference to the forgetful functor is often omitted and
the accompanying diagram is often drawn as follows:
/F
fe

It certainly looks a lot simpler. Why do you suppose I opted for the more complicated version?
3.4. PRODUCTS AND COPRODUCTS
33
3.3.3. Proposition. If two objects in some concrete category are free on the same set, then they
are isomorphic.
3.3.4. Definition. Let A be a subset of a nonempty set S and F be a field. Define A : S F,
the characteristic function of A, by
(
1, if x A
A (x) =
0, otherwise
3.3.5. Example. If S is an arbitrary nonempty set and F is a field, then there exists a vector
space V over F which is free on S. This vector space is unique (up to isomorphism).
Hint for proof . Given the set S let V be the set of all F-valued functions on S which have finite
support. Define addition and scalar multiplication pointwise. The map : s 7 {s} of each element
s S to the characteristic function of {s} is the desired injection. To verify that V is free over S it
must be shown that for every vector space W and every function S
linear map V
fe
/ |W | there exists a unique
/ W which makes the following diagram commute.

/ |V |
V
fe
|fe|

|W |
3.3.6. Proposition. Every vector space is free.

Hint for proof . Of course, part of the problem is to specify a set S on which the given vector
space is free.
3.3.7. Exercise. Let S = {a, , #}. Then an expression such as
3a 21 + 2 #
is said to be a formal linear combination of elements of S. Make sense of such expressions.
3.4. Products and Coproducts
In this section we define products and coproducts of vector spaces in terms of universal mapping
properties.
3.4.1. Definition. Let A1 and A2 be objects in a category C. We say that a triple (P, 1 , 2 ),
where P is an object and k : P Ak (k = 1, 2) are morphisms, is a product of A1 and A2 if
for every object B in C and every pair of morphisms fk : B Ak (k = 1, 2) there exists a unique
map g : B P such that fk = k g for k = 1, 2.
B
f1
A1 o
f2

/ A2
A triple (P, j1 , j2 ), where P is an object and jk : Ak P , (k = 1, 2) are morphisms, is a

coproduct of A1 and A2 if for every object B in C and every pair of morphisms Fk : Ak B
34
(k = 1, 2) there exists a unique map G : P B such that Fk = G jk for k = 1, 2.

? BO _
F1
A1
F2
j1
/P o
j2
A2
3.4.2. Proposition. In an arbitrary category products and coproducts (if they exist) are essentially
unique.
Essentially unique means unique up to isomorphism. Thus in the preceding proposition the
claim is that if (P, 1 , 2 ) and (Q, 1 , 2 ) are both products of two given objects, then P
= Q.
3.4.3. Definition. Let V and W be vector spaces over the same field F. To make the Cartesian
product V W into a vector space we define addition by
(v, w) + (v 0 , w0 ) = (v + v 0 , w + w0 )
(where v, v 0 V and w, w0 W ), and we define scalar multiplication by
(v, w) = (v, w)
(where F, v V , and w W ). The resulting vector space we call the (external) direct
sum of V and W . It is conventional to use the same notation V W for external direct sums that
we use for internal direct sums.
3.4.4. Example. The external direct sum of two vector spaces (as defined in 3.4.3) is a vector
space.
3.4.5. Example. In the category of vector spaces and linear maps the external direct sum is both
a product but also a coproduct.
3.4.6. Example. In the category of sets and maps (functions) the product and the coproduct are
not the same.
3.4.7. Proposition. Let U , V , and W be vector spaces. If U
= W , then U V
=W V.
3.4.8. Example. The converse of the preceding proposition is not true.
3.4.9. Definition. Let V0 , V1 , V2 , . . . be vector spaces (over the same field). Then their (ex
L
ternal) direct sum, which is denoted by
Vk , is defined to be the set of all functions
k=0
S
+
v : Z+
k=0 Vk with finite support such that v(k) = vk Vk for each k Z . The usual
pointwise addition and scalar multiplication make this set into a vector space.
3.5. Quotients
/B
3.5.1. Definition. Let A be an object in a concrete category C. A surjective morphism A
in C is a quotient map for A if a function g : B C (in SET) is a morphism (in C) whenever
g is a morphism. An object B in C is a quotient object for A if it is the range of some
quotient map for A.
3.5.2. Proposition. In the category of vector spaces and linear maps every surjective linear map
is a quotient map.
The next item shows how a particular quotient object can be generated by factoring out a
subspace.
3.6. EXACT SEQUENCES
35
3.5.3. Definition. Let M be a subspace of a vector space V . Define an equivalence relation on

V by
xy
if and only if
y x M.
For each x V let [x] be the equivalence class containing x. Let V /M be the set of all equivalence
classes of elements of V . For [x] and [y] in V /M define
[x] + [y] := [x + y]
and for R and [x] V /M define
[x] := [x].
Under these operations V /M becomes a vector space. It is the quotient space of V by M . The
notation V /M is usually read V mod M . The linear map
: V V /M : x 7 [x]
is called the quotient map.
3.5.4. Exercise. Verify the assertions made in definition 3.5.3. In particular, show that is an
equivalence relation, that addition and scalar multiplication of the set of equivalence classes is well
defined, that under these operations V /M is a vector space, and that the quotient map is linear.
The following result is called the fundamental quotient theorem or the first isomorphism theorem
for vector spaces.
3.5.5. Theorem. Let V and W be vector spaces and M V . If T L(V, W ) and ker T M ,
then there exists a unique Te L(V /M , W ) which makes the following diagram commute.
V
T
!
/W
V /M
Te
Furthermore, Te is injective if and only if ker T = M ; and Te is surjective if and only if T is.
3.5.6. Corollary. If T : V W is a linear map between vector spaces, then ran T
= V / ker T .
For obvious reasons the next result is usually called the rank-plus-nullity theorem. (It is also
sometimes listed as part of the fundamental theorem of linear algebra.)
3.5.7. Proposition. Let T : V W be a linear map between vector spaces. If V is finite dimensional, then
rank T + nullity T = dim V.
3.5.8. Corollary. If M is a subspace of a finite dimensional vector space V , then dim V /M =
dim V dim M .
3.6. Exact Sequences
3.6.1. Definition. A sequence of vector spaces and linear maps
/ Vn1
jn
/ Vn
jn+1
/ Vn+1
is said to be exact at Vn if ran jn = ker jn+1 . A sequence is exact if it is exact at each of its
constituent vector spaces. A sequence of vector spaces and linear maps of the form
0
/U
/V
/W
/0
36
is a short exact sequence. (Here 0 denotes the trivial 0-dimensional vector space, and the
unlabeled arrows are the obvious linear maps.)
3.6.2. Proposition. The sequence
j
/U
/V
/W
/0
of vector spaces is exact at U if and only if j is injective. It is exact at W if and only if k is

surjective.
3.6.3. Exercise. Suppose a < b. Let K be the family of constant functions on the interval [a, b],
C 1 be the family of all continuously differentiable functions on [a, b], and C be the family of all
continuous functions on [a, b]. (A function f is said to be continuously differentiable if its
derivative f 0 exists and is continuous.)
Specify linear maps j and k so that the following sequence is short exact:
j
/K
/ C1
/C
/ 0.
3.6.4. Exercise. Let C be the family of all continuous functions on the interval [0, 2]. Let E1 be
the mapping from C into R defined by E1 (f ) = f (1). (The functional E1 is called evaluation at 1.)
Find a subspace F of C such that the following sequence is short exact.
/F
E1
/C
/R
/ 0.
3.6.5. Exercise. If j : U V is an injective linear map between vector spaces, then the sequence
/U
/V
/ V / ran j
/0
is exact.
3.6.6. Example. Let U and V be vector spaces. Then the following sequence is short exact:
/U
/U V
/V
/ 0.
The indicated linear maps are the obvious ones:

1 : U U V : u 7 (u, 0)
and
2 : U V V : (u, v) 7 v.
3.6.7. Proposition. Consider the following diagram in the category of vector spaces and linear
maps.
0
/ U

/ U0
/V
j0

/V 0
/W
/0
k0

/W0
/0
If the rows are exact and the left square commutes, then there exists a unique linear map h : W
W 0 which makes the right square commute.
3.6.8. Proposition (The Short Five Lemma). Consider the following diagram of vector spaces
and linear maps
0
/ U

/ U0
/V
j0

/V 0
/W
/0
k0

/W0
/0
3.7. SOME MISCELLANEOUS RESULTS
37
where the rows are exact and the squares commute. Then the following hold.
(a) If g is surjective, so is h.
(b) If f is surjective and g is injective, then h is injective.
(c) If f and h are surjective, so is g.
(d) If f and h are injective, so is g.
j
/U
3.6.9. Proposition. Show that if 0

and linear maps, then V
= U W.
/V
/W
/ 0 is an exact sequence of vector spaces
Hint for proof . Consider the following diagram and use proposition 3.6.8.
j
/U
/U
/V
i1
/W
/0.
/W
/0
/ U W o
2
i2
The trick is to find the right map g.

3.6.10. Exercise. Prove the converse of the preceding exercise. That is, suppose that U , V , and
W are vector spaces and that V
= U W ; prove that there exist linear maps j and k such that the
/U j /V k /W
sequence 0
Define j and k in terms of g.
/ 0 is exact. Hint. Suppose g : U W V is an isomorphism.
3.6.11. Proposition. Let M and M 0 be subspaces of vector spaces V and V 0 , respectively. Prove
the following.
(a) Every linear map T : V V 0 which satisfies T (M ) M 0 induces a linear map Tb from
V /M into V 0 /M 0 which satisfies Tb(v + M ) = (T v) + M 0 for all v V .
(b) If T is an isomorphism and T (M ) = M 0 , then V /M
= V 0 /M 0 .
j
/V
/W
/ 0 is an exact sequence of vector spaces and
/U
3.6.12. Proposition. If 0
linear maps, then W = V / ran j. Thus, if U V and j is the inclusion map, then W
= V /U .
k
Give two different proofs of the preceding result: one using theorem 3.5.5 and the other using
proposition 3.6.13.
3.6.13. Proposition. The converse of proposition 3.6.12 is also true. That is, if j : U V is
an injective linear map between vector spaces and W
= V / ran j, then there exists a linear map k
which makes the sequence 0
/U
/V
/W
/ 0 exact.
3.6.14. Proposition. If V0 , V1 , . . . , Vn are finite dimensional vector spaces and the sequence
0
is exact, then
/ Vn
dn
/ Vn1
/ ...
/ V1
d1
/ V0
/0
n
X
(1)k dim Vk = 0.
k=0
3.7. Some Miscellaneous Results

3.7.1. Definition. Let T L(V, W ), where V and W are vector spaces. Define coker T , the
cokernel of T , to be W/ ran T .
3.7.2. Proposition. Let U , V , and W be vector spaces. If S L(U, V ), T L(V, W ), then the
sequence
0
is exact.
/ ker S
/ ker T S
/ ker T
/ coker S
/ coker T S
/ coker T
/0
38
3.7.3. Proposition. Let W be a vector space and M V W . Then

(W/M ) (V /M )
= W/V.
Hint for proof . Proposition 3.6.12.
3.7.4. Proposition. Let V be a vector space and M , M 0 V Then
(M + M 0 )/M
= M 0 /(M M 0 ) .
3.7.5. Proposition. Let M be a subspace of a vector space V . Then the following are equivalent:
(a) dim V /M < ;
(b) there exists a finite dimensional subspace F of V such that V = M F ; and
(c) there exists a finite dimensional subspace F of V such that V = M + F .
3.7.6. Exercise. Suppose that a vector space V is the direct sum of subspaces U and W . Some
authors define the codimension of U to be dim W . Others define it to be dim V /U . Show that
these are equivalent.
CHAPTER 4
The Spectral Theorem for Vector Spaces

4.1. Projections
Much of mathematical research consists analyzing complex objects by writing them as a combination of simpler objects. In the case of vector space operators the simpler objects, the fundamental
building blocks, are projection operators.
4.1.1. Definition. Let V be a vector space. An operator E L(V ) is a projection operator
if it is idempotent; that is, if E 2 = E.
4.1.2. Proposition. If E is a projection operator on a vector space V , then
V = ran E ker E.
4.1.3. Proposition. Let V be a vector space and E, F L(V ). If E + F = IV and EF = 0, then
E and F are projection operators and V = ran E ran F .
P
4.1.4. Proposition. Let V be a vector space and E1 , . . . , En L(V ).LIf nk=1 Ek = IV and
Ei Ej = 0 whenever i 6= j, then each Ek is a projection operator and V = nk=1 ran Ek .
4.1.5. Proposition. If E is a projection operator on a vector space V , then ran E = {x V : Ex =
x}.
4.1.6. Proposition. Let E and F be projection operators on a vector space V . Then E + F = IV
if and only if EF = F E = 0 and ker E = ran F .
4.1.7. Definition. Let V be a vector space and suppose that V = M N . We know from an
earlier theorem 1.5.20 that for each v V there exist unique vectors m M and n N such that
v = m + n. Define a function EN M : V V by EN M v = m. The function EN M is called the
projection of V along N onto M . (This terminology is, of course, optimistic. We must prove
that EN M is in fact a projection operator.)
4.1.8. Proposition. If M N is a direct sum decomposition of a vector space V , then the function
EN M defined in 4.1.7 is a projection operator whose range is M and whose kernel is N .
4.1.9. Proposition. If M N is a direct sum decomposition of a vector space V , then EN M +
EM N = IV and EN M EM N = 0.
4.1.10. Proposition. If E is a projection operator on a vector space V , then there exist M , N 4 V
such that E = EN M .
4.1.11. Exercise. Let M be the line y = 2x and N be the y-axis in R2 . Find [EM N ] and [EN M ].
4.1.12. Exercise. Let E be the projection of R3 onto the plane 3x y + 2z = 0 along the z-axis.
Find the matrix representation [E] (of E with respect to the standard basis of R3 ).
4.1.13. Exercise. Let F be the projection of R3 onto the z-axis along the plane 3x y + 2z = 0.
Where does F take the point (4, 5, 1)?
4.1.14. Exercise. Let P be the plane in R3 whose equation is x + 2y z = 0 and L be the line
z
x
= y = . Let E be the projection of R3 along L onto P and F be the
whose equations are
3
2
39
40
4. THE SPECTRAL THEOREM FOR VECTOR SPACES
projection of R3 along P onto L. Then
a
b
c
1
d
d
d
[E] =
3 a 2d b + 2d c + 2d
where a =
,b=
,c=
,d=
3d 3e 3d
1
d e d
[F ] =
3 2d 2e 2d
and
, and e =
4.1.15. Exercise. Let T : V W be linear and S : W V a left inverse for T . Then

(a) W = ran T ker S, and
(b) T S is the projection along ker S onto ran T .
4.2. Algebras
4.2.1. Definition. Let (A, +, M ) be a vector space over a field F which is equipped with another
binary operation : A A A : (a, b) 7 ab in such a way that (A, +, ) is a ring. If additionally
the equations
(ab) = (a)b = a(b)
(4.1)
hold for all a, b A and F, then (A, +, M, ) is an algebra over the field F (sometimes
referred to as a linear associative algebra). We abuse notation in the usual way by writing
such things as, Let A be an algebra. We say that an algebra A is unital if its underlying ring
(A, +, ) is. And it is commutative if its ring is.
4.2.2. Example. A field may be regarded as an algebra over itself.
4.2.3. Example. If S is a nonempty set, then the vector space F(S, F) (see example 1.4.5) is
a commutative unital algebra under pointwise multiplication, which is defined for all f , g
F(S, F) by
(f g)(s) = f (s) g(s)
for all s S. The constant function 1 (that is, the function whose value at each s S is 1) is the
multiplicative identity.
4.2.4. Example. If V is a vector space, then the set L(V ) of linear operators on V is a unital
algebra under pointwise addition, pointwise scalar multiplication, and composition.
4.2.5. Notation. In the following material we make the notational convention that if B and C are
subsets of (a ring or) an algebra A, then BC denotes the set of all sums of products of elements in
B and C. That is
BC := {b1 c1 + + bn cn : n N; b1 , . . . , bn B; and c1 , . . . , cn C}.
And, of course, if b A, then bC = {b}C.
4.2.6. Definition. A map f : A B between algebras is an (algebra) homomorphism if it
is a linear map between A and B as vector spaces which preserves multiplication (in the sense of
equation (1.2). In other words, an algebra homomorphism is a linear ring homomorphism. It is
a unital (algebra) homomorphism if it preserves identities (as in (1.3)). The kernel of an
algebra homomorphism f : A B is, of course, {a A : f (a) = 0}.
If f 1 exists and is also an algebra homomorphism, then f is an isomorphism from A to B.
If an isomorphism from A to B exists, then A and B are isomorphic.
Here are three essentially obvious facts about algebra homomorphisms.
4.2.7. Proposition. Every bijective algebra (or ring) homomorphism is an isomorphism.
4.2.8. Proposition. If f : A B is an isomorphism between algebras (or rings) and A is unital,
then so is B and f is a unital homomorphism.
4.2. ALGEBRAS
41
4.2.9. Proposition. Let A, B, and C be algebras (or rings). If f : A B and g : B C are

homomorphisms, so is gf : A C. (As is the case with group homomorphism and linear maps, gf
denotes the composite function g f .) If f and g are unital, so is gf .
Here is an example of another important algebra.
4.2.10. Example. We have
seen in example
1.4.8 that the set Mn of nn matrices of real numbers
is a vector space. If a = aij and b = bkl are n n matrices of real numbers, then the product
P
of a and b is the nn matrix c = ab whose entry in the ith row and k th column is cik = nj=1 aij bjk .
This definition makes Mn into a unital algebra.
Hint for proof . Proving associativity of matrix multiplication can be something of a nuisance if
one charges ahead without thinking. As an alternative to brute calculation look at exercise 2.2.16.
4.2.11. Definition. A subset of an algebra A which is closed under the operations of addition,
multiplication, and scalar multiplication is a subalgebra of A. If A is a unital algebra and B is a
subalgebra of A which contains the multiplicative identity of A, then B is a unital subalgebra
of A.
4.2.12. CAUTION. Be very careful with the preceding definition. It is possible for B to be a
subalgebra of an algebra A and to be a unital algebra but still not be a unital subalgebra of A!
The definition requires that for B to be a unital subalgebra of A the identity of B must be the
same as the identity of A. Example: Under pointwise operations A = R2 is a unital algebra. The
set B = {(x, 0) : x R} is a subalgebra of A. And certainly B is unital (the element (1, 0) is the
multiplicative identity of B). But B is not a unital subalgebra of A because it does not contain
the multiplicative identity (1, 1) of A.
4.2.13. Example. Let S be a nonempty set. The family B(S) of all bounded real valued functions
on S is a unital subalgebra of the algebra F(S) of all real valued functions on S.
4.2.14. Definition. A left ideal in an algebra A is a vector subspace J of A such that AJ J.
(For right ideals, of course, we require JA J.) We say that J is an ideal if it is a two-sided
ideal; that is, both a left and a right ideal. A proper ideal is an ideal which is a proper subset
of A.
The ideals {0} and A are often referred to as the trivial ideals of A. The algebra A is simple
if it has no nontrivial ideals.
4.2.15. Example. If : A B is an algebra homomorphism, then the kernel of is an ideal in
A and the range of is a subalgebra of B.
4.2.16. Definition. An element a of a unital algebra A is invertible if there exists an element
a1 A such that aa1 = a1 a = 1A .
4.2.17. Proposition. If a and b are invertible elements in a unital algebra, then ab is also invertible
and (ab)1 = b1 a1 .
4.2.18. Proposition. No invertible element in a unital algebra can belong to a proper ideal.
4.2.19. Proposition. Let a be an element of a commutative algebra A. Then aA is an ideal in A.
If A is unital and a is not invertible, then aA is a proper ideal in A.
T
4.2.20. Example. Let J be a family of ideals in an algebra A. Then the J is an ideal in A.
4.2.21. Definition. Let a be an element of an algebra A. Then the intersection of all the ideals
of A which contain a is the principal ideal generated by a.
4.2.22. Proposition. Let a be an element of a commutative algebra A. The ideal aA in proposition 4.2.19 is the principal ideal generated by a.
42
4.3. Quotients and Unitizations

4.3.1. Definition. Let J be a proper ideal in an algebra A. Define an equivalence relation on
A by
ab
if and only if
b a J.
For each a A let [a] be the equivalence class containing a. Let A/J be the set of all equivalence
classes of elements of A. For [a] and [b] in A/J define
[a] + [b] := [a + b]
and
[a][b] := [ab]
and for C and [a] A/J define

[a] := [a] .
Under these operations A/J becomes an algebra. It is the quotient algebra of A by J. The
notation A/J is usually read A mod J. The surjective algebra homomorphism
: A A/J : a 7 [a]
is called the quotient map.
4.3.2. Exercise. Verify the assertions made in the preceding definition.
4.3.3. Definition. Let A be an algebra over a field F. The unitization of A is the unital algebra
A = A F in which addition and scalar multiplication are defined pointwise and multiplication is
defined by
(a, ) (b, ) = (ab + a + b, ).
4.3.4. Exercise. Prove that the unitization A of an algebra A is in fact a unital algebra with (0, 1)
as its identity. Prove also that A is (isomorphic to) a subalgebra of A with codimension 1.
4.4. The Spectrum
4.4.1. Definition. Let a be an element of a unital algebra A over a field F. The spectrum of a,
denoted by A (a) or just (a), is the set of all F such that a 1 is not invertible.
If the algebra A is not unital we will still speak of the spectrum of the element a with the
understanding that we are speaking of the spectrum of a in the unitization of A.
4.4.2. Example. If z is an element of the algebra C of complex numbers, then (z) = {z}.
4.4.3. Example. Let f be an element of the algebra C([a, b]) of continuous complex valued functions on the interval [a, b]. Then the spectrum of f is its range.
4.4.4. Example. The operator which rotates (the real vector space) R2 by
spectrum.
radians has empty
For the next example you may assume that a square matrix of real or complex numbers is
invertible if and only if its determinant is nonzero.
4.4.5. Example. The family M3 (C) of 33 matrices of complex
numbers is a unital algebra under
5 6 6
2 is {1, 2}.
the usual matrix operations. The spectrum of the matrix 1 4
3 6 4
4.4.6. Example. Let a be an element of a unital complex algebra such that a2 = 1. Then either
(i) a = 1, in which case (a) = {1}, or
(ii) a = 1, in which case (a) = {1}, or
(iii) (a) = {1, 1}.
Hint for proof . In (iii) to prove (a) {1, 1}, consider
1
(a + 1).
1 2
4.5. POLYNOMIALS
43
4.4.7. Definition. An element a of an algebra is idempotent if a2 = a.

4.4.8. Example. Let a be an idempotent element of a unital complex algebra. Then either
(i) a = 1, in which case (a) = {1}, or
(ii) a = 0, in which case (a) = {0}, or
(iii) (a) = {0, 1}.

1
Hint for proof . In (iii) to prove (a) {0, 1}, consider
a
+
(
1)1
.
2
4.5. Polynomials
4.5.1. Notation. If S is a set and A is an algebra, l(S, A) denotes the vector space of all functions
from S into A with pointwise operations of addition and scalar multiplication, and lc (S, A) denotes
the subspace of functions with finite support.
4.5.2. Definition. Let A be a unital commutative algebra. On the vector space l(Z+ , A) define a
n
P
P
binary operation (often called convolution) by (f g)n =
fj g k =
fj g nj (where f ,
j+k=n
l(Z+ , A)
Z+ .
j=0
l(Z+ , A)
g
and n
An element of
is a formal power series (with coefficients
in A) and an element of lc (Z+ , A) is a polynomial (with coefficients in A). The reason for using
the expression with coefficients in A may seem at first a bit mysterious. For an explanation look
at example 4.5.10.
4.5.3. Proposition. If A is a unital commutative algebra, then under the operations defined
in 4.5.2 l(Z+ , A) is a unital commutative algebra (whose multiplicative identity is the sequence
(1A , 0, 0, 0, . . . )) and lc (Z+ , A) is a unital subalgebra of l(Z+ , A).
4.5.4. Proposition. If : A B is a unital algebra homomorphism between unital commutative
algebras, then the map

l(Z+ , ) : l(Z+ , A) l(Z+ , B) : f 7 (fn ) n=0
is also a unital homomorphism of unital commutative algebras. The pair of maps A 7 l(Z+ , A) and
7 l(Z+ , ) is a covariant functor from the category of unital commutative algebras and unital
algebra homomorphisms to itself.
4.5.5. Remark. We regard the algebra A as a subset of l(Z+ , A) by identifying the element
a A with the element (a, 0, 0, 0, . . . ) l(Z+ , A). Thus the map a 7 (a, 0, 0, 0, . . . ) becomes an
inclusion map. (Technically speaking, of course, the map : a 7 (a, 0, 0, 0, . . . ) is an injective
unital homomorphism and A
= ran .)
4.5.6. Convention. In the algebra l(Z+ , A) we will normally write ab for a b.
4.5.7. Definition. Let A be a unital commutative algebra. In the algebra l(Z+ , A) of formal power
series the special sequence x = (0, 1A , 0, 0, 0, . . . ) is called the indeterminant of l(Z+ , A). Notice
that the sequence x2 = x x = (0, 0, 1, 0, 0, 0, . . . ), the sequence x3 = x x x = (0, 0, 0, 1, 0, 0, 0, . . . ),
and so on. For each n N the sequence xn = x x x x (n factors) has the property that its nth
coordinate xn n is 1 while its k th coordinate xn k is 0 whenever k 6= n. It is conventional to take x0
to be the multiplicative identity (1A , 0, 0, 0, . . . ) in l(Z+ , A).
4.5.8. Remark. The algebra l(Z+ , A) of formal
power series with coefficients in a unital commutative algebra A is frequently denoted by A [x] and the subalgebra lc (Z+ , A) of polynomials is
denoted by A[x].
For many algebraists scalar
multiplication is of little interest so A is taken to be a unital
commutative ring, so that A [x] is ring of formal power series (with coefficients in A) and A[x] is
the polynomial ring (with coefficients in A). We will be primarily interested in the case where A
is a field F. Since a field can be regarded as a one-dimensional vector space over itself, it is also
44
an algebra. Thus we will take F[x] to be the polynomial algebra with coefficients in F; it has as its
basis {xn : n = 0, 1, 2, . . . }.
4.5.9. Definition. A nonzero polynomial p, being an element of lc (Z+ , A), has finite support. So
there exists n0 Z+ such that pn = 0 whenever n > n0 . The smallest such n0 is the degree of
the polynomial. We denote it by deg p. A polynomial of degree 0 is a constant polynomial.
The zero polynomial (the additive identity of l(Z+ , A)) is a special case; while it is also a constant
polynomial some authors assign it no degree whatever, while others let its degree be .
If p is a polynomial of degree n, then pn is the leading coefficient of p. A polynomial is
monic if its leading coefficient is 1.
4.5.10. Example. Let A be a unital commutative algebra. If p is a nonzero polynomial in lc (Z+ , A),
then
n
X
p=
pk xk
where n = deg p.
k=0
This is the standard form of the polynomial p. Keep in mind that each coefficient pk belongs to
P
the algebra A. Also notice that it does not really matter whether we write p as nk=0 pk xk or as
P
P
k
p k xk .
k=0 pk x ; so frequently we write just
4.5.11. Remark. Recall that there is occasionally a slight ambiguity in notation for sets. For
example, if we consider the (complex) solutions to an algebraic equation E of degree n, we know
that, counting multiplicities, there are n solutions to the equation. So it is common practice to
write, Let {x1 , x2 , . . . , xn } be the set of solutions to E. Notice that in this context there may be
repeated elements of the set. The cardinality of the set may be strictly less than n. However, when
we encounter the expression, Let {x1 , x2 , . . . , xn } be a set of . . . , it is usually the intention of the
author that the elements are distinct, that the cardinality of the set is n.
P
A similar ambiguity arises in polynomial notation. If, for example, p = nk=0 pk xk and q =
Pn
Pn
k
k=0 (pk +
k=0 qk x are both polynomials of degree n, we ordinarily write their sum as p + q =
k
qk )x even though the resulting sum may very well have degree strictly less than n. On the other
P
hand when one sees, Consider a polynomial p = nk=0 pk xk such that . . . , it is usually intended
that p have degree n; that is, that p is written in standard form.
4.5.12. Proposition. If p and q are polynomials with coefficients in a unital commutative algebra
A, then
(i) deg(p + q) max{deg p, deg q}, and
(ii) deg(pq) deg p + deg q.
If A is a field, then equality holds in (ii).
4.5.13. Example. If A is a unital commutative algebra, then so is l(A, A) under pointwise operations of addition, multiplication, and scalar multiplication.
4.5.14.
P Definition. Let A be a unital commutative algebra over a field F. For each polynomial
p = nk=0 pk xk with coefficients in F define
pe: A A : a 7
n
X
pk ak .
k=0
Then pe is the polynomial function on A determined by the polynomial p. Also for fixed a A
define
a : F[x] A : p 7 pe(a).
The mapping a is the polynomial functional calculus determined by the element a.
It is important to distinguish between the concepts of polynomials with coefficients in an algebra
and polynomial functions. Also important is the distinction between the indeterminant x in l(Z+ , A)
and x used as a variable for a polynomial function. (See 4.5.18.)
4.6. MINIMAL POLYNOMIALS
45
4.5.15. Exercise. Let A be a unital commutative algebra over a field F. Then for each a A the
polynomial functional calculus a : F[x] A defined in 4.5.14 is a unital algebra homomorphism.
4.5.16. Proposition. Let A be a unital commutative algebra over a field F. The map
: F[x] l(A, A) : p 7 pe
is a unital algebra homomorphism.
4.5.17. Exercise. Under the homomorphism (defined in 4.5.16) what is the image of the indeterminant x? Under the homomorphism a (defined in 4.5.14) what is the image of the indeterminant x?
The following example is intended to illustrate the importance of distinguishing between polynomials and polynomial functions.
4.5.18. Example. Let F = {0, 1} be the two-element field. The polynomials p = x + x2 + x3 and
q = x in the polynomial algebra F[x] show that the homomorphism (defined in 4.5.16) need not
be injective.
4.5.19. Proposition. Let A be a unital algebra of finite dimension m over a field F. For every
a A there exists a polynomial p F[x] such that 1 deg p m and pe(a) = 0.
4.6. Minimal Polynomials
4.6.1. Definition. Let V be a vector space over a field F and T L(V ). A nonzero polynomial
p F[x] such that pe(T ) = 0 is an annihilating polynomial for T . A monic polynomial of
smallest degree that annihilates T is a minimal polynomial for T .
4.6.2. Proposition. Let V be a finite dimensional vector space over a field F. Then every T L(V )
has a minimal polynomial.
Hint for proof . Use example 4.5.19.
4.6.3. Theorem (Division Algorithm). Let f and d be polynomials with coefficients in a field F
with d 6= 0. Then there exist unique polynomials q and r in F[x] such that
(i) f = dq + r and
(ii) r = 0 or deg r < deg d.
P
P
j
Hint for proof . Let f = kj=0 fj xj and d = m
j=0 dj x be in standard form. The case k < m
is trivial. For k m suppose the result to be true for all polynomials of degree strictly less than
km d?
k. What can you say about fb = f p where p = (fk d1
m )x
4.6.4. Notation. If T is an operator on a finite dimensional vector space over a field F, we denote
its minimal polynomial in F[x] by mT .
4.6.5. Proposition. Let V be a finite dimensional vector space over a field F and T L(V ). Then
the minimal polynomial mT for T is unique.
4.6.6. Proposition. An operator T on a finite dimensional vector space is invertible if and only
if the constant term of its minimal polynomial is not zero.
4.6.7. Exercise. Explain how, for an invertible operator T on a finite dimensional vector space,
we can write its inverse as a polynomial in T .
4.6.8. Definition. If F is a field and p, p1 F[x], we say that p1 divides p if there exists q F[x]
such that p = p1 q.
4.6.9. Proposition. Let T be an operator on a finite dimensional vector space over a field F. If
p F[x] and pe(T ) = 0, then mT divides p.
46
4.6.10. Definition. A polynomial p F[x] is reducible over F if there exist polynomials f ,

g F[x] both of degree at least one such that p = f g. A polynomial p of degree at least one is
irreducible (or prime) over F provided that whenever p = f g with f , g F[x], then either f or
g is constant. That is, a polynomial p of degree at least one is irreducible if and only if it is not
reducible.
2
4.6.11. Example. Let T be the operator
on the
real vector space R whose matrix representation
0 1
(with respect to the standard basis) is
. Find the minimal polynomial mT of T and show
1 0
that it is irreducible (over R).
4.6.12. Example. Let T be the operator on the complex

vector space C2 whose matrix represen
0 1
tation (with respect to the standard basis) is
. Find the minimal polynomial mT of T and
1 0
show that it is reducible (over C).
4.6.13. Definition. A field F is algebraically closed if every prime polynomial in F[x] has
degree 1.
4.6.14. Example. The field R of real numbers is not algebraically closed.
4.6.15. Proposition. Let F be a field and p be a polynomial of degree m 1 in F[x]. If Jp is the
principal ideal generated by p in F[x], then dim F[x]/Jp = m.
Hint for proof . See proposition 4.2.22. Show that B = { [xk ] : k = 0, 1, . . . , m 1} is a basis
for the vector space F[x]/Jp .
4.6.16. Corollary. Let T be an operator on a finite dimensional vector space V over a field F,
: F[x] L(V ) be its associated polynomial functional calculus, and Jm be the principal ideal
T
generated by its minimal polynomial. Then the sequence
0
/ Jm
/ F[x]
/ ran
/0
is exact. Furthermore, the dimension of the range of the functional calculus associated with the
operator T is the degree of its minimal polynomial.
4.6.17. Definition. Let t0 , t1 , . . . , tn be distinct elements of a field F. For 0 k n define
pk F[x] by
n
Y
x tj
pk =
.
tk tj
j=0
j6=k
4.6.18. Proposition (Lagrange Interpolation Formula). The polynomials defined in 4.6.17 form
a basis for the vector space V of all polynomials with coefficients in F and degree less than or equal
to n and that for each polynomial q V
q=
n
X
q(tk )pk .
k=0
4.6.19. Exercise. Use the Lagrange Interpolation Formula to find the polynomial with coefficients
in R and degree no greater than 3 whose values at 1, 0, 1, and 2 are, respectively, 6, 2, 2,
and 6.
4.6.20. Proposition. Let F be a field and p, q, and r be polynomials in F[x]. If p is a prime in
F[x] and p divides qr, then p divides q or p divides r.
4.6.21. Proposition. Let F be a field. Then every nonzero ideal in F[x] is principal.
4.7. INVARIANT SUBSPACES
47
Hint for proof . If J is a nonzero ideal in F[x] consider the principal ideal generated by any
member of J of smallest degree.
4.6.22. Definition. Let p1 , . . . , pn be polynomials, not all zero, with coefficients in a field F. A
monic polynomial d such that d divides each pk (k = 1, . . . , n) and such that any polynomial which
divides each pk also divides d is the greatest common divisor of the pk s. The polynomials pk
are relatively prime if their greatest common denominator is 1.
4.6.23. Proposition. Any finite set of polynomials (not all zero), with coefficients in a field F has
a greatest common divisor.
4.6.24. Theorem (Unique Factorization). Let F be a field. A nonconstant monic polynomial in
F[x] can be factored in exactly one way (except for the order of the factors) as a product of monic
primes in F[x] .
4.6.25. Definition. Let F be a field and p(x) F[x]. An element r F is a root of p(x) if
p(r) = 0.
4.6.26. Proposition. Let F be a field and p(x) F[x]. Then r is a root of p(x) if and only if x r
is a factor of p(x).
4.7. Invariant Subspaces
4.7.1. Definition. Let T be an operator on a vector space V . A subspace M of V is invariant
under T (or T -invariant) if T (M ) M . Since the subspaces {0} and V are invariant under
any operator on V , they are called the trivial invariant subspaces.
3 4 2
4.7.2. Exercise. Let S be the operator on R3 whose matrix representation is 0 1 2. Find
0 0 0
3
three one dimensional subspaces U , V , and W of R which are invariant under S.
0 0 2
4.7.3. Exercise. Let T be the operator on R3 whose matrix representation is 0 2 0. Find a
2 0 0
two dimensional subspace U of R3 which is invariant under T .
4.7.4. Exercise. Find infinitely many subspaces of the vector space of polynomial functions on R
which are invariant under the differentiation operator.
4.7.5. Definition. An operator T on a vector space V is reduced by a pair (M, N ) of subspaces
M and N of V if V = M N and both M and N are invariant under T . In this case M and N
are reducing subspaces for T .
2
0 0
4.7.6. Exercise. Let T be the operator on R3 whose matrix representation is 1 3 2. Find
1 1 0
a plane and a line in R3 which reduce T .
4.7.7. Proposition. Let M be a subspace of a vector space V and T L(V ). If M is invariant
under T , then ET E = T E for every projection E onto M . And if ET E = T E for some projection
E onto M , then M is invariant under T .
4.7.8. Proposition. Suppose a vector space V has the direct sum decomposition V = M N .
Then an operator T on V is reduced by the pair (M, N ) if and only if ET = T E, where E = EM N
is the projection along M onto N .
4.7.9. Proposition. Suppose a finite dimensional vector space V has the direct sum decomposition
V = M N and that E = EM N is the projection along M onto N . Show that E is the projection
in L(V ) along N onto M .
48
4.7.10. Proposition. Let M and N be complementary subspaces of a vector space V (that is, V
is the direct sum of M and N ) and let T be an operator on V . If M is invariant under T , then
M is invariant under T and if T is reduced by the pair (M, N ), then T is reduced by the pair
(M , N ).
4.8. Burnsides Theorem

4.8.1. Notation. Let V be a vector space. For T L(V ) let
Lat T := {M V : M is invariant under T }.
If T L(V ) let
Lat T :=
Lat T.
T T
We say that Lat T (or Lat T) is trivial if it contains only the trivial invariant subspaces {0} and V .
4.8.2. Example. If V is a vector space, then Lat L(V ) is trivial.
Hint for proof . For dim V 2 let M be a nonzero proper subspace of V . Choose nonzero
vectors x M and y M c . Define T : V V : v
7 f (v)y where f is a functional in V such that
f (x) = 1.
2
4.8.3. Example.
Let

A be the subalgebra of L(R ) whose members have matrix representations
a b
of the form
. Then Lat A is trivial.
b a
2
4.8.4. Example.
Let

A be the subalgebra of L(C ) whose members have matrix representations
a b
of the form
. Then Lat A is not trivial.
b a
Hint for proof . Try span{(1, i)}.

4.8.5. Definition. Let V be a vector space. A subalgebra A of L(V ) is transitive if for every
x 6= 0 and y in V there exists an operator T in A such that y = T x.
4.8.6. Proposition. Let V be a vector space. A subalgebra A of L(V ) is transitive if and only if
Lat A is trivial.
4.8.7. Definition. A field F is algebraically closed if every nonconstant polynomial in F[x]
has a root in F.
4.8.8. Example. The field C of complex numbers is algebraically closed; the field R of real numbers
is not.
4.8.9. Theorem (Burnsides Theorem). Let V be a finite dimensional vector space over an algebraically closed field. Then L(V ) has no proper subalgebra which is transitive.
Proof. See [13], theorem 3.15.
4.8.10. Corollary. Let V be a finite dimensional complex vector space of dimension at least 2.
Then every proper subalgebra of L(V ) has a nontrivial invariant subspace.
4.8.11. Example. The preceding result does not hold for real vector spaces.
4.9. EIGENVALUES AND EIGENVECTORS
49
4.9. Eigenvalues and Eigenvectors

4.9.1. Definition. Suppose that on a vector space V there exist projection operators E1 , . . . , En
such that
(i) IV = E1 + E2 + + En and
(ii) Ei Ej = 0 whenever i 6= j.
Then we say that the family {E1 , E2 , . . . , En } of projections is a resolution of the identity.
Recall that it was shown in proposition
Ln 4.1.4 that if {E1 , E2 , . . . , En } is a resolution of the
identity on a vector space V , then V = k=1 ran Ek .
4.9.2. Definition. Let M1 Mn be a direct sum decomposition of a vector space V . For
each k Nn let Nk be the following subspace of V complementary to Mk :
Nk := M1 Mk1 Mk+1 Mn .
Also (for each k) let
Ek := EN
k Mk
be the projection onto Mk along the complementary subspace Nk . The projections E1 , . . . En are
the projections associated with the direct sum decomposition V = M1 Mn .
4.9.3. Proposition. If M1 Mn is a direct sum decomposition of a vector space V , then the
family {E1 , E2 , . . . , En } of the associated projections is a resolution of the identity.
In the following definition we make use of the familiar notion of the determinant of a matrix
even though we have not yet developed the theory of determinants. We will eventually do this.
4.9.4. Definition. Let V be a vector space over a field F and T L(V ). An element F
is an eigenvalue of T if ker(T I) 6= {0}. The collection of all eigenvalues of T is its point
spectrum, denoted by p (T ).
4.9.5. Definition. If F is a field and A is an n n matrix of elements of F, we define the
characteristic polynomial cA of A to be the determinant of A xI. (Note that det(A xI)
is a polynomial.) Some authors prefer the characteristic polynomial to be monic, and consequently
define it to be the determinant of xI A. As you would expect, the characteristic polynomial
cT of an operator T on a finite dimensional space (with basis B) is the characteristic polynomial
of the matrix representation of that operator (with respect to B). Making use of some standard
facts (which we have not yet proved) about determinants (see section 7.3) we see that F is an
eigenvalue of the matrix A (or of its associated linear transformation) if and only if it is a root of
the characteristic polynomial cA .
4.9.6. Proposition. If T is an operator on a finite dimensional vector space, then p (T ) = (T ).
1 1 1
4.9.7. Exercise. Let A = 1 1 1.
1 1 1
The characteristic polynomial of A is p ( 3)q where p =
The minimal polynomial of A is r ( 3)s where r =
and q =
and s =
.
.
0
1
2
3
4.9.8. Exercise. Let T be the operator on R4 whose matrix representation is
2 1
p
2 1
.
The characteristic polynomial of T is ( 2) where p =
The minimal polynomial of T is ( 2)r where r =
0 1
0 1
.
2 1
0 3
50
0 1 0
4.9.9. Exercise. Choose a, b and c in the matrix A = 0 0 1 so that the characteristic polya b c
nomial of A is 3 + 42 + 5 + 6.
4.9.10. Proposition. Let V be a finite dimensional vector space over a field F. An operator T on
V is invertible if and only if T is not a zero divisor in L(V ).
Hint for proof . In one direction the proof is very easy. For the other, use proposition 4.6.6.
4.9.11. Corollary. Let V be a finite dimensional vector space over a field F, T L(V ), and F.
Then T I fails to be invertible if and only if is a root of the minimal polynomial of T .
4.9.12. Corollary. If T is an operator on a finite dimensional vector space, then its minimal
polynomial and characteristic polynomial have the same roots.
4.9.13. Theorem (Cayley-Hamilton Theorem). If T is an operator on a finite dimensional vector
space, then the characteristic polynomial of T annihilates T . Moreover, the minimal polynomial of
T is a factor of the characteristic polynomial of T .
Proof. See [13], proposition 3.19.
4.9.14. Definition. Let V be a vector space, T be an operator on V , and be an eigenvalue
of T . A nonzero vector x in the kernel of T I is an eigenvector of T associated with (or
corresponding to, or belonging to) the eigenvalue .
4.9.15. Definition. Let V be a vector space, T be an operator on V , and be an eigenvalue of T .
The eigenspace associated with (or corresponding to, or belonging to) the eigenvalue
is the kernel of T I.
3 1 1
4.9.16. Exercise. Let T be the operator on R3 whose matrix representation is 2 2 1.
2 2 0
(a) Find the characteristic polynomial of T .
(b) Find the minimal polynomial of T .
(c) Find the eigenspaces V1 and V2 of T .
1 0
0
1 1
0 1 2 3 3
5
.
0
0
1
2
2
4.9.17. Exercise. Let T be the operator on R whose matrix representation is
1 1 1
0
1
1 1 1 1 2
4.9.18. Proposition. If 1 6= 2 are eigenvalues of an operator T , then the eigenspaces M1 and
M2 associated with 1 and 2 , respectively, have only 0 in common.
4.9.19. Proposition. Let V be a vector space over a field F. If v is an eigenvector associated with
an eigenvalue of an operator T L(V ) and p is a polynomial in F[x], then p(T )v = p()v.
4.9.20. Definition. Two operators on a vector space (or two n n matrices) R and T are similar
if there exists an invertible operator (or matrix) S such that R = S 1 T S.
4.9.21. Proposition. If R and T are operators on a vector space and R is invertible, then RT is
similar to T R.
4.9.22. Example. If R and T are operators on a vector space, then RT need not be similar to T R.
4.9.23. Proposition. Let R and T be operators on a vector space. If R is similar to T and p F[x]
is a polynomial, then p(R) is similar to p(T ).
4.9.24. Proposition. If R and T are operators on a vector space, R is similar to T , and R is
invertible, then T is invertible and T 1 is similar to R1 .
4.10. THE SPECTRAL THEOREM - VECTOR SPACE VERSION
51
4.9.25. Proposition. If two matrices A and B are similar, then they have the same spectrum.
Hint for proof . You may use familiar facts about determinants that we have not yet proved.
4.9.26. Definition. An operator on a vector space is nilpotent if some power of the operator
is 0.
4.9.27. Proposition. An operator T on a finite dimensional complex vector space is nilpotent if
and only if (T ) = {0}.
4.9.28. Notation. Let 1 , . . . , n be elements of a field F. Then diag(1 , . . . , n ) denotes the
n n matrix whose entries are all zero except on the main diagonal where they are 1 , . . . , n .
Such a matrix is a diagonal matrix.
4.9.29. Definition. Let V be a vector space of finite dimension n. An operator T on V is
diagonalizable if it has n linearly independent eigenvectors (or, equivalently, if V has a basis of
eigenvectors of T ).
4.9.30. Proposition. Let A be an n n matrix with entries from a field F. Then A, regarded as
an operator on Fn , is diagonalizable if and only if it is similar to a diagonal matrix.
4.10. The Spectral Theorem - Vector Space Version
4.10.1. Proposition. Let E1 , . . . , En be the projections associated with a direct sum decomposition
V = M1 Mn of a vector space V and let T be an operator on V . Then each subspace Mk is
invariant under T if and only if T commutes with each projection Ek .
4.10.2. Theorem (Spectral Theorem for Vector Spaces). If T is a diagonalizable operator on a
finite dimensional vector space V , then
n
X
k Ek
T =
k=1
where 1 , . . . , n are the (distinct) eigenvalues of T and {E1 , . . . En } is the resolution of the identity
whose projections are associated with the corresponding eigenspaces M1 , . . . , Mn .
4.10.3. Proposition. Let T be an operator on a finite dimensional vector space V . If 1 , . . . , n
are distinct scalars and E1 , . . . , En are nonzero operators on V such that
P
(i) T = P nk=1 k Ek ,
(ii) I = nk=1 Ek , and
(iii) Ej Ek = 0 whenever j 6= k,
then T is diagonalizable, the scalars 1 , . . . , n are the eigenvalues of T , and the operators E1 ,
. . . , En are projections whose ranges are the eigenspaces of T .
4.10.4. Proposition. If T is a diagonalizable operator on a finite dimensional vector space V over
a field F and p F[x], then
n
X
p(T ) =
p(k )Ek
k=1
where 1 , . . . , n are the (distinct) eigenvalues of T and E1 , . . . En are the projections associated
with the corresponding eigenspaces M1 , . . . , Mn .
4.10.5. Proposition. If T is a diagonalizable operator on a finite dimensional vector L
space V ,
then the projections E1 , . . . , En associated with the decomposition of V as a direct sum
Mk of
its eigenspaces can be expressed as polynomials in T .
Hint for proof . Apply the Lagrange interpolation formula with the tk s being the eigenvalues
of T .
52
0 0 2
4.10.6. Exercise. Let T be the operator on R3 whose matrix representation is 0 2 0. Use
2 0 0
proposition 4.10.5 to write T as a linear combination of projections.
2 2 1
4.10.7. Exercise. Let T be the operator on R3 whose matrix representation is 1 1 1. Use
1 2 0
proposition 4.10.5 to write T as a linear combination of projections.
1
23 23
3
2
5
76
.
3
6
5
7
2
3 6
6
Write T as a linear combination of projections.
4.10.9. Proposition. An operator T on a finite
Qn dimensional vector space is diagonalizable if and
only if its minimal polynomial is of the form k=1 (x k ) for some distinct elements 1 , . . . , n
of the scalar field F.
Proof. See [16], page 204, theorem 6.
4.11. Two Decomposition Theorems
4.11.1. Theorem (Primary Decomposition Theorem). Let T L(V ) where V is a finite dimensional vector space. Factor the minimal polynomial
n
Y
mT =
p rk
k
k=1
r
into powers of distinct irreducible monic polynomials p1 , . . . , pn and let Wk = ker pk (T ) k for
each k. Then
L
(i) V = nk=1 Wk ,
(ii) each Wk is
invariant under T , and
(iii) if Tk = T W , then mT = p rk .
k

In the preceding theorem the spaces Wk are the generalized eigenspaces of the operator T .
4.11.2. Theorem (Diagonalizable Plus Nilpotent Decomposition). Let T be an operator on a
finite dimensional vector space V . Suppose that the minimal polynomial for T factors completely
into linear factors
mT (x) = (x 1 )d1 . . . (x r )dr
where 1 , . . . r are the (distinct) eigenvalues of T . For each k let Wk be the generalized eigenspace
ker(T k I)dk and let E1 , . . . , Er be the projections associated with the direct sum decomposition
V = W1 W2 Wr .
Then this family of projections is a resolution of the identity, each Wk is invariant under T , the
operator
D = 1 E1 + + r Er
is diagonalizable, the operator
N =T D
is nilpotent, and N commutes with D.
Furthermore, if D1 is diagonalizable, N1 is nilpotent, D1 + N1 = T , and D1 N1 = N1 D1 , then
D1 = D and N1 = N .
4.11. TWO DECOMPOSITION THEOREMS
53
4.11.3. Corollary. Every operator on a finite dimensional complex vector space can be written as
the sum of two commuting operators, one diagonalizable and the other nilpotent.

2 1
2
.
1 4
(a) Explain briefly why T is not diagonalizable.
(b) Find the diagonalizable

and nilpotent
parts
of T .
a b
c c
,b=
, and c =
.
Answer: D =
and N =
where a =
b a
c c
0
0 3
4.11.5. Exercise. Let T be the operator on R3 whose matrix representation is 2 1 2.
2 1 5
(a) Find D and N , the diagonalizable and nilpotent parts of T . Express these as polynomials
in T .
(b) Find a matrix S which diagonalizes D.
2 1 1
2 1 2
(c) Let [D1 ] = 1 2 1 and [N1 ] = 1 1 1. Show that D1 is diagonalizable,
1 1 2
3
0
3
that N1 is nilpotent, and that T = D1 + N1 . Why does this not contradict the uniqueness
claim made in theorem 4.11.2?
0
1 0 1
2 3 0 1

2 1 2 1.
2 1 0 3
p
.
(a) The characteristic polynomial of T is ( 2) where p =
r
(b) The minimal polynomial of T is ( 2) where r =
.
a b b b
b a b b
(c) The diagonalizable part of T is D =

and b =
.
b b a b where a =
b b b a
a b c b
a b c b
(d) The nilpotent part of T is N =

,b=
, and
a b c b where a =
c=
.
a b c b
1 0
0
1 1
0 1 2 3 3
5
.
0
0
1
2
2
1 1 1
0
1
1 1 1 1 2
Answer: cT () = ( + 1)p ( 1)q where p =
and q =

Answer: mT () = ( + 1)r ( 1)s where r =
and s =
(c) Find the eigenspaces V1 and V2 of T .

Answer: V1 = span{(a, 1, b, a, a)} where a =
V2 = span{(1, a, b, b, b), (b, b, b, 1, a)} where a =
and b =
and b =
; and
.
54
(d) Find the diagonalizable part of T .
a b b b b
b a c c c
where a =
b
b
a
c
c
Answer: D =
b b b a b
b b b b a
(e) Find the nilpotent part of
a a
a a
Answer: N =
a a
b b
b b
T.
a b b
a b b
a a
a
where a =
b b b
b b b
,b=
and b =
, and c =
(f) Find a matrix S which diagonalizes the diagonalizable part D of T . What is the diagonal
form of D associated
with this
matrix?
a b a a a
b a b a a
and b =
.
Answer: S =
b a a b a where a =
a a a b b
a a a a b
a 0 0 0 0
0 a 0 0 0
where a =
0
0
a
0
0
and
=
.
0 0 0 a 0
0 0 0 0 a
CHAPTER 5
THE SPECTRAL THEOREM FOR INNER PRODUCT SPACES

In this chapter all vector spaces (and algebras) have complex or real scalars.
5.1. Inner Products
5.1.1. Definition. Let V be a complex (or a real) vector space. A function which associates to
each pair of vectors x and y in V a complex number (or, in the case of a real vector space, a real
number) hx, yi is an inner product (or a dot product) on V provided that the following four
conditions are satisfied:
(a) If x, y, z V , then
hx + y, zi = hx, zi + hy, zi.
(b) If x, y V , then
hx, yi = hx, yi.
(c) If x, y V , then
hx, yi = hy, xi.
(d) For every nonzero x in V we have hx, xi > 0.
Conditions (a) and (b) show that an inner product is linear in its first variable. Conditions (a) and
(b) of proposition 5.1.2 say that an inner product is conjugate linear in its second variable.
When a mapping is linear in one variable and conjugate linear in the other, it is often called
sesquilinear (the prefix sesqui- means one and a half). Taken together conditions (a)(d)
say that the inner product is a positive definite conjugate symmetric sesquilinear form. Of course,
in the case of a real vector space, the complex conjugation indicated in (c) has no effect and the
inner product is a positive definite symmetric bilinear form. A vector space on which an inner
product has been defined is an
inner product space.
5.1.2. Proposition. If x, y, and z are vectors in an inner product space and C, then
(a) hx, y + zi = hx, yi + hx, zi,
(b) hx, yi = hx, yi, and
(c) hx, xi = 0 if and only if x = 0.
5.1.3. Example. For vectors x = (x1 , x2 , . . . , xn ) and y = (y1 , y2 , . . . , yn ) belonging to Cn define
hx, yi =
n
X
xk yk .
k=1
Then
Cn
is an inner product space.
5.1.4. Example. For a < b let C([a, b], C) be the family of all continuous complex valued functions
on the interval [a, b]. For every f , g C([a, b], C) define
Z b
hf, gi =
f (x)g(x) dx.
a
Then C([a, b], C) is a complex inner product space. In a similar fashion, C([a, b]) = C([a, b], R) is
made into a real inner product space.
55
56
5. THE SPECTRAL THEOREM FOR INNER PRODUCT SPACES
5.1.5. Definition. When a (real or complex) vector space has been equipped with an inner product
we define the norm of a vector x by
p
kxk := hx, xi;
(This somewhat optimistic terminology is justified in proposition 5.1.13 below.)
5.1.6. Theorem. In every inner product space the Schwarz inequality
|hx, yi| kxk kyk.
holds for all vectors x and y.
Hint for proof . Let V be an inner product space and fix vectors x, y V . For every scalar
we know that
0 hx y, x yi.
(5.1)
Expand the right hand side of (5.1) into four terms and write hy, xi in polar form: hy, xi = rei ,
where r > 0 and R. Then in the resulting inequality consider those of the form tei where
t R. Notice that now the right side of (5.1) is a quadratic polynomial in t. What can you say
about its discriminant?
5.1.7. Exercise. If a1 , . . . , an > 0, then
X
n
j=1
aj
X
n
k=1
1
ak
n2 .
The proof of this is obvious from the Schwarz inequality if we choose x and y to be what?
5.1.8. Exercise. In this exercise notice that part (a) is a special case of part (b).
2
(a) Show that if a, b, c > 0, then 12 a + 13 b + 61 c 12 a2 + 13 b2 + 16 c2 .
P
(b) Show that if a1 , . . . , an , w1 , . . . , wn > 0 and nk=1 wk = 1, then
2 X
X
n
n
ak 2 wk .
ak wk
k=1
5.1.9. Exercise. Show that if
k=1 ak
k=1
converges, then
k=1 k
1 a
k
converges absolutely.
5.1.10.
Example. A sequence (ak ) of (real or) complex numbers is said to be square summable
P
2
if
|a
k=1 k | < . The vector space of all square summable sequences of real numbers (respectively,
complex numbers) is denoted by l2 (R) (respectively, l2 (C)). When no confusion will result, both
are denoted by l2 . If a, b l2 , define
X
ha, bi =
ak bk .
k=1
(It must be shown that this definition makes sense and that it makes l2 into an inner product
space.)
5.1.11. Definition. Let V be a complex (or real) vector space. A function k k : V R : x 7 kxk
is a norm on V if
(i) kx + yk kxk + kyk
for all x, y V ;
(ii) kxk = || kxk
for all x V and C (or R); and
(iii) if kxk = 0, then x = 0.
The expression kxk may be read as the norm of x or the length of x.
A vector space on which a norm has been defined is a normed linear space (or normed
vector space). A vector in a normed linear space which has norm 1 is a unit vector.
5.1.12. Proposition. If k k is norm on a vector space V , then kxk 0 for every x V and
k0k = 0.
5.2. ORTHOGONALITY
57
As promised in definition 5.1.5 we can verify the (somewhat obvious) fact that every inner
product space is a normed linear space (and therefore a topologicalin fact, a metricspace).
5.1.13. Proposition. Let V be an inner product space. The map x 7 kxk defined on V in 5.1.5
is a norm on V .
5.1.14. Proposition (The parallelogram law). If x and y are vectors in an inner product space,
then
kx + yk2 + kx yk2 = 2kxk2 + 2kyk2 .
5.1.15. Example. Consider the space C([0, 1]) of continuous complex valued functions defined on
[0, 1]. Under the uniform norm kf ku := sup{|f (x)| : 0 x 1} the vector space C([0, 1]) is a
normed linear space, but there is no inner product on C([0, 1]) which induces this norm.
Hint for proof . Use the preceding proposition.
5.1.16. Proposition (The polarization identity). If x and y are vectors in a complex inner product
space, then
hx, yi = 41 (kx + yk2 kx yk2 + i kx + iyk2 i kx iyk2 ) .
5.1.17. Exercise. What is the corresponding formula for the polarization identity in a real inner
product space?
5.2. Orthogonality
5.2.1. Definition. Vectors x and y in an inner product space H are orthogonal (or perpendicular) if hx, yi = 0. In this case we write x y. Subsets A and B of H are orthogonal if
a b for every a A and b B. In this case we write A B.
5.2.2. Proposition. Let a be a vector in an inner product space H. Then a x for every x H
if and only if a = 0.
5.2.3. Proposition (The Pythagorean theorem). If x y in an inner product space, then
kx + yk2 = kxk2 + kyk2 .
5.2.4. Definition. If M and N are subspaces of an inner product space H we use the notation
H = M N to indicate not only that H is the sum of M and N but also that M and N are
orthogonal. Thus we say that H is the (internal) orthogonal direct sum of M and N .
5.2.5. Proposition. If M and N are subspaces of an inner product space H and H is the orthogonal
direct sum of M and N , then it is also the vector space direct sum of M and N .
As is the case with vector spaces in general, we make a distinction between internal and external
direct sums.
5.2.6. Definition. Let V and W be inner product spaces. For (v, w) and (v 0 , w0 ) in V W and
C define
(v, w) + (v 0 , w0 ) = (v + v 0 , w + w0 )
and
(v, w) = (v, w) .
This results in a vector space, which is the (external) direct sum of V and W . To make it into an
inner product space define
h(v, w), (v 0 , w0 )i = hv, v 0 i + hw, w0 i.
This makes the direct sum of V and W into an inner product space. It is the (external orthogonal) direct sum of V and W and is denoted by V W .
58
5.2.7. CAUTION. Notice that the same notation is used for both internal and external direct
sums and for both vector space direct sums (see definitions 1.5.14 and 3.4.3) and orthogonal direct
sums. So when we see the symbol V W it is important to be alert to context, to know which
category we are in: vector spaces or inner product spaces, especially as it is common practice to
omit the word orthogonal as a modifier to direct sum even in cases when it is intended.
5.2.8. Example. In R2 let M be the x-axis and L be the line whose equation is y = x. If we
think of R2 as a (real) vector space, then it is correct to write R2 = M L. If, on the other
hand, we regard R2 as a (real) inner product space, then R2 6= M L (because M and L are not
perpendicular).
5.2.9. Notation. Let V be an inner product space, x V , and A, B V . If x a for every
a A, we write x A; and if a b for every a A and b B, we write A B. We define A ,

the orthogonal complement of A, to be {x V : x A}. We write A for A .
5.2.10. CAUTION. The superscript is here used quite differently than in our study of vector
spaces (see 2.6.1). These two uses are, however, by no means unrelated! It is an instructive exercise
to make explicit exactly what this relationship is.
5.2.11. Proposition. If A is a subset of an inner product space V , then A is a subspace of V
and A = (span A) . Furthermore, if A B V , then B A .
5.2.12. Definition. When a nonzero vector x in an inner product space V is divided by its norm
x
is clearly a unit vector. We say that u results from normalizing the
the resulting vector u =
kxk
vector x. A subset E of V is orthonormal if every pair of distinct vectors in E are orthogonal and
every vector in E has length one. If, in addition, V is the span of E, then E is an orthonormal
basis for V .
5.2.13. Definition. Let V be an inner product space and E = {e1 , e2 , . . . , en } be a finite orthonork
mal subset of V . For each k Nk let xk :=
Pnhx, e i.k This scalar is called the Fourier coefficient
of x with respect to E. The vector s := k=1 xk e is the Fourier sum of x with respect to E.
5.2.14. Proposition. Let notation be as in 5.2.13. Then s = x if and only if x span E.
5.2.15. Proposition. Let notation be as in 5.2.13. Then x s ek for k = i, . . . , n and therefore
x s s.
The next result gives us a recipe for converting a finite linearly independent subset of an inner
product space into an orthonormal basis for the span of that set.
5.2.16. Theorem (Gram-Schmidt Orthonormalization). Let A = {a1 , a2 , . . . , an } be a finite linearly independent subset of an inner product space V . Define vectors e1 , . . . , en recursively by
setting
e1 := ka1 k
1 1
and for 2 m n
em := kam sm k1 (am sm )
P
m k k
m with respect to E
1
m1 }.
where sm := m1
m1 := {e , . . . , e
k=1 ha , e ie is the Fourier sum for a
Then En is an orthonormal basis for the span of A.
It should be clear from the proof of the preceding theorem that finiteness plays no essential
role. The theorem remains true for countable linearly independent sets (as does its proof).
5.2.17. Corollary. Every finite dimensional inner product space has an orthonormal basis.
5.2. ORTHOGONALITY
59
5.2.18. Example. Let R[x] be the inner product space of real polynomials whose inner product
is defined by
Z 1
pe(x)e
q (x) dx
hp, qi :=
1
for all p, q R[x]. Application of the Gram-Schmidt process to the set {1, x, x2 , x3 , . . . } of real
polynomials produces an orthonormal sequence of polynomials known as the Legendre polynomials.
Compute the first four of these.
5.2.19. Example. Let R[x] be the inner product space of real polynomials whose inner product
is defined by
Z
hp, qi :=
pe(x)e
q (x)ex dx
0
for all p, q R[x]. Application of the Gram-Schmidt process to the set {1, x, x2 , x3 , . . . } of real
polynomials produces an orthonormal sequence of polynomials known as the Laguerre polynomials.
Compute the first four of these.
Hint forR proof . Integration by parts or familiarity with the gamma function allows us to con
clude that 0 xn ex dx = n! for each n N.
5.2.20. Proposition. If M is a subspace of a finite dimensional inner product space V then
V = M M .
5.2.21. Example. The subspace lc (N, R) of the inner product space l2 (R) (see example 5.1.10)
shows that the preceding proposition does not hold for infinite dimensional spaces.
5.2.22. Proposition. Let M be a subspace of an inner product space V . Then
(a) M M ;
(b) equality need not hold in (a); but
(c) if V is finite dimensional, then M = M .
5.2.23. Proposition. If S is a set of mutually perpendicular vectors in an inner product space
and 0
/ S, then the set S is linearly independent.
5.2.24. Proposition. Let M and N be subspaces of an inner product space V . Then
(a) (M + N ) = (M N ) = M N and
(b) if V is finite dimensional, then (M N ) = M + N .
5.2.25. Proposition. Let S, T : H K be linear maps between inner product spaces H and K.
If hSx, yi = hT x, yi for every x H and y K, then S = T .
5.2.26. Proposition. If H is a complex inner product space and T L(V ) satisfies hT z, zi = 0
for all z H, then T = 0.
Hint for proof . In the hypothesis replace z first by x + y and then by x + iy.
5.2.27. Example. Give an example to show that the preceding result does not hold for real inner
product spaces.
5.2.28. Example. Let H be a complex inner product space and a H. Define a : H C by
a (x) = hx, ai for all x H. Then a is a linear functional on H.
5.2.29. Theorem (Riesz-Frechet Theorem). If f is a linear functional on a finite dimensional
inner product space H, then there exists a unique vector a H such that
f (x) = hx, ai
for every x H.
60
P
P
k
5.2.30. Example. Consider the function : lc (N) C : x 7
k=1 k e , the
k=1 k where x =
ek s being the standard basis vectors for lc (N). This linear functional provides an example which
shows that the Riesz-Frechet theorem does not hold (as stated) in infinite dimensional spaces.
Here is another example of the failure of this result in infinite dimensional spaces.
5.2.31. Example.
On the vector space H of polynomials over C define an inner product by
R1
hp, qi = 0 p(t)q(t) dt. For a fixed z C define the functional Ez , called evaluation at z, by
Ez (p) = p(z) for every p H. Then Ez belongs to H but that there is no polynomial p such that
Ez (q) = hq, pi for every q H.
5.2.32. CAUTION. It is important not to misinterpret the two preceding examples. There is
indeed a (very important!) version of the Riesz-Frechet theorem which does in fact hold for infinite
dimensional spaces. If we restrict our attention to continuous linear functionals on complete inner
product spaces, then the conclusion of theorem 5.2.29 does indeed hold for infinite dimensional
spaces.
5.2.33. Exercise. Use vector methods (as described in 1.4.15no coordinates, no major results
from Euclidean geometry) to show that the midpoint of the hypotenuse of a right triangle is
equidistant from the vertices. Hint. Let 4ABC be a right triangle and O be the midpoint of the

hypotenuse AB. What can you say about h AO + OC, CO + OB i?
5.2.34. Exercise. Use vector methods (as described in 1.4.15) to show that an angle inscribed in
a semicircle is a right angle.
5.2.35. Exercise. Use vector methods (as described in 1.4.15) to show that if a parallelogram has
perpendicular diagonals, then it is a rhombus (that is, all four sides have equal length). Hint. Let
ABCD be a parallelogram. Express the inner product of the diagonals AC and DB in terms of
the lengths of the sides AB and BC.

5.2.36. Exercise. Use vector methods (as described in 1.4.15) to show that the diagonals of a
rhombus are perpendicular.
5.3. Involutions and Adjoints
5.3.1. Definition. An involution on a complex (or real) algebra A is a map x 7 x from A into
A which satisfies
(i) (x + y) = x + y ,
(ii) (x) = x ,
(iii) x = x, and
(iv) (xy) = y x
for all x, y A and C (or R). An algebra on which an involution has been defined is a algebra (pronounced star algebra). An algebra homomorphism between -algebras which
preserves involution (that is, such that (a ) = ((a)) ) is a -homomorphism (pronounced star
homomorphism. A -homomorphism : A B between unital algebras is said to be unital if
(1A ) = 1B .
5.3.2. Proposition. If a and b are elements of a -algebra, then a commutes with b if and only if
a commutes with b .
5.3.3. Proposition. In a unital -algebra 1 = 1.
5.3.4. Proposition. If a -algebra A has a left multiplicative identity e, then A is unital and
e = 1A .
5.3.5. Proposition. An element a of a unital -algebra is invertible if and only if a is. And
when a is invertible we have

(a )1 = a1 .
5.3. INVOLUTIONS AND ADJOINTS
61
5.3.6. Proposition. Let a be an element of a unital -algebra. Then (a) if and only if
(a ).
5.3.7. Definition. An element a of a complex -algebra A is normal if a a = aa . It is selfadjoint (or Hermitian) if a = a. It is skew-Hermitian if a = a. And it is unitary if
a a = aa = 1. The set of all self-adjoint elements of A is denoted by H(A), the set of all normal
elements by N (A), and the set of all unitary elements by U(A).
Oddly, and perhaps somewhat confusingly, history has dictated an alternative, but parallel,
language for real algebrasespecially algebras of matrices and linear maps. An element a of a real
-algebra A is symmetric if a = a. It is skew-symmetric if a = a. And it is orthogonal
if a a = aa = 1.
5.3.8. Example. Complex conjugation is an involution on the algebra C of complex numbers.
5.3.9. Example. Transposition (see definition 1.7.31) is an involution on the real algebra Mn of
n n matrices.
5.3.10. Example. Let a < b. The map f
7 f of a function to its complex conjugate is an
involution on the complex algebra C([a, b], C) of continuous complex valued functions on [a, b]
5.3.11. Proposition. For every element a of a -algebra A there exist unique self-adjoint elements
u and v in A such that a = u + iv.
Hint for proof . The self-adjoint element u is called the real part of a and v the imaginary part
of a.
5.3.12. Corollary. An element of a -algebra is normal if and only if its real part and its imaginary
part commute.
5.3.13. Definition. Let H and K be complex inner product spaces and T : H K be a linear
map. If there exists a function T : K H which satisfies
hT x, yi = hx, T yi
for all x H and y K, then T is the adjoint of T . If a linear map T has an adjoint we say
that T is adjointable. Denote the set of all adjointable maps from H to K by A(H, K) and write
A(H) for A(H, H).
When H and K are real vector spaces, the adjoint of T is usually called the transpose of T
and the notation T t is used (rather than T ).
5.3.14. Proposition. Let T : H K be a linear map between complex inner product spaces. If
the adjoint of T exists, then it is unique. (That is, there is at most one function T : K H that
satisfies hT x, yi = hx, T yi for all x H and y K.)
Similarly, of course, if T : H K is a linear map between real inner product spaces and if the
transpose of T exists, then it is unique.
5.3.15. Example. Let C = C([0, 1]) be the inner product space defined in example 5.1.4 and
J0 = {f C : f (0) = 0}. Then the inclusion map : J0 C is an example of a map which is not
adjointable.
5.3.16. Example. Let U be the unilateral shift operator on l2 (see example 5.1.10)
U : l2 l2 : (x1 , x2 , x3 , . . . ) 7 (0, x1 , x2 , . . . ),
then its adjoint is given by
U : l2 l2 : (x1 , x2 , x3 , . . . ) 7 (x2 , x3 , x4 , . . . ).
62
5.3.17. Example (Multiplication operators). Let be a fixed continuous complex valued function
on the interval [a, b]. On the inner product space C = C([a, b], C) (see example 5.1.4) define
M : C C : f 7 f .
Then M is an adjointable operator on C.
the adjoint of T exists, then it is linear.
And, similarly, of course, if the transpose of a linear map between real inner product spaces
exists, then it is linear. In the sequel we will forgo the dubious helpfulness of mentioning every
obvious real analog of results holding for complex inner product spaces.
the adjoint of T exists, then so does the adjoint of T and T = T .
5.3.20. Proposition. Let S : H K and T : K L be linear maps between complex inner product
space. Show that if S and T both have adjoints, then so does their composite T S and
(T S) = S T .
5.3.21. Proposition. If T : H K is an invertible linear map between complex inner product
spaces and both T and T 1 have adjoints, then T is invertible and (T )1 = (T 1 ) .
5.3.22. Proposition. Let S and T be operators on a complex inner product space H. Then
(S + T ) = S + T and (T ) = T for every C.
5.3.23. Example. If H is a complex inner product space, then A(H) is a unital complex -algebra.
It is a unital subalgebra of L(H).
5.3.24. Theorem. Let T be an adjointable operator on an inner product space H. Then
(a) ker T = (ran T ) and
(b) ran T (ker T ) . If H is finite dimensional, then equality holds in (b).
5.3.25. Theorem. Let T be an adjointable operator on an inner product space H. Then
(a) ker T = (ran T ) and
(b) ran T (ker T ) . If H is finite dimensional, then equality holds in (b).
5.3.26. Proposition. Every linear map between finite dimensional complex inner product spaces
is adjointable.
Hint for proof . Use the Riesz-Frechet theorem.
5.3.27. Corollary. If H is a finite dimensional inner product space then L(H) is a unital -algebra.
5.3.28. Proposition. An adjointable operator T on a complex inner product space H is unitary
if and only if it preserves inner products (that is, if and only if hT x, T yi = hx, yi for all x, y H).
Similarly, an adjointable operator on a real inner product space is orthogonal if and only if it
preserves inner products.
5.3.29. Exercise. Let T : H K be a linear map between finite dimensional complex inner
product spaces. Find the matrix representation of T in terms of the matrix representation of T .
Also, for a linear map T : H K between finite dimensional real inner product spaces find the
matrix representation of T t in terms of the matrix representation of T .
5.3.30. Proposition. Every eigenvalue of a Hermitian operator on a complex inner product space
is real.
Hint for proof . Let x be an eigenvector associated with an eigenvalue of an operator A.
Consider kxk2 .
5.4. ORTHOGONAL PROJECTIONS
63
5.3.31. Proposition. Let A be a Hermitian operator on a complex inner product space. Prove
that eigenvectors associated with distinct eigenvalues of A are orthogonal.
Hint for proof . Let x and y be eigenvectors associated with distinct eigenvalues and of A.
Start your proof by showing that hx, yi = hx, yi.
5.3.32. Proposition. Let N be a normal operator on a complex inner product space H. Then
kN xk = kN xk for every x H.
5.4. Orthogonal Projections
5.4.1. Proposition. Let H be an inner product space and M and N be subspaces of H such that
H = M + N and M N = {0}. (That is, suppose that H is the vector space direct sum of M
and N .) Also let P = EN M be the projection of H along N onto M . Prove that P is self-adjoint
(P exists and P = P ) if and only if M N .
5.4.2. Definition. A projection in a -algebra A is an element p of the algebra which is idempotent (p2 = p) and self-adjoint (p = p). The set of all projections in A is denoted by P(A).
An operator P on a complex inner product space H is an orthogonal projection if it is selfadjoint and idempotent; that is, if it is a projection in the -algebra A(H) of adjointable operators
on H. (On a real inner product space, of course, the appropriate adjectives are symmetric and
idempotent.)
Notice that a vector space projection EM N is, in general, not a projection in any -algebra.
5.4.3. Convention. It is standard practice to refer to orthogonal projections on inner product
spaces simply as projections. This clearly invites confusion with the vector space notion of
projection. So one must be careful: just as the symbols and have different meanings depending
on context (vector spaces or inner product spaces), so does the word projection. In these notes
and elsewhere, when the context is inner product spaces the word projection frequently means
orthogonal projection
5.4.4. Proposition. If P is an orthogonal projection on an inner product space, then the space is
the orthogonal direct sum of the range of P and the kernel of P .
5.4.5. Proposition. Let p and q be projections in a -algebra. Then the following are equivalent:
(a) pq = 0;
(b) qp = 0;
(c) qp = pq;
(d) p + q is a projection.
5.4.6. Definition. Let p and q be projections in a -algebra. If any of the conditions in the
preceding result holds, then we say that p and q are orthogonal and write p q. (Thus for
operators on an inner product space we can correctly speak of orthogonal orthogonal projections!)
5.4.7. Proposition. Let P and Q be projections on an inner product space H. Then P Q if and
only if ran P ran Q. In this case P + Q is an orthogonal projection whose kernel is ker P ker Q
and whose range is ran P + ran Q.
5.4.8. Example. On an inner product space projections need not commute. For example, let P
be the projection of the (real) inner product space R2 onto the line y = x and Q be the projection
of R2 onto the x-axis. Then P Q 6= QP .
5.4.9. Proposition. Let p and q be projections in a -algebra. Then pq is a projection if and only
if pq = qp.
5.4.10. Proposition. Let P and Q be projections on an inner product space H. If P Q = QP ,
then P Q is a projection whose kernel is ker P + ker Q and whose range is ran P ran Q.
64
5.4.11. Proposition. Let p and q be projections in a -algebra. Then the following are equivalent:
(a) pq = p;
(b) qp = p;
(c) q p is a projection.
5.4.12. Definition. Let p and q be projections in a -algebra. If any of the conditions in the
preceding result holds, then we write p q. In this case we say that p is a subprojection of q or
that p is smaller than q.
5.4.13. Proposition. If A is a -algebra, then the relation defined in 5.4.12 is a partial ordering
on P(A). If A is unital, then 0 p 1 for every p P(A).
5.4.14. Notation. If H, M , and N are subspaces of an inner product space, then the assertion
H = M N , may be rewritten as M = H N (or N = H M ).
5.4.15. Proposition. Let P and Q be projections on an inner product space H. Then the following
are equivalent:
(a) P Q;
(b) kP xk kQxk for all x H; and
(c) ran P ran Q.
In this case Q P is a projection whose kernel is ran P + ker Q and whose range is ran Q ran P .
The next two results are optional: they will not be used in the sequel.
5.4.16. Proposition. Suppose p and q are projections on a -algebra A. If pq = qp, then the
infimum of p and q, which we denote by p f q, exists with respect to the partial ordering and
p f q = pq. The infimum p f q may exist even when p and q do not commute. A necessary and
sufficient condition that p q hold is that both p f q = 0 and pq = qp hold.
5.4.17. Proposition. Suppose p and q are projections on a -algebra A. If p q, then the
supremum of p and q, which we denote by p g q, exists with respect to the partial ordering and
p g q = p + q. The supremum p g q may exist even when p and q are not orthogonal.
5.5. The Spectral Theorem for Inner Product Spaces
5.5.1. Definition. Two elements a and b of a -algebra A are unitarily equivalent if there
exists a unitary element u of A such that b = u au.
5.5.2. Definition. An operator T on a complex inner product space V is unitarily diagonalizable if there exists an orthonormal basis for V consisting of eigenvectors of T .
5.5.3. Proposition. Let A be an n n matrix A of complex numbers. Then A, regarded as an
operator on Cn , is unitarily diagonalizable if and only if it is unitarily equivalent to a diagonal
matrix.
5.5.4. Definition. Let M1 Mn be an orthogonal direct sum decomposition of an inner
product space H. For each k let Pk be the orthogonal projection onto Mk . The projections P1 ,
. . . Pn are the orthogonal projections associated with the orthogonal direct sum
decomposition H = M1 Mn . The family {P1 , . . . , Pn } is an orthogonal resolution
of the identity. (Compare this with definitions 4.9.1 and 4.9.2.)
5.5.5. Theorem (Spectral Theorem: Complex Inner Product Space Version). An operator N
on a finite dimensional complex inner product space V is normal if and only if it is unitarily
diagonalizable in which case it can be written as
n
X
N=
k Pk
k=1
5.5. THE SPECTRAL THEOREM FOR INNER PRODUCT SPACES
65
where 1 , . . . , n are the (distinct) eigenvalues of N and {P1 , . . . , Pn } is the orthogonal resolution
of the identity whose orthogonal projections are associated with the corresponding eigenspaces M1 ,
. . . , Mn .
Proof. See [29], theorems 10.13 and 10.21; or [16], chapter 8, theorems 20 and 22, and
chapter 9. theorem 9.
5.5.6. Exercise. Let N be the operator on C2 whose matrix representation is

0 1
.
1 0
(a) The eigenspace M1 associated with the eigenvalue i is the span of ( 1 ,
).
).
(b) The eigenspace M2 associated with the eigenvalue i is the span of ( 1 ,
(c) The (matrix representations of the) orthogonal

projections P1 andP2 onto the eigenspaces
a b
a b
M1 and M2 , respectively, are P1 =
; and P2 =
where a =
and
b a
b a
b=
.
(d) Write N as a linear combination of the projections found in (c).
Answer: [N ] =
P1 +
P2 .

a a
and b =
.
(e) A unitary matrix U which diagonalizes [N ] is
where
a =
b b
The associated diagonal form = U [N ] U of [N ] is

2
1+i
5.5.7. Exercise. Let H be the self-adjoint matrix
.
1i
3
(a) Use the spectral theorem to write H as a linear combination of orthogonal projections.

1
2
1 i
, =
, P1 =
,
Answer: H = P1 + P2 where =
3

1
1
1+i
and P2 =
.
3
(b) Find a square root of H.

1
4
1+i
.
Answer: H =
3
4 + 2i 1 i 1 i
1
1 i 4 + 2i 1 i .
5.5.8. Exercise. Let N =
3 1 i 1 i 4 + 2i
a b b
(a) The matrix N is normal because N N = N N = b a b where a =
and
b b a
b=
.
(b) According to the spectral theorem N can be written as a linear combination of orthogonal
projections. Written in this
,
2 where 1 =
form N= 1 P1 + 2 P
a a a
b a a
2 =
, P1 = a a a, and P2 = a b a where a =
and
a a a
a a b
b=
.
66
a b c
(c) A unitary matrix U which diagonalizes N is a b c where a =
a d 2c
c=
, and d =
.
The associated diagonal form = U N U of N is
,b=
CHAPTER 6
A BRIEF REVIEW OF DIFFERENTIAL CALCULUS

We now pause for a very brief review of differential calculus. The central concept here is
differentiability. A function f between normed linear spaces is said to be differentiable at a point p
if (when the point (p, f (p)) is translated to the origin) the function is tangent to some continuous
linear map. In this chapter (much of which is just chapter 13 of my online text [12]) we make this
idea precise and record a few important facts about differentiability. A more detailed nd leisurely
treatment can be found in my ProblemText in Advanced Calculus [11], chapters 2529.
There are two sorts of textbooks on differential calculus: concept oriented and computation
oriented. It is my belief that students who understand the concepts behind differentiation can
do the calculations, while students who study calculations only often get stuck. Among the most
masterful presentations of concept oriented differential calculus are [8] (volume I, chapter 8) and
[22] (chapter 3). As of this writing the latter book is available without charge at the website of
one of the authors:
http://www.math.harvard.edu/~shlomo/docs/Advanced_Calculus.pdf
The material in this chapter will benefit primarily those whose only encounter with multivariate
calculus has been through partial derivatives and a chain rule that looks something like
w x w y
w z
w
=
+
+
u
x u
y u
z u
(6.1)
The approach here is intended to be more geometric, emphasizing the role of tangency.
6.1. Tangency
6.1.1. Notation. Let V and W be normed linear spaces and a V . (If you are unfamiliar with,
or uncomfortable working in, normed linear spaces, just pretend that all the spaces involved are
n-dimensional Euclidean spaces. The only thing you may lose by so doing is the pleasant feeling of
assurance that differential calculus is no harder in infinite dimensional spaces than on the real line.)
We denote by Fa (V, W ) the family of all functions defined on a neighborhood of a taking values
in W . That is, f belongs to Fa (V, W ) if there exists an open set U such that a U dom f V
and if the image of f is contained in W . We shorten Fa (V, W ) to Fa when no confusion will result.
Notice that for each a V , the set Fa is closed under addition and scalar multiplication. (As
usual, we define the sum of two functions f and g in Fa to be the function f + g whose value at
x is f (x) + g(x) whenever x belongs to dom f dom g.) Despite the closure of Fa under these
operations, Fa is not a vector space. (Why not?)
6.1.2. Definition. Let V and W be normed linear spaces. A function f in F0 (V, W ) belongs to
O(V, W ) if there exist numbers c > 0 and > 0 such that
kf (x)k c kxk
whenever kxk < .
A function f in F0 (V, W ) belongs to o(V, W ) if for every c > 0 there exists > 0 such that
kf (x)k c kxk
67
68
6. A BRIEF REVIEW OF DIFFERENTIAL CALCULUS
whenever kxk < . Notice that f belongs to o(V, W ) if and only if f (0) = 0 and
kf (h)k
= 0.
h0 khk
lim
When no confusion seems likely we will shorten O(V, W ) to O and o(V, W ) to o.

6.1.3. Exercise. Here is a list summarizing the important facts about the families O and o. State
precisely what each of these says and give a proof. (Here B is the set of continuous linear maps
between normed linear spaces V and W , and C0 is the set of all functions in F0 (V, W ) which are
continuous at 0.)
(1)
B o O C0 .
(2)
B o = 0.
(3)
O + O O;
(4)
o + o o;
(5)
o O o.
(6)
O o o.
(7)
o(V, R) W o(V, W ) .
(8)
O(V, R) O(V, W ) o(V, W ) .
O O.
o o.
6.1.4. Definition. Let V and W be normed linear spaces. Two functions f and g in F0 (V, W )
are tangent (at zero), in which case we write f ' g, if f g o(V, W ).
6.1.5. Proposition. The relation of tangency enjoys the following properties.
(a) Tangency at zero is an equivalence relation on F0 .
(b) Let S, T B and f F0 . If S ' f and T ' f , then S = T .
(c) If f ' g and j ' k, then f + j ' g + k, and furthermore, f ' g for all R.
(d) Let , F0 (V, R) and w W . If ' , then w ' w.
(e) Let f , g F0 (V, W ) and T B(W, X). If f ' g, then T f ' T g.
(f) Let h O(V, W ) and f , g F0 (W, X). If f ' g, then f h ' g h.
6.2. The Differential
6.2.1. Definition. Let V and W be normed linear spaces, a V , and f Fa (V, W ). Define the
function fa by
fa (h) := f (a + h) f (a)
for all h such that a + h is in the domain of f . Notice that since f is defined in a neighborhood
of a, the function fa is defined in a neighborhood of 0; that is, fa belongs to F0 (V, W ). Notice
also that fa (0) = 0.
6.2.2. Proposition. If V and W are normed linear spaces and a V , then the function has
the following properties.
(a) If f Fa (V, W ) and R, then
(f )a = fa .
(b) If f , g Fa (V, W ), then
(f + g)a = fa + ga .
(c) If Fa (V, R) and f Fa (V, W ), then
(f )a = (a) fa + a f (a) + a fa .
(d) If f Fa (V, W ), g Ff (a) (W, X), and g f Fa (V, X), then
(g f )a = gf (a) fa .
6.2. THE DIFFERENTIAL
69
(e) A function f : V W is continuous at the point a in V if and only if fa is continuous

at 0.
(f) If f : U U1 is a bijection between subsets of arbitrary vector spaces, then for each a in
U the function fa : U a U1 f (a) is invertible and
1

fa
= f 1 f (a) .
6.2.3. Definition. Let V and W be normed linear spaces, a V , and f Fa (V, W ). We say that
f is differentiable at a if there exists a continuous linear map which is tangent at 0 to fa . If
such a map exists, it is called the differential of f at a and is denoted by dfa . Thus dfa is just a
member of B(V, W ) which satisfies dfa ' fa . We denote by Da (V, W ) the family of all functions
in Fa (V, W ) which are differentiable at a. We often shorten this to Da .
We establish next that there can be at most one bounded linear map tangent to fa .
6.2.4. Proposition. Let V and W be normed linear spaces and a V . If f Da (V, W ), then its
differential is unique.
6.2.5. Exercise. Let
f : R3 R2 : (x, y, z) 7 (x2 y 7, 3xz + 4y)
and a = (1, 1, 0). Use the definition of differential to find dfa . Hint. Work with the matrix
3
2
representation of dfa. Since the
differential must belong to B(R , R ), its matrix representation is
r s t
a 2 3 matrix M =
. Use the requirement that khk1 kfa (h) M hk 0 as h 0 to
u v w
discover the identity of the entries in M .
6.2.6. Exercise. Let F : R2 R4 be defined by F(x, y) = (y, x2 , 4 xy, 7x), and let p = (1, 1).
Use the definition of differentiable to show that F is differentiable at p. Find the (matrix
representation of the) differential of F at p.
6.2.7. Proposition. Let V and W be normed linear spaces and a V . If f Da , then fa O;
thus, every function which is differentiable at a point is continuous there.
6.2.8. Proposition. Let V and W be normed linear spaces and a V . Suppose that f , g
Da (V, W ) and that R. Then
(1) f is differentiable at a and
d(f )a = dfa ;
(2) also, f + g is differentiable at a and
d(f + g)a = dfa + dga .
Suppose further that Da (V, R). Then
(c) f Da (V, W ) and
d(f )a = da f (a) + (a) dfa .
It seems to me that the version of the chain rule given in (6.1), although (under appropriate
hypotheses) a correct equation, really says very little. The idea that should be conveyed is that
the best linear approximation to the composite of two smooth functions is the composite of their
best linear approximations.
6.2.9. Theorem (The Chain Rule). Let V , W , and X be normed linear spaces with a V . If
f Da (V, W ) and g Df (a) (W, X), then g f Da (V, X) and
d(g f )a = dgf (a) dfa .
70
6. A BRIEF REVIEW OF DIFFERENTIAL CALCULUS
Proof. Our hypotheses are fa ' dfa and gf (a) ' dgf (a) . By proposition 6.2.7 fa O.
Then by proposition 6.1.5(f)
gf (a) fa ' dgf (a) fa
(6.2)
and by proposition 6.1.5(e)
dgf (a) fa ' dgf (a) dfa .
(6.3)
(g f )a ' gf (a) fa .
(6.4)
According to proposition 6.2.2(d)
From (6.2), (6.3),(6.4), and proposition 6.1.5(a) it is clear that

(g f )a ' dgf (a) dfa .
Since dgf (a) dfa is a bounded linear transformation, the desired conclusion is an immediate consequence of proposition 6.2.4.

6.2.10. Exercise. Derive (under appropriate hypotheses) equation (6.1) from theorem 6.2.9.
6.2.11. Exercise. Let T be a linear map from Rn to Rm and p Rn . Find dTp .
6.2.12. Example. Let T be a symmetric nn matrix and let p Rn . Define a function f : Rn R
by f (x) = hT x, xi. Then
dfp (h) = 2hT p, hi
for every h Rn .
6.3. The Gradient of a Scalar Field in Rn
6.3.1. Definition. A scalar field on Rn is a scalar valued function on a subset of Rn .
6.3.2. Definition. Let U be an open subset of Rn and : U R be a
differentiable at a point a in U , then its differential da is a (continuous)
into R. That is, da (Rn ) . Thus according to the Riesz-Frechet theorem
unique vector, which we denote by (a), representing the linear functional
is the unique vector in Rn such that
scalar field. If is
linear map from Rn
5.2.29 there exists a
da . That is, (a)
da (x) = hx, (a)i

for all x in Rn . The vector (a) is the gradient of at a. If U is an open subset of Rn and
is differentiable at each point of U , then the function
: U Rn : u 7 (u)
is the gradient of . Notice two things: first, the gradient of a scalar field is a vector field (that
is, a map from Rn into Rn ); and second, the differential da is the zero linear functional if and only
if the gradient at a, (a), is the zero vector in Rn .
Perhaps the most useful fact about the gradient of a scalar field at a point a in Rn is that it
is the vector at a which points in the direction of the most rapid increase of , a fact to be proved
in proposition 6.3.7.
6.3.3. Definition. Let f be a member of Fa (V, W ) and v be a nonzero vector in V . Then Dv f (a),
the derivative of f at a in the direction of v, is defined by
1
Dv f (a) := lim fa (tv)
t0 t
teaux differential (or
if this limit exists. This directional derivative is also called the Ga
teaux variation) of f , and is sometimes denoted by f (a; v). Many authors require that in
Ga
the preceding definition v be a unit vector. We will not adopt this convention.
6.3. THE GRADIENT OF A SCALAR FIELD IN Rn
71
Recall that for 0 6= v V the curve ` : R V defined by `(t) = a + tv is the parametrized line
through a in the direction of v. In the following proposition, which helps illuminate our use of the
adjective directional, we understand the domain of f ` to be the set of all numbers t for which
the expression f (`(t)) makes sense; that is,
dom(f `) = {t R : a + tv dom f } .
Since a is an interior point of the domain of f , the domain of f ` contains an open interval about 0.
6.3.4. Proposition. If f Da (V, W ) and 0 6= v V , then the directional derivative Dv f (a) exists
and is the tangent vector to the curve f ` at 0 (where ` is the parametrized line through a in the
direction of v). That is,
Dv f (a) = D(f `)(0) .
1
7
6.3.5. Example. Let f (x, y) = ln x2 + y 2 2 . Then Dv f (a) = 10
when a = (1, 1) and v = ( 35 , 45 ).
6.3.6. Proposition. If f Da (V, W ), then for every nonzero v in V
Dv f (a) = dfa (v) .
6.3.7. Proposition. Let : U R be a scalar field on a subset U of Rn . If is differentiable at a
point a in U and da is not the zero functional, then the maximum value of the directional derivative
Du (a), taken over all unit vectors u in Rn , is achieved when u points in the direction.of the gradient
(a). The minimum value is achieved when u points in the opposite direction (a).
What role do partial derivatives play in all this? Conceptually, not much of one. They are
just directional derivatives in the directions of the standard basis vector of Rn . They are, however,
useful for computation. For example, if F is a mapping from Rn to Rm differentiable at a point a,
then the matrix representation of dFa is an m n matrix (the so-called Jacobian matrix ) whose
F j
(where F j is the j th
entry in the j th row and k th column is the partial derivative Fkj =
xk
n
coordinate function
of F ). And

if is a differentiable scalar field on R . then its gradient can be
represented as
,...,
.
x1
xn
6.3.8. Exercise. Take any elementary calculus text and derive every item called a chain rule in
that text from theorem 6.2.9.
CHAPTER 7
MULTILINEAR MAPS AND DETERMINANTS

7.1. Permutations
A bijective map : X X from a set X onto itself is a permutation of the set. If x1 ,
x2 , . . . , xn are distinct elements of a set X, then the permutation of X that maps x1 7 x2 ,
x2 7 x3 , . . . , xn1 7 xn , xn 7 x1 and leaves all other elements of X fixed is a cycle (or cyclic
permutation) of length n. A cycle of length 2 is a transposition. Permutations 1 , . . . , n
of a set X are disjoint if each x X is moved by at most one j ; that is, if j (x) 6= x for at most
one j Nn := {1, 2, . . . , n}.
7.1.1. Proposition. If X is a nonempty set, the set of permutations of X is a group under
composition.
Notice that if and are disjoint permutations of a set X, then = . If X is a set
with n elements, then the group of permutations of X (which we may identify with the group of
permutations of the set Nn ) is the symmetric group on n elements (or on n letters); it is
denoted by Sn .
7.1.2. Proposition. Any permutation 6= idX of a finite set X can be written as a product
(composite) of cycles of length at least 2.
Proof. See[28], chapter 8, theorem 1.
A permutation of a finite set X is even if it can be written as the product of an even number of
transpositions, and it is odd if it can be written as a product of an odd number of transpositions.
7.1.3. Proposition. Every permutation of a finite set is either even or odd, but not both.
Proof. See[28], chapter 8, theorem 3.
7.1.4. Definition. The sign of a permutation , denoted by sgn , is +1 if is even and 1 if
is odd.
7.2. Multilinear Maps

7.2.1. Definition. Let V1 , V2 , . . . , Vn , and W be vector spaces over a field F. We say that
a function f : V1 Vn W is multilinear (or n-linear) if it is linear in each of its n
variables. We ordinarily call 2-linear maps bilinear and 3-linear maps trilinear. We denote by
Ln (V1 , . . . , Vn ; W ) the family of all n-linear maps from V1 Vn into W . A multilinear map
from the product V1 Vn into the scalar field F is a multilinear form (or a multilinear
functional.
7.2.2. Exercise. Let V and W be vector spaces over a field F; u, v, x, y V ; and F.
(a) Expand T (u + v, x + y) if T is a bilinear map from V V into W .
(b) Expand T (u + v, x + y) if T is a linear map from V V into W .
(c) Write T (x, y) in terms of and T (x, y) if T is a bilinear map from V V into W .
(d) Write T (x, y) in terms of and T (x, y) if T is a linear map from V V into W .
7.2.3. Example. Composition of operators on a vector space V is a bilinear map on L(V ).
73
74
7. MULTILINEAR MAPS AND DETERMINANTS
7.2.4. Proposition. If U , V , and W are vector spaces over a field F, then so is L2 (U, V ; W ).
Furthermore the spaces L(U, L(V, W )) and L2 (U, V ; W ) are (naturally) isomorphic.
Hint for proof . The isomorphism is implemented by the map
F : L(U, L(V, W )) L2 (U, V ; W ) : 7
v) := ((u))(v) for all u U and v V .
where (u,
7.2.5. Definition. A multilinear map f : V n W from the n-fold product V V of a vector
space V into a vector space W is alternating if f (v1 , . . . , vn ) = 0 whenever vi = vj for some
i 6= j.
7.2.6. Exercise. Let V = R2 and f : V 2 R : (v, w) 7 v1 w2 . Is f bilinear? Is it alternating?
7.2.7. Exercise. Let V = R2 and g : V 2 R : (v, w) 7 v1 + w2 . Is g bilinear? Is it alternating?
7.2.8. Exercise. Let V = R2 and h : V 2 R : (v, w) 7 v1 w2 v2 w1 . Is h bilinear? Is it
alternating? If {e1 , e2 } is the usual basis for R2 , what is h(e1 , e2 )?
7.2.9. Definition. If V and W are vector spaces, a multilinear map f : V n W is skewsymmetric if

f (v 1 , . . . , v n ) = (sgn )f v (1) , . . . , v (n)
for all Sn .
7.2.10. Proposition. Suppose that V and W be vector spaces. Then every alternating multilinear
map f : V n W is skew-symmetric.
Hint for proof . Consider f (u + v, u + v) in the bilinear case.
7.2.11. Remark. If a function f : Rn R is differentiable, then at each point a in Rn the
differential of f at a is a linear map from Rn into R. Thus we regard df : a 7 dfa (the differential
of f ) as a map from Rn into L(Rn , R). It is natural to inquire whether the function df is itself
differentiable. If it is, its differential at a (which we denote by d 2fa ) is a linear map from Rn into
L(Rn , R); that is
d 2fa L(Rn , L(Rn , R)).
In the same vein, since d 2f maps Rn into L(Rn , L(Rn , R)), its differential (if it exists) belongs to
L(Rn , L(Rn , L(Rn , R))). It is moderately unpleasant to contemplate what an element of L(Rn , L(Rn , R))
or of L(Rn , L(Rn , L(Rn , R))) might look like. And clearly as we pass to even higher order differentials things look worse and worse. It is comforting to discover that an element of L(Rn , L(Rn , R))
may be regarded as a map from (Rn )2 into R which is bilinear (that is, linear in both of its variables), and that L(Rn , L(Rn , L(Rn , R))) may be thought of as a map from (Rn )3 into R which is
linear in each of its three variables. More generally, if V1 , V2 , V3 , and W are arbitrary vector spaces
it will be possible to identify the vector space L(V1 , L(V2 , W ))) with the space of bilinear maps from
V1 V2 to W , the vector space L(V1 , L(V2 , L(V3 , W ))) with the trilinear maps from V1 V2 V3
to W , and so on (see, for example, proposition 7.2.4).
7.3. Determinants
7.3.1. Definition. A field F is of characteristic zero if n1 = 0 for no n N.
7.3.2. Convention. In the following material on determinants, we will assume that the scalar
fields underlying all the vector spaces and algebras we encounter are of characteristic zero. Thus
multilinear functions will be alternating if and only if they are skew-symmetric. (See exercises 7.2.10
and 7.3.7.)
7.3.3.
Let A be a unital commutative algebra. In the sequel we identify the algebra
n Remark.
n
n
A
= A An (n factors) with the algebra Mn (A) of n n matrices of elements of A by
n
regarding the term ak in (a1 , . . . , an ) An as the k th column vector of an nn matrix of elements
7.3. DETERMINANTS
75
of A. There are many standard notations for the same thing: Mn (A), An An (n factors),
n
2
An , Ann , and An , for example.
The identity matrix, which we usually denote by I, in Mn (A) is (e1 , . . . , en ), where e1 , . . . , en
are the standard basis vectors for An ; that is, e1 = (1A , 0, 0, . . . ), e2 = (0, 1A , 0, 0, . . . ), and so on.
7.3.4. Definition. Let A be a unital commutative algebra. A determinant function is an
alternating multilinear map D : Mn (A) A such that D(I) = 1A .
7.3.5. Proposition. Let V = Rn . Define
: V n R : (v 1 , . . . , v n ) 7
1
n
(sgn )v(1)
. . . v(n)
.
Sn
Then is a determinant function which satisfies (e1 , . . . , en ) = 1.

Note: If A is an n n matrix of real numbers we define det A, the determinant of A, to be
(v 1 , . . . , v n ) where v 1 , . . . , v n are the column vectors of the matrix A.
1
3 2
7.3.6. Exercise. Let A = 1 0 3. Use the definition above to find det A.
2 2 1
7.3.7. Proposition. If V and W are vector spaces over a field F of characteristic zero and f : V n
W is a skew-symmetric multilinear map, then f is alternating.
7.3.8. Proposition. Let V and W be vector spaces (over a field of characteristic 0) and f : V n
W be a multilinear map. If f (v 1 , . . . , v n ) = 0 whenever two consecutive terms in the n-tuple
(v 1 , . . . , v n ) are equal, then f is skew-symmetric and therefore alternating.
7.3.9. Proposition. Let f : V n W be an alternating multilinear map, j 6= k in Nn , and be a
scalar. Then
f (v 1 , . . . , v j + v k , . . . , v n ) = f (v 1 , . . . , v j , . . . , v n ).
7.3.10. Proposition. Let A be a unital commutative algebra (over a field of characteristic zero)
and n N. A determinant function exists on Mn (A). Hint. Consider
X
det : Mn (A) A : (a1 , . . . an ) 7
(sgn )a1(1) . . . an(n) .
Sn
7.3.11. Proposition. Let D be an alternating multilinear map on Mn (A) where A is a unital

commutative algebra and n N. For every C Mn (A)
D(C) = D(I) det C.
7.3.12. Proposition. Show that the determinant function on Mn (A), where A is a unital commutative algebra, is unique.
7.3.13. Proposition. Let A be a unital commutative algebra and B, C Mn (A). Then
det(BC) = det B det C.
Hint for proof . Consider the function D(C) = D(c1 , . . . , cn ) := det(Bc1 , . . . , Bcn ), where Bck
is the product of the n n matrix B and the k th column vector of C.
7.3.14. Proposition. For an n n matrix B let B t , the transpose
jof
B, be thet matrix
i obtained
from B by interchanging its rows and columns; that is, if B = bi , then B = bj . Then
det B t = det B.
76
7.4. Tensor Products of Vector Spaces

For a modern and very careful exposition of tensor products, which is more extensive than given
here, I recommend chapter 14 of [29].)
7.4.1. Definition. Let U and V be vector spaces over a field F. A vector space U V together
with a bilinear map : U V U V is a tensor product of U and V if for every vector space
e : U V W which
W and every bilinear map B : U V W , there exists a unique linear map B
makes the following diagram commute.
U V
/ U V
e
B

7.4.2. Proposition. In the category of vector spaces and linear maps if tensor products exist, then
they are unique (up to isomorphism).
7.4.3. Proposition. In the category of vector spaces and linear maps tensor products exist.
Hint for proof . Let U and V be vector spaces over a field F. Consider the free vector space
lc (U V ) = lc (U V , F). Define
: U V lc (U V ) : (u, v) 7 {(u,v)} .
Write u v instead of (u, v). Then let
S1 = {(u1 + u2 ) v u1 v u2 v : u1 , u2 U and v V },
S2 = {(u) v (u v) : F, u U , and v V },
S3 = {u (v1 + v2 ) u v1 u v2 : u U and v1 , v2 V },
S4 = {u (v) (u v) : F, u U , and v V },
S = span(S1 S2 S3 S4 ), and
U V = lc (U V )/S .
Also define
: U V U V : (u, v) 7 [u v].
Then show that U V and satisfy the conditions stated in definition 7.4.1.

7.4.4. Notation. It is conventional to write u v for (u, v) = [u v]. Tensors of the form
uv are called elementary tensors (or decomposable tensors or homogeneous tensors).
Keep in mind that
7.4.5. Proposition. Let u and v be elements of finite dimensional vector spaces U and V , respectively. If u v = 0, then either u = 0 or v = 0.
Hint for proof . Argue by contradiction. Suppose there exist u0 6= 0 in U and v 0 6= 0 in V such
that u0 v 0 = 0. Use propositon 2.5.8 to choose linear functionals f U and g V such that
f (u0 ) = g(v 0 ) = 1. Consider the map B defined on U V by B(u, v) = f (u)g(v).
7.4.6. CAUTION. One needs to exercise some care in dealing with elementary tensors: keep in
mind that
(1) not every member of U V is of the form u v;
(2) the representation of a tensor as an elementary tensor, even when it is possible, fails to be
unique; and
(3) the family of elementary tensors (although it spans U V ) is by no means linearly independent.
7.4. TENSOR PRODUCTS OF VECTOR SPACES
77
We do, however, have the following useful result.

7.4.7. Proposition. Let e1 , . . . , en be linearly independent
vectors in a vector space U and v 1 , . . . , v n
Pn
j
be arbitrary vectors in a vector space V . Then j=1 e v j = 0 if and only if v k = 0 for each
k Nn .
Hint for proof . Extend {e1 , . . . , en } to a basis E for U . Fix k Nn . Let V . Consider the
scalar valued function B defined on U V by B(u, v) = ek (u)(v), where ek is as defined in

proposition 2.5.4. Prove that (v k ) = 0.
j n
7.4.8. Proposition. If {ei }m
i=1 and {f }j=1 are bases for the finite dimensional vector spaces U
m n is a basis for U V .
and V , respectively, then the family {ei f j }i=1,
j=1
7.4.9. Corollary. If U and V are finite dimensional vector spaces, then so is U V and
dim(U V ) = (dim U )(dim V ).
7.4.10. Proposition. Let U and V be finite dimensional vector spaces and {f j }nj=1 be a basis
for V . Then for every element t U V there exist unique vectors u1 , . . . un U such that
n
X
t=
uj f j .
j=1
7.4.11. Proposition. If U and V are vector spaces, then

U V
= V U.
7.4.12. Proposition. If V is a vector space over a field F, then
V F
=V
= F V.
7.4.13. Proposition. Let U , V , and W be vector spaces. For every vector space X and every
trilinear map k : U V W X there exists a unique linear map e
k : U (V W ) : X such
that

e
k u (v w) = k(u, v, w)
for all u U , v V , and w W .
7.4.14. Proposition. If U , V , and W are vector spaces, then
U (V W )
= (U V ) W.
7.4.15. Proposition. Let U and V be vector spaces. Then there exists an injective linear map
from U V into L(U, V ). If U and V are finite dimensional, then U V
= L(V, U ).
Hint for proof . Consider the map B : U V L(V, U ) : (u, g) 7 B(u, g) where
B(u, g)(v) = g(v) u
for every v V .
7.4.16. Proposition. If U , V , and W are vector spaces, then
U (V W )
= (U V ) (U W ).
7.4.17. Proposition. If U and V are finite dimensional vector spaces, then
(U V )
= U V .
7.4.18. Proposition. If U , V , and W are finite dimensional vector spaces, then
L(U V, W )
= L(U, L(V, W ))
= L2 (U, V ; W ).
7.4.19. Proposition. Let u1 , u2 U and v 1 , v 2 V where U and V are finite dimensional vector
spaces. If u1 v 1 = u2 v 2 6= 0, then u2 = u1 and v 2 = v 1 where = 1.
78
7.5. Tensor Products of Linear Maps

7.5.1. Definition. Let S : U W and T : V X be linear maps between vector spaces. We
define the tensor product of the linear maps S and T by
S T : U V W X : u v 7 S(u) T (v) .
7.5.2. Exercise. Definition 7.5.1 defines the tensor product S T of two maps only for homogeneous elements of U V . Explain exactly what is needed to convince ourselves that S T is well
defined on all of U V . Then prove that S T is a linear map.
7.5.3. Exercise. Some authors hesitate to use the notation S T for the mapping defined in 7.5.1
on the (very reasonable) grounds that S T already has a meaning; it is a member of the vector
space L(U, W ) L(V, X). Discuss this problem and explain, in particular, why the use of the
notation S T in 7.5.1 is not altogether unreasonable.
7.5.4. Proposition. Suppose that R, S L(U, W ) and that T L(V, X) where U , V , W , and X
are finite dimensional vector spaces. Then
(R + S) T = R T + S T.
7.5.5. Proposition. Suppose that R L(U, W ) and that S, T L(V, X) where U , V , W , and X
R (S + T ) = R S + R T.
7.5.6. Proposition. Suppose that S L(U, W ) and that T L(V, X) where U , V , W , and X are
finite dimensional vector spaces. Then for all scalars and
(S) (T ) = (S T ).
7.5.7. Proposition. Suppose that Q L(U, W ), R L(V, X), S L(W, Y ), and that T L(X, Z)
where U , V , W , X, Y , and Z are finite dimensional vector spaces. Then
(S T )(Q R) = SQ T R.
7.5.8. Proposition. If U and V are finite dimensional vector spaces, then
IU IV = IU V .
7.5.9. Proposition. Suppose that S L(U, W ) and that T L(V, X) where U , V , W , and X are
finite dimensional vector spaces. If S and T are invertible, then so is S T and
(S T )1 = S 1 T 1 .
7.5.10. Proposition. Suppose that S L(U, W ) and that T L(V, X) where U , V , W , and X
are finite dimensional vector spaces. If S T = 0, then either S = 0 or T = 0.
ran(S T ) = ran S ran T.
ker(S T ) = ker S V + U ker T.
CHAPTER 8
TENSOR ALGEBRAS
8.1. Grassmann Algebras
8.1.1. Definition. Let V be an d-dimensional vector space over a field F. We say that
the Grassmann algebra (or the exterior algebra) over V if
V
(1) (V ) is a unital algebra
V over F (multiplication is denoted by ),
(2) V is contained in (V ),
(3) v vV= 0 for every v V ,
(4) V
dim( (V )) = 2d , and
(5) (V ) is generated by 1V(V ) and V .
V
(V ) is
The multiplication in a Grassmann algebra is called the wedge product (or the exterior
product).
8.1.2. Exercise. There are two instances in the preceding definition where I have opted for brevity
over precision. Explain why in definition 8.1.1 contained in appears in quotation marks. Give a
more precise version
of condition (2). Also, explain more precisely what is meant, in condition (5),
V
by saying that (V ) is generated by 1 and V .
V
8.1.3. Proposition. If (V ) is a Grassmann
algebra over a vector space V , then
V
V the zero vector
of V is an annihilator in the algebra (V ). That is, 0V g = 0V for every g (V ).
V
8.1.4. Proposition. If (V ) is a Grassmann algebra over a vector space V , then 1V(V )
/ V.
8.1.5. Proposition.
V Let v and w be elements of a finite dimensional vector space V . In the
Grassmann algebra (V ) generated by V
v w = w v .
8.1.6. Proposition. Let V be a d-dimensional vector space with basis E = {e1 , . . . , ed }. For each
nonempty subset S = {ei1 , ei2 , . . . , eip } of E with i1 < i2 < < ip let eS =Vei1 ei2 eip .
Also let e = 1V(V ) . Then {eS : S E} is a basis for the Grassmann algebra (V ).
L
8.1.7. Definition. An algebra A is a Z+ -graded algebra if it is a direct sum A =
Ak of
k0
vector subspaces Ak and its multiplication takes elements in Aj Ak to elements in Aj+k for all
j, k Z+ . An element in Ak is said to be a homogeneous element of degree k (or of grade k).
The definitions of Z-graded algebras, N-graded algebras and Z2 -graded algebras are similar. (In
the case of a Z2 -graded algebra the indices are 0 and 1 and A1 A1 A0 .) Usually the unmodified
expression graded algebra refers to an Z+ -graded algebra.
V
Proposition 8.1.8 says that every Grassmann algebra (V ) over a vector space V is a graded
V
algebra. The set of elements homogeneous of degree k is denoted by k (V ). An element of
Vk
(V ) which can be written in the form v1 v2 vk (where v1 , . . . , vk all belong to V ) is a
decomposable element of degree k (or of grade k).
8.1.8. Proposition. Every Grassmann algebra is a Z+ -graded algebra.
V
V
We denote by k (V ) the subspace of all homogeneous elements of degree k in (V ). In
V
V
V
particular, 0 (V ) = F and 1 (V ) = V . If the dimension of V is d, take k (V ) = {0} for all k > d.
V
Vk
(And if you wish to regard (V ) as a Z-graded algebra also take
(V ) = {0} whenever k < 0.)
79
80
8. TENSOR ALGEBRAS
8.1.9. Example. If the dimension of a vector space V is 3 or less, then every homogeneous element
of the corresponding Grassmann algebra is decomposable.
8.1.10. Example. If the dimension of a (finite dimensional) vector space V is at least four,
then there exist homogeneous elements in the corresponding Grassmann algebra which are not
decomposable.
(e3
Hint for proof . Let e1 , e2 , e3 , and e4 be distinct basis elements of V and consider (e1 e2 ) +
e4 ).
8.1.11. Proposition. The elements v1 , v2 , . . . , vp in a vector space V areVlinearly independent if

and only if v1 v2 vp 6= 0 in the corresponding Grassmann algebra (V ).
8.1.12. Proposition. Let T : V W be a linear map between finite dimensional
spaces.
V
V vector V
Then there exists a unique extension of T to a unital algebra homomorphism (T ) : (V ) (W ).
V
V
This extension maps k (V ) into k (W ) for each k N.
V
V
8.1.13. Example. The pair of maps V 7 (V ) and T 7 (T ) is a covariant functor from the
category of vector spaces and linear maps to the category of unital algebras and unital algebra
homomorphisms.

V
8.1.14. Proposition. If V is a vector space of dimension d, then dim p (V ) = dp for 0 p d.
8.1.15. Convention. Let V be a d-dimensional vector space. Since the map 7 1V(V ) is
V
an obvious isomorphism between F and the one-dimensional space 0 (V ), we identify these two
Vd
V0
(V ) is also one-dimensional,
(V ) as a scalar. And since
spaces, and refer to an element of
its elements are frequently referred to as pseudoscalars.
V
V
8.1.16. Proposition. If V is a finite dimensional vector space, p (V ), and q (V ), then
= (1)pq .
8.2. Existence of Grassmann Algebras
8.2.1. Definition. Let V0 , V1 , V2 , . . . be vector spaces (over the same field). Then their (ex
L
ternal) direct sum, which is denoted by
Vk , is defined to be the set of all functions
k=0
S
+
v : Z+
k=0 Vk with finite support such that v(k) = vk Vk for each k Z . The usual
pointwise addition and scalar multiplication make this set into a vector space.
8.2.2. Definition. Let V be a vector space over a field F. Define T 0 (V ) = F, T 1 (V ) = V ,
T 2 (V ) = V V , T 3 (V ) = V V V , . . . , T k (V ) = V V (k factors), . . . . Then let
L
T (V ) =
T k (V ). Define multiplication on T (V ) by using the obvious isomorphism
k=0
T k (V ) T m (V )
= T k+m (V )
and extending by linearity to all of T (V ). The resulting algebra is the tensor algebra of V (or
generated by V ).
8.2.3. Proposition. The tensor algebra T (V ) as defined in 8.2.2 is in fact a unital algebra.
8.2.4. Proposition. Let V be a finite dimensional vector space and J be the ideal in the tensor
algebra T (V ) generated by the set of all elements of the form v v where v V . Then the quotient
algebra T (V )/J is the Grassmann algebra over V (or, equivalently, over V ).
8.2.5. Notation. If x and y are elements of the tensor algebra T (V ), then in the quotient algebra
T (V )/J the product of [x] and [y] is written using wedge notation; that is,
[x] [y] = [x y].
8.2. EXISTENCE OF GRASSMANN ALGEBRAS
81
8.2.6. Notation. If V is a vector space over F and k N we denote by Altk (V ) the set of all
alternating k-linear maps from V k into F. (The space Alt1 (V ) is just V .) Additionally, take
Alt0 (V ) = F.
8.2.7. Example. If V is a finite dimensional vector space and k > dim V , then Altk (V ) = {0}.
8.2.8. Definition. Let p, q N. We say that a permutation Sp+q is a (p, q)-shuffle if
(1) < < (p) and (p + 1) < < (p + q). The set of all such permutations is denoted
by S(p, q).
8.2.9. Example. Give an example of a (4, 5)-shuffle permutation of the set N9 = {1, . . . , 9} such
that (7) = 4.
8.2.10. Definition. Let V be a vector space over a field of characteristic 0. For p, q N define
: Altp (V ) Altq (V ) Altp+q (V ) : (, ) 7
where
( )(v 1 , . . . , v p+q ) =
(sgn )(v (1) , . . . , v (p) )(v (p+1) , . . . , v (p+q) ).
S(p,q)
8.2.11. Exercise. Show that definition 8.2.10 is not overly optimistic by verifying that if
Altp (V ) and Altq (V ), then Altp+q (V ).
8.2.12. Proposition. The multiplication defined in 8.2.10 is associative. That is if Altp (V ),
Altq (V ), and Altr (V ), then
( ) = ( ) .
8.2.13. Exercise. Let V be a finite dimensional vector space over a field of characteristic zero.
Explain in detail how to make Altk (V ) (or, if you prefer, Altk (V ) ) into a vector space for each
k Z and how to make the collection of these into a Z-graded algebra. Show that this algebra is
the Grassmann algebra generated by V . Hint. Take Altk (V ) = {0} for each k < 0 and extend the
definition of the wedge product so that if Alt0 (V ) = F and Altp (V ), then = .
8.2.14. Proposition. Let 1 , . . . , p be members of Alt1 (V ) (that is, linear functionals on V ).
Then

p
(1 p )(v 1 , . . . , v p ) = det j (v k ) j,k=1
for all v 1 , . . . , v p V .
8.2.15. Proposition. If {e1 , . . . , en } is a basis for an n-dimensional vector space V , then
{e(1) e(p) : S(p, n p)}
is a basis for Altp (V ).
8.2.16. Proposition. For T : V W a linear map between vector spaces define
Altp (T ) : Altp (W ) Altp (V ) : 7 Altp (T )()

where Altp (T )() (v 1 , . . . , v p ) = (T v 1 , . . . , T v p ) for all v 1 , . . . , v p V . Then Altp is a contravariant functor from the category of vector space and linear maps into itself.
8.2.17. Exercise. Let V be an n-dimensional vector space and T L(V ). If T is diagonalizable,
then
n
X
cT () =
(1)k [Altnk (T )]k .
k=0
82
8. TENSOR ALGEBRAS
8.3. The Hodge -operator

8.3.1. Definition. Let E be a basis for an n-dimensional vector space V . Then the n-tuple
(e1 , . . . , en ) is an ordered basis for V if e1 , . . . , en are distinct elements of E.
8.3.2. Definition. Let E = (e1 , . . . , en ) be an ordered basis for Rn . We say that the basis E is
right-handed if det[e1 , . . . , en ] > 0 and left-handed otherwise.
8.3.3. Definition. Let V be a real n-dimensional vector space and T : Rn V be an isomorphism.
Then the set of all n-tuples of the form (T (e1 ), . . . , T (en )) where (e1 , . . . , en ) is a right-handed basis
in Rn is an orientation of V . Another orientation consists of the set of n-tuples (T (e1 ), . . . , T (en ))
where (e1 , . . . , en ) is a left-handed basis in Rn . Each of these orientations is the opposite (or
reverse) of the other. A vector space together with one of these orientations is an oriented
vector space.
8.3.4. Exercise. Let V be an n-dimensional real inner product space. In exercise 2.5.14 we established an isomorphism : v 7 v between V and its dual space V . Show how this isomorphism
can be used to induce an inner product on V . Then show how this may be used to create an
inner product on Altp (V ) for 2 p n. Hint. For v, w V let hv , w i = hv, wi. Then for
1 , . . . , p , 1 , . . . , p Alt1 (V ) let h1 p , 1 p i = det[hj , k i].
8.3.5. Proposition. Let V be an d-dimensional oriented real inner product space. Fix a unit
vector vol Altd (V ). This vector is called a volume element. (In the case where V = Rd , we
will always choose vol = e1 ed where (e1 , . . . , ed ) is the usual ordered basis for Rd .)
Let Altp (V ) and q = d p. Then there exists a vector Altq (V ) such that
h , i vol =
for each Altq (V ). Furthermore, the map 7 from Altp (V ) into Altq (V ) is a vector space
isomorphism. This map is the Hodge star operator.
8.3.6. Proposition. Let V be a finite dimensional oriented real inner product space of dimension n.
Suppose that p + q = n. Then = (1)pq for every Altp (V ).
CHAPTER 9
DIFFERENTIAL MANIFOLDS
The purpose of this chapter and the next is to examine an important and nontrivial example
of a Grassmann algebra: the algebra of differential forms on a differentiable manifold. If your
background includes a study of manifolds, skip this chapter. If you are pressed for time, reading just
the first section of the chapter should enable you to make sense out of most of the ensuing material.
That section deals with familiar manifolds in low (three or less) dimensional Euclidean spaces. The
major weakness of this presentation is that it treats manifolds in a non-coordinate-free manner as
subsets of some larger Euclidean space. (A helix, for example, is a 1-manifold embedded in 3-space.)
The rest of the chapter gives a (very brief) introduction to a more satisfactory coordinate-free way of
viewing manifolds. For a less sketchy view of the subject read one of the many splendid introductory
books in the field. I particularly like [3] and [21].
9.1. Manifolds in R3
A 0-manifold is a point (or finite collection of points).
A function is smooth if it is infinitely differentiable (that is, if it has derivatives of all orders).
A curve is a continuous image of a closed line segment in R. If C is a curve, the choice of an
interval [a, b] and a continuous function f such that C = f ([a, b]) is a parametrization of C.
If the function f is smooth, we say that C is a smooth curve.
A 1-manifold is a curve (or finite collection of curves). A 1-manifold is flat if it is contained
in some line in R3 . For example, the line segment connecting two points in R3 is a flat 1-manifold.
A surface is a continuous image of a closed rectangular region in R2 . If S is a surface, the the
choice of a rectangle R = [a1 , b1 ] [a2 , b2 ] and a continuous function f such that S = f (R) is a
parametrization of S. If the function f is smooth, we say that S is a smooth surface.
A 2-manifold is a surface (or finite collection of surfaces). A 2-manifold is flat if it is
contained in some plane in R3 . For example, the triangular region connecting the points (1, 0, 0),
(0, 1, 0), and (0, 0, 1) is a flat 2-manifold.
A solid is a continuous image of the 3-dimensional region determined by a closed rectangular
parallelepiped (to avoid a six-syllable word many people say rectangular solid or even just box )
in R3 . If E is a solid, then the choice of a rectangular parallelepiped P = [a1 , b1 ] [a2 , b2 ] [a3 , b3 ]
and a continuous function f such that E = f (P ) is a parametrization of E. If the function f
is smooth, we say that E is a smooth solid.
A 3-manifold is a solid (or finite collection of solids).
9.2. Charts, Atlases, and Manifolds
9.2.1. Definition. Let M and S be sets; U , V M , : U S, and : V S be injective maps.
Then the composite 1 , is taken to be the function
1 : (U V ) (U V ).
These composite maps are called, variously, transition maps or overlap maps or connecting
maps.
9.2.2. Proposition. The preceding definition makes sense and the composite map 1 is a
bijection.
83
84
9. DIFFERENTIAL MANIFOLDS
9.2.3. Definition. Let m, n N, and U Rm . A function F : U Rn is smooth (or infinitely

differentiable, or C ) if the differential dp Fa exists for every p N and every a U . We denote
by C (U, Rn ) the family of all smooth functions from U into Rn .
9.2.4. Definition. Let M be a topological space and n N. A pair (U, ), where U is an open
e is a homeomorphism from U to an open subset U
e of Rn , is called an
subset of M and : U U
(n-dimensional coordinate) chart. (The notation here is a bit redundant. If we know the
function , then we also know its domain U . Indeed, do not be surprised to see reference to the
chart or to the chart U .)
Let p M . A chart (U, ) is said to contain p if p U and is said to be a chart (centered)
at p if (p) = 0. A family of n-dimensional coordinate charts whose domains cover M is an
(n-dimensional) atlas for M . If such an atlas exists, the space M is said to be locally
Euclidean.
9.2.5. Notation. Let n N. For 1 k n the function k : Rn R : x = (x1 , x2 , . . . , xn ) 7 xk
is the k th coordinate projection. If : M Rn is a chart on a topological space, one might
reasonably expect the n component functions of (that is, the functions k ) to be called 1 ,
. . . , n . But this is uncommon. People seem to like and as names for charts. But then the
components of these maps have names such as x1 , . . . , xn , or y 1 , . . . , y n . Thus we usually end up
with something like (p) = x1 (p), . . . , xn (p) . The numbers x1 (p), . . . , xn (p) are called the local
coordinates of p.
Two common exceptions to this notational convention occur in the cases when n = 2 or n = 3.
In the former case you are likely to see things like = (x, y) and = (u, v) for charts on 2-manifolds.
Similarly, for 3-manifolds expect to see notations such as = (x, y, z) and = (u, v, w).
9.2.6. Definition. A second countable Hausdorff topological space M equipped with an n-dimensional
atlas is a topological n-manifold (or just a topological manifold).
9.2.7. Definition. Charts and of a topological n-manifold M are said to be (smoothly)
compatible if the transition maps 1 and 1 are smooth. An atlas on M is a smooth
atlas if every pair of its charts is smoothly compatible. Two atlases on M are (smoothly)
compatible (or equivalent) if every chart of one atlas is smoothly compatible with every chart
of the second; that is, if their union is a smooth atlas on M .
9.2.8. Proposition. Every smooth atlas on a topological manifold is contained in a unique maximal
smooth atlas.
9.2.9. Definition. A maximal smooth atlas on a topological manifold M is a differential
structure on M . A topological n-manifold which has been given a differential structure is a
smooth n-manifold (or a differential n-manifold, or a C n-manifold).
NOTE: From now on we will be concerned only with differential manifolds; so the modifier
smooth will ordinarily be omitted when we refer to charts, to atlases, or to manifolds. Thus it
will be understood that by manifold we mean a topological n-manifold (for some fixed n) equipped
with a differential structure to which all the charts we mention belong.
9.2.10. Example. Let U be an open subset of Rn for some n Z+ and : U Rn : x 7 x be
the inclusion map. Then {} is a smooth atlas for U . We make the convention that when an open
subset of Rn is regarded as an n-manifold we will suppose, unless the contrary is explicitly stated,
that the inclusion map is a chart in its differentiable structure.
9.2.11. Example (An atlas for S1 ). Let S1 = {(x, y) R2 : x2 + y 2 = 1} be the unit circle in R2
and U = {(x, y) S1 : y 6= 1}. Define : U R to be the projection of points
in U from the

point (0, 1) onto the x-axis; that is, if p = (x, y) is a point in U , then (p), 0 is the unique point
on the x-axis which is collinear with both (0, 1) and (x, y).
(1) Find an explicit formula for .
9.3. DIFFERENTIABLE FUNCTIONS BETWEEN MANIFOLDS
85
(2) Let V = {(x, y) S1 : y 6= 1}. Find an explicit formula for the projection of points in
V from (0, 1) onto the x-axis.
(3) The maps and are bijections between U and V , respectively, and R.
(4) The set {, } is a (smooth) atlas for S1 .
9.2.12. Example (An atlas for the n-sphere). The previous example can be generalized to the
n-sphere Sn = {x Rn+1 : kxk = 1}. The generalization of the mapping is called the stereographic projection from the south pole and the generalization of is the stereographic
projection from the north pole. (Find a simple expression for the transition maps.)
9.2.13. Example (Another atlas for S1 ). Let U = {(x, y) S1 : x 6= 1}. For (x, y) U let (x, y)
be the angle (measured counterclockwise) at the origin from (1, 0) to (x, y). (So (x, y) (0, 2).)
Let V = {(x, y) S1 : y 6= 1}. For (x, y) V let (x, y) be /2 plus the angle (measured
counterclockwise) at the origin from (0, 1) to (x, y). (So (x, y) (/2, 5/2).) Then {, } is a
(smooth) atlas for S1 .
9.2.14. Example (The projective plane P2 ). Let P2 be the set of all lines through the origin in R3 .
Such a line is determined by a nonzero vector lying on the line. Two nonzero vectors x = (x1 , x2 , x3 )
and y = (y 1 , y 2 , y 3 ) determine the same line if there exists R, 6= 0, such that y = x. In this
case we write x y. It is clear that is an equivalence relation. We regard a member of P2 as an
equivalence class of nonzero vectors. Let Uk = { [x] P2 : xk 6= 0} for k = 1, 2, 3. Also let

: U1 R2 : [(x, y, z)] 7 xy , xz ;

: U2 R2 : [(x, y, z)] 7 xy , yz ; and

: U3 R2 : [(x, y, z)] 7 xz , yz .
The preceding sets and maps are well defined; and {, , } is a (smooth) atlas for P2 .
9.2.15. Example (The general linear group). Let G = GL(n, R) be the group of nonsingular n n
matrices of real numbers. If a = [ajk ] and b = [bjk ] are members of G define
1
X
n
2
2
(ajk bjk )
d(a, b) =
.
j,k=1
The function d is a metric on G. Define

2
: G Rn : a = [ajk ] 7 a11 , . . . , a1n , a21 , . . . , a2n , . . . , an1 , . . . , ann .
Then {} is a (smooth) atlas on G. (Be a little careful here. There is one point that is not
completely obvious.)
9.2.16. Example. Let
I : R R : x 7 x ,
a: R R: x
7 arctan x , and
c : R R : x 7 x3 .
Each of {I}, {a}, and {c} is a (smooth) atlas for R. Which of these are equivalent?
9.3. Differentiable Functions Between Manifolds
9.3.1. Definition. A function F : M N between (smooth) manifolds is smooth at a point
m M if there exist charts (U, ) containing m and (V, ) containing F (m) such that F (U ) V
and the local representative F := F 1 : (U ) (V ) is smooth at (m). The
map F is smooth if it is smooth at every m M .
In the case that N is a subset of some Euclidean space Rn , it is the usual convention to use the
inclusion mapping : N Rn as the preferred chart on N . In this case the local representative of
F is written as F rather than F .
86
NOTE: It now makes sense to say (and is true) that a single chart on a manifold is a smooth map.
9.3.2. Proposition. Suppose that a map F : M N between manifolds is smooth at a point m
and that (W, ) and (X, ) are charts at m and F (m), respectively, such that F (W ) X. Then
F 1 is smooth at (m).
9.3.3. Proposition. If F : M N and G : N P are smooth maps between manifolds, then
G F is smooth.
9.3.4. Proposition. Every smooth map F : M N between manifolds is continuous.
9.3.5. Example. Consider the 2-manifold S2 with the differential structure generated by the
stereographic projections form the north and south poles (see example 9.2.12) and the 2-manifold
P2 with the differentiable structure generated by the atlas given in example 9.2.14. The map
F : S2 P2 : (x, y, z) 7 [(x, y, z)] is smooth. (Think of F as taking a point (x, y, z) in S2 to the
line in R3 passing through this point and the origin.)
9.4. The Geometric Tangent Space
9.4.1. Definition. Let J R be an interval with nonempty interior and M be a manifold. A
smooth function c : J M is a curve in M . If, in addition, 0 J and c(0) = m M , then we
say that c is a curve at m.
9.4.2. Example. The function b : t 7 t2 is a curve at 0 in R. Sketch the range of b.
The next example illustrates the point that the ranges of smooth curves may not look smooth.
9.4.3. Example. The function c : t 7 (cos3 t, sin3 t) is a curve at (1, 0) in R2 . Sketch its range.
Let V and W be normed linear spaces. Recall that a W -valued function f defined on some
neighborhood of 0 in V is said to belong to the family o(V, W ) provided that for every > 0 there
exists > 0 such that kf (x)k kxk whenever kxk . Recall also that W -valued functions f
and g, each defined on a neighborhood of 0 in V , are said to be tangent (at 0) if f g o(V, W ).
In this case we write f ' g.
Thus in the special case when b and c are curves at a point w in a normed linear space W ,
we should say that b and c are tangent at 0 if b c o(R, W ). As a matter of fact, it is almost
universal custom in this situation to say that b and c are tangent at w. This use of at w for
at 0 results from a tendency to picture curves in terms of their ranges. (For example, asked to
visualize the curve c : t 7 (cos t, sin t), most people see a circle. Of course, the circle is only the
range of c and not its graph, which is a helix in R3 .) We will follow convention and say that the
curves b and c are tangent at w. This convention will also apply to curves in manifolds.
9.4.4. Example. It is important to note that it cannot be determined whether two curves are
tangent just by looking at their ranges. The curves b : t 7 (t, t2 ) and c : t 7 (2t, 4t2 ) have identical
ranges; they both follow the parabola y = x2 in R2 (and in the same direction). They are both
curves at the origin. Nevertheless, they are not tangent at the origin.
Definition 9.4.6 below says that curves in a manifold are tangent if their composites with a
chart are tangent in the sense described above. Before formally stating this definition it is good
to know that tangency thus defined does not depend on the particular chart chosen.
9.4.5. Proposition. Let m be a point in a manifold and b and c be curves at m. If b ' c
for some chart centered at m, then b ' c for every chart centered at m.
Hint for proof . Use proposition 25.4.7 and problem 25.4.10 in [11].
9.4.6. Definition. Let m be a point in a manifold and b and c be curves at m. Then b and c are
tangent at m (we write b ' c) if b ' c for some (hence all) charts centered at m.
It is useful to know that smooth mappings between manifolds preserve tangency.
9.4. THE GEOMETRIC TANGENT SPACE
87
9.4.7. Proposition. If F : M N is a smooth mapping between manifolds and b and c are curves
tangent at a point m M , then the curves F b and F c are tangent at F (m) in N .
9.4.8. Definition. Since the family of little-oh functions is closed under addition it is obvious
that tangency at a point m is an equivalence relation on the family of curves at m. We denote the
equivalence class containing the curve c by e
c or, if we wish to emphasize the role of the point m, by
e
cm . Each equivalence class e
cm is a geometric tangent vector at m and the family of all such
vectors is the geometric tangent space at m. The geometric tangent space at m is denoted
by Tem (or, if we wish to emphasize the role of the manifold M , by Tem (M )).
The language (involving the words vector and space) in the preceding definition is highly
optimistic. So far we have a set of equivalence classeswith no vector space structure. The key to
providing Tem with a such a structure is exercise 2.2.16. There we found that a set S may be given
a vector space structure by transferring the structure from a known vector space V to the set S by
means of a bijection f : S V . We will show, in particular, that if M is an n-manifold, then for
each m M the geometric tangent space Tem can be given the vector space structure of Rn .
9.4.9. Definition. Let be a chart containing a point m in an n-manifold M and let u be a
nonzero vector in Rn . For every t R such that (m) + tu belongs to the range of , let

cu (t) = 1 (m) + tu .
Notice that since cu is the composite of smooth functions and since cu (0) = m, it is clear that cu
is a curve at m in M .
9.4.10. Example. If M is an n-manifold, then the curves ce1 , . . . , cen obtained by means of the
preceding definition from the standard basis vectors e1 , . . . , en of Rn will prove to be very useful. We will shorten the notation somewhat and write ck for cek (1 k n). We think of the
curves c1 , . . . , cn as being linearly independent directions in the tangent space at m (see proposition 9.4.14). We call these curves the standard basis curves at m determined by . It is important
to keep in mind that these curves depend on the choice of the chart ; the notation c1 , . . . , cn fails
to remind us of this.
9.4.11. Proposition. Let be a chart at a point m in an n-manifold. Then the map
C : Tem Rn : e
c 7 D( c)(0)
is well-defined and bijective.
Notice that initially we had no way of adding curves b and c at a point m in a manifold or of
multiplying them by scalars. Now, however, we can use the bijection C to transfer the vector
space structure from Rn to the tangent space Tem . Thus we add equivalence classes eb and e
c in the
obvious fashion. The formula is

eb + e
c = C 1 C (eb) + C (e
c) .
(9.1)
Similarly, if b is a curve at m and is a scalar, then

eb = C 1 C (eb) .
(9.2)
9.4.12. Corollary. At every point m in an n-manifold the geometric tangent space Tem may be
regarded as a vector space isomorphic to Rn .
As remarked previously, we have defined the vector space structure of the geometric tangent
space at m in terms of the mapping C , which in turn depends on the choice of a particular chart .
From this it might appear that addition and scalar multiplication on the tangent space depend on .
Happily this is not so.
9.4.13. Proposition. Let m be a point in an n-manifold. The vector space structure of the geometric tangent space Tem is independent of the particular chart used to define it.
88
9.4.14. Proposition. Let be a chart containing a point m in an n-manifold. Let c1 , . . . , cn be

the standard basis curves determined by (see example 9.4.10). Then {e
c1 , . . . , e
cn } is a basis for
e
the tangent space Tm .
9.4.15. Exercise. We know from thePpreceding proposition that if e
c belongs to Tem , then there
n
exist scalars 1 , . . . , n such that e
c = k=1 k e
ck . Find these scalars.
If F : M N is a smooth mapping between manifolds and m is a point in M , we denote by
e
dFm the mapping that takes each geometric tangent vector e
c in Tem to the corresponding geometric
e m is well-defined is clear from proposition 9.4.7. The

tangent vector (F c) in TeF (m) . That dF
point of proposition 9.4.17 below is to make the notation for this particular map plausible.
9.4.16. Definition. If F : M N is a smooth mapping between manifolds and m M , then the
function
e m : Tem TeF (m) : e
dF
c 7 (F c)
is the differential of F at m.
9.4.17. Proposition. Let F : M N be a smooth map from an n-manifold M to a p-manifold N .
For every m M and every pair of charts at m and at F (m) the following diagram commutes
Tem
e m
dF
/ TeF (m)
Rn
d F

/ Rp

(m)
e m is a linear map. (The maps C and C are defined as in proposition 9.4.11.)

and consequently dF
The map C has been used to provide the geometric tangent space Tem at a point m on an
n-manifold with the vector space structure of Rn . With equal ease it may be used to provide Tem
with a norm. By defining
ke
ck = kC (e
c)k
it is clear that we have made Tem into a normed linear space (but in a way that does depend on the
choice of the chart ). Furthermore, under this definition C is an isometric isomorphism between
Tem and Rn . Thus, in particular, we may regard Tm and TF (m) as Euclidean spaces. From the
e m is (continuous and) linear, being the composite of
preceding proposition we see that the map dF

1
the mapping d F (m) with two isometric isomorphisms C
and C .
If we use the mapping C to identify Tem and Rn as Banach spaces and, similarly, C to identify

e m and d F
are also identified. The
TeF (m) and Rp , then the continuous linear maps dF
(m)
e m is thus a reasonable one since, as we have just

notation which we have used for the mapping dF
seen, this mapping can be identified with the differential at (m) of the local representative of F
and since its definition does not depend on the charts and . To further strengthen the case for
the plausibility of this notation. consider what happens if M and N are open subsets of Rn and
Rp , respectively (regarded as manifolds whose differential structure is generated in each case by the
appropriate identity map). The corresponding local representative of F is F itself, in which case
the bottom map in the diagram for proposition 9.4.17 is simply dFm .
The preceding definition also helps to justify the usual intuitive picture for familiar n-manifolds
of the tangent space being a copy of Rn placed at m. Suppose that for a particular n-manifold
M there exists a smooth inclusion map of M into a higher dimensional Euclidean space Rp . (For
example, the inclusion map of the 2-manifold S2 into R3 is smooth.) Then taking F in the preceding
9.5. THE ALGEBRAIC TANGENT SPACE
89
e m maps Tem into a subspace of TeF (m) = Rp .

discussion to be this inclusion map, the function dF
Picture this subspace being translated to the point m = F (m) in Rp . To see how this works it is
best to work through the details of some examples.
9.4.18. Example. Let U = {(x, y, z) S2 : x > 0}. The map : U R2 : (x, y, z) 7 (y, z) is
a chart for
2-manifold S2 . Let : S2 R3 be the inclusion map of S2 into R3
and let m =
the
1 1 1
2
3
2 , 2 , 2 S . Then the range of d (m) is the plane in R whose equation is x + 2y + z = 0.
If we translate this plane to the point m, we obtain the plane whose equation is x + 2y + z = 2,
which is exactly the result we obtain using techniques of beginning calculus.to
find the equation

of the tangent plane to the surface x2 + y 2 + z 2 = 1 at the point 21 , 12 , 12 .
9.4.19. Example. Same example as the preceding except this time let U = {(x, y, z) S2 : z 6= 1}
and : U R2 be the stereographic projection of S2 from the south pole. That is,

x
y
.
(x, y, z) =
,
1+z 1+z
9.4.20. Proposition (A Chain Rule for Maps Between Manifolds). If F : M N and G : N P
are smooth mappings between manifolds, then
e F )m = dG
e F (m) dF
e m
d(G
(9.3)
for every m M .
9.5. The Algebraic Tangent Space
There is another way of looking at the tangent space at a point m in a manifold. Instead
of regarding tangent vectors as directions determined by equivalence classes of curves at m, we
will now consider them to be directional derivatives of (germs of) smooth functions defined in
neighborhoods of m.
(M, R) (the family of all
9.5.1. Definition. Let m be a point in a manifold M and f , g Cm
smooth real valued functions defined on a neighborhood of m). We write f g if there exists
a neighborhood of m on which f and g agree. Then is clearly an equivalence relation on
(M, R). The corresponding equivalence classes are germs of smooth functions at m. If f is a
Cm
(M, R), we denote the germ containing f by fb. The family of all germs of smooth
member of Cm
real valued functions st m is denoted by Gm (M ) (or just Gm ). Addition, multiplication, and scalar
multiplication of germs are defined as you would expect. For fb, gb Gm and R let
fb + gb = (f + g) b
fb gb = (f g) b
fb = (f ) b
(As usual, the domain of f + g and f g is taken to be dom f dom g.)
9.5.2. Proposition. If m is a point in a manifold, then the set Gm of germs of smooth functions
at m is (under the operations defined above) a unital commutative algebra.
9.5.3. Definition. Let m be a point in a manifold. A derivation on the algebra Gm of germs of
which satisfies Leibnizs rule
smooth functions at m is a linear functional v Gm
v(fb gb) = f (m)v(b
g ) + v(fb)g(m)
for all fb, gb Gm . Another name for a derivation on the algebra Gm is an algebraic tangent
vector at m. The set of all algebraic tangent vectors at m (that is, derivations on Gm ) will be
called the algebraic tangent space at m and will be denoted by Tbm (M ) (or just Tbm ).
90
The idea here is to bring to manifolds the concept of directional derivative; as the next example
shows directional derivatives at points in a normed linear space are derivations on the algebra of
(germs of) smooth functions.
9.5.4. Example. Let V be a normed linear space and let a and v be vectors in V . Define Dv,a on
functions f in Ga (V ) by
Dv,a (fb) := Dv f (a).
(Here, Dv f (a) is the usual directional derivative of f at a in the direction of v from beginning
calculus.) Then Dv,a is well-defined and is a derivation on Ga . (Think of the operator Dv,a as being
differentiation in the direction of the vector v followed by evaluation at a.)
9.5.5. Proposition. If m is a point in a manifold, v is a derivation on Gm , and k is a smooth
real valued function constant in some neighborhood of m, then v(b
k) = 0.
Hint for proof . Recall that 1 1 = 1.
It is an easy matter to show that the terminology algebraic tangent space adopted in definition 9.5.3 is not overly optimistic. It is a vector space.
9.5.6. Proposition. If m is a point in a manifold, then the tangent space Tbm is a vector space
under the usual pointwise definition of addition and scalar multiplication.
We now look at an example which establishes a first connection between the geometric and the
algebraic tangent spaces.
9.5.7. Example. Let m be a point in a manifold M and e
c be a vector in the geometric tangent
e
space Tm . Define
vec : Gm R : fb 7 D(f c)(0).
Then vec is well-defined and belongs to the algebraic tangent space Tbm .
The notation vec (fb) of the preceding example is not particularly attractive. In the following
material we will ordinarily write just vc (f ). Although strictly speaking this is incorrect, it should
not lead to confusion. We have shown that vec (fb) depends only on the equivalence classes e
c and fb,
not on the representatives chosen. Thus we do not distinguish between vb (g) and vc (f ) provided
that b and c belong to the same member of Tem and f and g to the same germ.
Example 9.5.7 is considerably more general than it may at first appear. Later in this section we
will show that the association e
c 7 vec is an isomorphism between the tangent spaces Tem and Tbm .
In particular, there are no derivations on Gm other than those induced by curves at m.
plausible for an n-manifold the use of the notation
For the moment, however, we wish to make

th standard basis curve at the point m determined by
for
the
derivation
v
,
where
c
is
the
k
ck
k
xk m
the chart = (x1 , . . . , xn ) (see the notational convention in 9.2.5 and example 9.4.10). The crux
of the matter is the following proposition, which says that if cu = 1 bu , where is a chart at
m and bu is the parametrized line through (m) in the direction of the vector u, then the value at
a germ fb of the derivation vcu may be found by taking the directional derivative in the direction u
of the local representative f .
9.5.8. Proposition. Let be a chart containing the point m in an n-manifold. If u Rn and
bu : R Rn :
t 7 (m) + tu, then cu := 1 bu is a curve at m and
vcu (f ) = Du (f )((m))
for all fb G(m).
Hint for proof . Use proposition 25.5.9 in [11].
91
9.5.9. Definition. In proposition 9.4.14 we saw that if is a chart containing a point m in an

n-manifold, then the vectors e
c1 , . . . , e
cn form a basis for the geometric tangent space Tem . We call
these vectors the basis vectors of Tem determined by .
9.5.10. Corollary. Let be a chart containing the point m in an n-manifold. If e
c1 , . . . , e
cn are the
basis vectors for Tem determined by , then

vck (f ) = f k ((m))
(9.4)

for every fb in Gm . (The subscript k on the right side of the equation indicates differentiation; f k
is the k th partial derivative of the local representative f , in another notation the right hand side
f
is xk ((m).)
9.5.11. Notation. The preceding corollary says that (in an n-manifold M ) the action of the
derivation vck on a function f (technically, on the germ fb) is that of partial differentiation of the
local representative f followed by evaluation at the point (m). In particular, if M happens to be
an open subset of Rn , then (9.4) becomes
vck (f ) = fk (m)
so that the value of vck at f is the k th partial derivative of f (evaluated at m). It is helpful
for the
notation to remind us of this fact; so in the following material we will usually write x k m for vck ,
where 1 k n and = (x1 , . . . , xn ) is the chart containing m in terms of which the curves ck
f
are defined. The value of this derivation at f will be denoted by x
k (m). Thus

f
(f ) = vck (f ) = veck (fb ).
(m) =
k
x
xk m
Let = (x1 , . . . , xn ) and = (y 1 , . . . , y n ) be charts containing a point m on an n-manifold.

It
is perhaps tempting, but utterly wrong, to believe that if xk = y k for some k, then x k m = y k m .
The formula
f
(m) = (f 1 )k ((m))
xk
(see 9.5.10 and 9.5.11) should make it clear that x k depends on all the components of the chart
and not just on the single component xk . In any case, here is a concrete counterexample.
9.5.12. Example. Consider R2 as a 2-manifold (with the usual differential structure generated by
the atlas whose only member is the identity map on R2 ). Let = (x1 , x2 ) be the identity map on
R2 and = (y 1 , y 2 ) be the map defined by : R2 R2 : (u, v) 7 (u, u + v). Clearly and are
charts containing the point m = (1, 1) and x1 = y 1 . We see, however, that x 1 6= y 1 by computing
f
(1, 1)
x1
and
f
(1, 1)
y 1
for the function f : R2 R : (u, v) 7 u2 v.
9.5.13. Proposition (Change of Variables Formula). Let = (x1 , . . . , xn ) and = (y 1 , . . . , y n )

be charts at a point m on an n-manifold. Then

n
X
y j

=
(m) j
(9.5)
y m
xk m
xk
j=1
for 1 k n.
9.5.14. Remark. If = (x1 , . . . , xn ) and = (y 1 , . . . , y n ) are charts on an n-manifold which
overlap (that is, dom dom 6= ), then the preceding change of variables formula holds for
all m in dom dom 6= . Thus in the interests of economy of notation, symbols indicating
evaluation at m are normally omitted. Then (9.5) becomes
n
X
y j
=
(9.6)
xk
xk y j
j=1
92
with the understanding that it may be applied to any real valued function which is smooth
on dom dom .
9.5.15. Exercise. Regard the 2-sphere in R3 as a 2-manifold whose differentiable structure is is
generated by the stereographic projections from the north and south poles (see example 9.2.12).
Let be the stereographic projection from the south pole. That is,

x
y
(x, y, z) =
,
1+z 1+z
for all (x, y, z) S2 such that z 6= 1. Let be defined by
(x, y, z) = (y, z)
for all (x, y, z)
function f by
S2
such that x > 0. (It is easy to see that is a chart.) Also define a real valued
x y
+
y
z

1 1 1
2
for all (x, y, z) S such that x, y, z > 0. Let m = 2 , 2 , 2 .
For these data verify by explicit computation formula (9.5). That is, by computing both sides
show that for k = 1 and k = 2 the formula
f
y 1
f
y 2
f
(m)
=
(m)
(m)
+
(m) 2 (m)
1
k
k
k
y
y
x
x
x
is correct for the functions , , and f and the point m given in the preceding paragraph.
f (x, y, z) =
9.5.16. Proposition (Another version of the chain rule). Let G : M N be a smooth map between
manifolds of dimensions n and p, respectively, = (x1 , . . . , xn ) be a chart containing the point m
in M , and = (y 1 , . . . , y p ) be a chart containing G(m) in N . Then
p
X (y j G)
(f G)
f
(m)
=
(m) j (G(m))
k
k
y
x
x
(9.7)
j=1
whenever f CG(m)
and 1 k n.
9.5.17. Remark. If one is willing to adopt a sufficiently relaxed attitude towards notation many
complicated looking formulas can be put in simpler form. Convince yourself that it is not beyond
the realm of possibility for one to encounter equation (9.7) written in the form
p
X z y j
z
=
.
y j xk
xk
j=1
9.5.18. Exercise. Consider the 2-manifold S2 with the differential structure generated by the
stereographic projections form the north and south poles (see example 9.2.12) and the 2-manifold
P2 with the differentiable structure generated by the atlas given in example 9.2.14. Recall from
example 9.3.5 that the map F : S2 P2 : (x, y, z) 7 [(x, y, z)] is smooth. Define
x+y
h [(x, y, z)] =
2z

1 1 1
2
whenever [(x, y, z)] P and z 6= 0; and let m = 2 , 2 , 2 . It is clear that h is well-defined.
(You may assume it is smooth in a neighborhood of F (m).) Let = (u, v) be the stereographic
projection of S2 from the south pole (see 9.2.12).

(a) Using the definitions of and compute (hF ) (m) and (hF ) (m).
u m
v m
(b) Let be the chart in P2 defined in example 9.2.14. Use this chart, which contains F (m),
and the version of the chain rule given in proposition 9.5.16 to compute (independently
)
(hF )
of part (a) ) (hF
u (m) and
v (m).
93
Let m be a point in an n-manifold. In example 9.5.7 we defined, for each e

c in the geometric
e
tangent space Tm , a function
vec : Gm R : fb 7 D(f c)(0)
and showed that vec is (well-defined and) a derivation on the space Gm of germs at m. Thus the
map v : e
c 7 vec takes members of Tem to members of Tbm . The next few propositions lead to the
conclusion that v : Tem Tbm is a vector space isomorphism and that consequently the two definitions
of tangent space are essentially the same. Subsequently we will drop the diacritical marks tilde
and circumflex that we have used to distinguish the geometric and algebraic tangent spaces and
instead write just Tm for the tangent space at m. We will allow context to dictate whether a
tangent vector (that is, a member of the tangent space) is to be interpreted as an equivalence class
of curves or as a derivation. Our first step is to show that the map v is linear.
9.5.19. Proposition. Let m be a point in an n-manifold and
v : Tem Tbm : e
c 7 vec
be the map defined in example 9.5.7. Then
(i) veb+ec = veb + vec and
(ii) vec = vec
for all eb, e
c Tem and R.
9.5.20. Proposition. Let m be a point in an n-manifold. Then the map
v : Tem Tbm : e
c 7 vec
(defined in 9.5.7) is injective.
e
b
In order to show that
the map v : Tm Tm is surjective we will need to know that the tangent

vectors x1 m , . . . , xn m span the tangent space Tbm . The crucial step in this argument depends
on adapting the (second order) Taylors formula so that it holds on finite dimensional manifolds.
9.5.21. Lemma. Let = (x1 , . . . , xn ) be a chart containing a point m in an n-manifold and f be a
. Then there exist a neighborhood U of m and smooth functions sjk (for 1 j, k n)
member of Cm
such that
n
n
X
X

f = f (m) +
(xj aj ) f j (a) +
(xj aj )(xk ak )sjk
j=1
j,k=1
where a = (m).
Hint for proof . Apply Taylors formula to the local representative f .
1
n
9.5.22. Proposition.
If = (x

, . . . , x ) is a chart containing a point m in an n-manifold, then

the derivations x1 m , . . . , xn m span the algebraic tangent space Tbm . In fact, if w is an arbitrary
element of Tbm , then

n
X

b
j
.
w=
w x
xj m
j=1
9.5.23. Proposition. Let m be a point in an n-manifold. Then the map

v : Tem Tbm : e
c 7 vec
(defined in 9.5.7) is surjective.
9.5.24. Corollary. Let m be a point in an n-manifold. Then the map v (of the preceding proposition) is an isomorphism between the tangent spaces Tem and Tbm .
1
n
9.5.25. Corollary.
If = (x

, . . . , x ) is a chart containing a point m in an n-manifold, then the

derivations x1 m , . . . , xn m constitute a basis for the tangent space Tbm .
94
e m of a smooth map F : M N between finite dimensional

In 9.4.16 we defined the differential dF
manifolds at a point n M . This differential between the geometric tangent spaces at m and F (m)
turned out to be a linear map (see 9.4.17). In a similar fashion F induces a linear map, which we
b m , between the algebraic tangent spaces at m and F (m). We define this new differential
denote by dF
and then show that it is essentially same as the one between the corresponding geometric tangent
spaces.
9.5.26. Definition. Let F : M N be a smooth map between finite dimensional manifolds,

b m : Tbm TbF (m) by setting dF
b m (w) (b
m M , and w Tbm . Define dF
g ) = w(g[
F )or in
somewhat less cluttered notation
b m (w)(g) = w(g F )
dF
for each g CF(m) .
b m (w), defined above, is well-defined and is a derivation
9.5.27. Proposition. The function dF
on GF (m) .
Now we show that this new differential is essentially the same as the one defined in 9.4.16.
9.5.28. Proposition. Let F : M N be a smooth mapping between finite dimensional manifolds
and m M . Then the following diagram commutes.
Tem
e m
dF
/ TeF (m)
Tbm
b m
dF
/ TbF (m)
In light of the isomorphism between the geometric and algebraic tangent spaces to a manifold
(see 9.5.24) and the equivalence of the respective differential maps (proved in the preceding proposition), we will for the most part write just Tm for either type of tangent space and dFm for either
differential of a smooth map. In situations where the difference is important context should make
it clear which one is intended.
9.5.29. Corollary. If F : M N is a smooth map between finite dimensional manifolds and
b m is a linear transformation from the algebraic tangent space Tbm into TbF (m) .
m M , then dF
9.5.30. Proposition (Yet another chain rule). If F : M N and G : N P are smooth maps
between finite dimensional manifolds and m M , then
b F )m = dG
b F (m) dF
b m.
d(G
In the next exercise we consider the tangent space Ta at a point a in R. Here, as usual,
the differential structure on R is taken to be the one generated by the atlas whose only chart is
the identity
map I on R. The tangent space is one-dimensional; it is generated by the tangent

. This particular
notation
is, as far as I know,
vector I
a

never used; some alternative standard

d d
d
d
df
notations are
, , and
. And, of course,
(f ) is written as
(a).

dx a dt a
dI a
dx a
dx
dg
(a).
dx
9.5.32. Exercise. If f Ca , where m is a point in some n-manifold, and w Tm , then dfm (w)
belongs to Tf (m)
. Since the tangent space at f (m) is one-dimensional, there exists R such that
d
dfm (w) =
. Show that = w(f ).
dI
9.5.31. Exercise. Let a R and g Ca . Find
f (m)
95
9.5.33. Convention. Let a be a point

in R. It is natural in the interests of simplifying notation
d
to make use of the isomorphism 7 between the tangent space Ta and R to identify these
dI a
where m is a point in an n-manifold and if we regard T
one-dimensional spaces. If f Cm
f (m) = R,
then corollary 9.5.29 says that dfm is a linear map from Tm into R; that is, dfm belongs to the dual
space of Tm ; that is
dfm Tm .
(9.8)

d
we conclude from exercise 9.5.32 that
Furthermore, under the identification =
dI
f (m)
dfm (w) = w(f ) .
(9.9)
for every w Tm . From now on we adopt (9.8) and (9.9) even though they are, strictly speaking,
abuses of notation. They should cause little confusion and are of considerable help in reducing
notational clutter.
9.5.34. Definition. If m is a point in an n-manifold, the dual space Tm of the tangent space at
m is called the cotangent space at m.
Notice that in (9.8) we have adopted the convention that at every point m the differential of a
smooth real valued function f belongs to the cotangent space at m. In particular, if = (x1 , . . . , xn )
is a chart on an n-manifold, then each of its components xk is a smooth real valued function and
therefore dxk m belongs to the cotangent space Tm for every m in the domain of . The next
proposition shows that in fact the set {dx1 m , . . . , dxn m } of cotangent vectors is the basis for Tm
dual to the basis for Tm given in corollary 9.5.25.
9.5.35. Proposition. Let = (x1 , . . . , xn ) be a chart containing the point m in an
n-manifold.

1
n
Then {dx m , . . . , dx m } is a basis for Tm ; it is dual to the basis x1 m , . . . , xn m for Tm .
, and = (x1 , . . . , xn ) be a chart
9.5.36. Proposition. Let m be a point in an n-manifold, f Cm
containing m. Then
n
X
f
dfm =
(m) dxk m .
xk
k=1
P
Hint for proof . There exist scalars 1 , . . . n such that dfm = nj=1 j dxj m . (Why?) Consider

Pn

j
j=1 j dx m xk m .
Notice that this proposition provides some meaning (and justification) for the conventional
formula frequently trotted out in beginning calculus courses.
f
f
f
df =
dx +
dy +
dz .
x
y
z
CHAPTER 10
DIFFERENTIAL FORMS ON MANIFOLDS

In everything that follows all vector spaces are
assumed to be real, finite dimensional, and oriented; and all manifolds are smooth oriented differentiable manifolds.
10.1. Vector Fields
10.1.1. Definition. Let M be a manifold. The set
[
T M :=
Tm
mM
is the tangent bundle of M . The tangent bundle projection is the map M : T M M

such that M (w) = m whenever w Tm .
10.1.2. Definition. A vector field on a manifold M is a map v : M T M such that M v =
IdM (the identity map on M ). It is an equivalent formulation to say that v is a vector field if it is
a right inverse of the tangent bundle projection or to say that it is a section of the tangent bundle.
10.1.3. Notation. For a manifold M denote by C (M ) the family of smooth real valued functions
(M, R) for every
on M . That is, a function f belongs to C (M ) provided that it belongs to Cm
m M (see definition 9.5.1).
10.1.4. Definition. If v is a vector field on a manifold M and f C (M ), define the function
vf : M R by
(vf )(m) = v(m)(fb) .
We will say that v is a smooth vector field if vf is smooth for every f C (M ).
10.1.5. Exercise. Let M be an n-manifold and = (x1 , . . . , xn ) : U Rn be a chart. Regard U
as a submanifold of M in the obvious fashion. For 1 k n take x k to be the vector field on U
defined by

: m 7
xk
xk m
where, as before (see 9.5.10 and 9.5.11),

b
1
f
=
f
((m))
k
xk
m
for all f C (U ). Show that the vector field
xk
is smooth.
10.1.6. Exercise. Notation as in the preceding exercise. Let v be a smooth vector field on U . For
every m U there exist numbers 1 (m), . . . , n (m) such that

n
X

v(m) =
k (m) k
x m
k=1
97
98
(see 9.5.25). Thus we may write

v=
n
X
k=1
xk
where k is the function m 7 k (m). Show that each k is a smooth function.

10.2. Differential 1-forms
10.2.1. Definition. Let M be a manifold. The set
[
T M :=
Tm
mM
is the cotangent bundle of M . The cotangent bundle projection is the map M : T M M

such that M (u) = m whenever u Tm .
10.2.2. Definition. A differential one-form (or differential 1-form) on a manifold M is
a map : M T M such that M = IdM (the identity map on M ). Equivalently, it is a right
inverse of the cotangent bundle projection, or it is a section of the cotangent bundle.
10.2.3. Definition. If is a differential one-form on an n-manifold M and v is a vector field
on M , we define

(v) : M R : m 7 (m) v(m) .
10.2.4. Definition. A differential one-form is smooth if (v) C (M ) whenever v is a smooth
vector field on M .
10.2.5. Proposition. Let M be an n-manifold and f C (M ). Then the function df : m 7 dfm
is a smooth differential one-form on M .
10.2.6. Convention. From now on lets drop the words smooth and differential in the phrase
smooth differential one-form. There is no other kind of one-form that we will be interested in.
10.3. Differential k-forms
10.3.1. Notation. For an n-manifold M and k 0 let
^k
[ ^k

(M ) =
(Tm ) : m M
and
^
[ ^

(M ) =
(Tm ) : m M .
V
10.3.2. Definition.
A differential form is a section of (M ). Thus is a differential form if
V
(m) (Tm ) for every m M . Similarly, is a differential k-form (or just a k-form) if
V
it a section of k (M ). Notice that for 1-forms this definition agrees with the one given in 10.2.2
V
since 1 (M ) = T M . Also notice that a 0-form is just a real valued function on M (because of our
V
identification of 0 (Tm ) with Rsee 8.1.15).
10.3.3. Exercise. Let M be a 3-manifold and = (x, y, z) : U R3 be a chart. Regard U as a
V
V
V
V
V
submanifold of M . Exhibit bases for 0 (U ), 1 (U ), 2 (U ), 3 (U ), and (U ).
10.3.4. Definition. Given a chart = (x1 , . . . , xn ) : U Rn on an n-manifold M , we may express
a k-form locally (that is, on U ) by
X
(m) =
aj1 ...jk (m) dxj1 m dxjk m
j1 <<jk
for all m U . More succinctly,

=
X
j1 <<jk
aj1 ...jk dxj1 dxjk .
10.3. DIFFERENTIAL k-FORMS
99
It should be kept in mind that the coefficients aj1 ...jk in this expression are functions and that
they depend on the choice of coordinate system (chart). The k-form is smooth with respect to
the chart if all the coefficients aj1 ...jk are smooth real valued functions on U . A k-form defined
on all of M is smooth if it is smooth with respect to every chart on M . A differential form is
V
smooth if its component in k(M ) is smooth for every k. The set of smooth differential
forms on

V
Vk
M is denoted by C M, (M ) and the set of smooth k-forms by C M, (M ) .

10.3.5. Convention. In the sequel all k-forms are smooth differential k-forms and all differential
forms are smooth.
The next theorem defines a mapping d on differential forms called the exterior differentiation operator.
10.3.6. Theorem. If M is an n-manifold, then there exists a unique linear map
^
^

d : C M, (M ) C M, (M )
which satisfies

V
V
(1) d C M, k (M ) C M, k+1 (M ) ;
(2) d(f ) = df (the ordinary differential of f ) for every 0-form f ;
(3) if is a k-form and is any differential form, then
d( ) = (d) + (1)k d;
and
(4) d2 = 0.
Proof. Proofs of the existence and uniqueness of such a function can be found in [21] (theorem
12.14), [32] (chapter 1, theorem 11.1), and [3] (section 4.6).

10.3.7. Exercise. Let M be a 3-manifold, = (x, y, z) : U R3 be a chart on M , and f : U R
be a 0-form on U . Compute d(f dy). (If f is a 0-form and is any differential form, it is conventional
to write f for f .)
10.3.8. Example. Let M be a 3-manifold and = (x, y, z) : U R3 be a chart on M . Then
d cos(xy 2 ) dx dz = 2xy sin(xy 2 ) dx dy dz.
10.3.9. Exercise. Let M be a 3-manifold and = (x, y, z) : U R3 be a chart on M . Compute
d(x dy dz + y dz dx + z dx dy).
10.3.10. Exercise. Let M be a 3-manifold and = (x, y, z) : U R3 be a chart on M . Compute
d[(3xz dx + xy 2 dy) (x2 y dx 6xy dz)].
10.3.11. Exercise. In beginningRRcalculus texts some curious arguments are given for replacing the
expression dx dy in the integral R f dx dy by r dr d when we change from rectangular to polar
coordinates in the plane. Show that if we interpret dx dy as the differential form dx dy, then this
is a correct substitution. (Assume additionally that R is a region in the open first quadrant and
that the integral of f over R exists.)
10.3.12. Exercise. Give an explanation similar to the one in the preceding exercise of the change
in triple integrals from rectangular to spherical coordinates.
10.3.13. Exercise. Generalize the two preceding exercises.
10.3.14. Proposition. If f is a 0-form and is a k-form on U , then (f ) = f ( ).
10.3.15. Proposition. If and are k-forms on U , then ( + ) = + .
10.3.16. Proposition. If is a k-form on U , then = (1)k(nk) .
Notice that, in consequence of the preceding proposition, every k-form on a 3-manifold satisfies
= .
100
10.3.17. Exercise. For real valued functions a, b, and c on U compute

(1) (a1),
(2) (a dx + b dy + c dz),
(3) (a dy dz + b dz dx + c dx dy), and
(4) a(dx dy dz).
10.4. Some Classical Vector Analysis
10.4.1. Definition. In beginning calculus we learn that the gradient of a smooth scalar field f on
f
f
Rn can be represented at a point m as the vector x
1 (m), . . . , xn (m) . For a smooth function f
on the domain U of a chart = (x1 , . . . , xn ) in an n-manifold we define the gradient of f at a
point m in U to be the vector in the cotangent space Tm whose components with respect to the
f
f
usual basis {dx1 m , . . . , dxn m } for Tm are just x
1 , . . . , xn . We denote this vector by grad f (m)
or f (m). Thus we make no distinction between the 1-forms grad f and df , since
grad f =
f
f
dx1 + + n dxn = df
1
x
x
(see proposition 9.5.36).

10.4.2. Definition. Let be a 1-form on the domain of a chart on a manifold. The curl of ,
denoted by curl or is defined by
curl = d.
(Notice that on a 3-manifold the curl of a 1-form is again a 1-form.)
10.4.3. Example. If = a dx + b dy + c dz is a 1-form on a 3-manifold, then

c
b
a
c
b
a
curl =
dx +
dy +
dz.
y z
z x
x y
10.4.4. Remark. Some depraved souls who completely abandon all inhibitions concerning notation
have been known to write
dx dy dz
curl = det x
y
z .
a
b
c
10.4.5. Definition. Let be a 1-form on the domain of a chart on a manifold. The divergence
of , denoted by div or is defined by
div = d .
(Notice that on a 3-manifold the divergence of a 1-form is a 0-form; that is, a real valued function.)
10.4.6. Example. If = a dx + b dy + c dz is a 1-form on a 3-manifold, then
div =
a
b
c
+
+
.
x y z
10.4.7. Exercise. If f is a 0-form on the domain of a chart, prove (without using partial derivatives)
that curl grad f = 0.
10.4.8. Exercise. If is a 1-form on the domain of a chart, prove (without using partial derivatives) that div curl = 0.
10.4.9. Definition. Let and be 1-forms on the domain of a chart. Define the cross-product
of and , denoted by , by
= ( ).
LEMMA
10.6. POINCARES
101
10.4.10. Example. If = a dx+b dy +c dz and = e dx+f dy +g dz are 1-forms on a 3-manifold,

then
= (by cf ) dx + (ce ay) dy + (af be) dz.
dx dy dz
b
c .
10.4.11. Remark. Occasionally as a memory aid some people write = det a
e f
g
10.4.12. Exercise. Suppose we wish to define the dot product h, i of two 1-forms = a dx +
b dy + c dz and = e dx + f dy + g dz on a 3-manifold to be the 0-form ae + bf + cg. Rephrase this
definition without mentioning the components of and .
10.4.13. Exercise. Suppose we wish to define the triple scalar product [, , ] of the 1-forms
= a dx + b dy + c dz, = e dx + f dy + g dz, and = j dx + k dy + l dz on a 3-manifold to be
the 0-form bgj f cj + cek agk + af l bel. Rephrase this definition without mentioning the
components of , , and .
10.5. Closed and Exact Forms
10.5.1. Definition. Let U be an open subset of a manifold and
^k1
dk1 ^k
dk ^k+1
/
/
/
(U )
(U )
(U )
where dk1 and dk are exterior differentiation operators. Elements of ker dk are called closed
k-forms and elements of ran dk1 are exact k-forms. In other words, a k-form is closed if
d = 0. It is exact if there exists a (k 1)-form such that = d.
10.5.2. Proposition. Every exact differential form is closed.
10.5.3. Proposition. If and are closed differential forms, so is .
10.5.4. Proposition. If is an exact form and is a closed form, then is exact.
10.5.5. Example. Let = (x, y, z) : U R3 be a chart on a 3-manifold and = a dx + b dy + c dz
c
b a
c
b
be a 1-form on U . If is exact, then y
= z
, z = x
, and x
= a
y .
10.5.6. Exercise. Determine if each of the following 1-forms is exact in R2 . If it is, specify the
0-form of which it is the differential.
(1) yexy dx + xexy dy;
(2)
x sin y dx + x cos y dy;and

arctan y
x
arcsin x
x2
2
y
2 + e dy.
(3)
+ + 3x dx +
y
1 + y2
2y
1 x2
10.5.7. Exercise. Explain why solving the initial value problem
ex cos y + 2x ex (sin y)y 0 = 0,
y(0) = /3
is essentially the same thing as showing that the 1-form (ex cos y + 2x) dx ex (sin y) dy is exact.
Do it.
10.6. Poincar
es Lemma
10.6.1. Notation (for the entire section). Let n N, U be a nonempty open convex subset of Rn ,
and x = (x1 , . . . , xn ) be the identity map on U .
Whenever 1 k n and = b dxi1 . . . , dxik is a k-form on U , we define a 0-form g on U
by
Z 1
g (x) :=
b(tx)tk1 dt if 6= 0
and
g 0 (x) = 0 ;
0
102
and we define (k 1)-forms and h() by

:=
k
X
d
ij dxik
(1)j1 xij dxi1 dx
j=1
and
h() := g .
In the definition of , the circumflex above the term dxij indicates that the term is deleted. For
example, if k = 3, then
= xi1 dxi2 dxi3 xi2 dxi1 dxi3 + xi3 dxi1 dxi2 .
V
For each k extend h to all of k (U ) by requiring it to be linear. Thus
^k
^k1
(U ) .
(U )
h
10.6.2. Theorem (Poincares lemma). If U is a nonempty open convex subset of Rn and p 1,
then every closed p-form on U is exact.
Hint for proof. Let p be a fixed integer such that 1 p n. Let i1 , . . . , ip be distinct
integers between 1 and n, and let a be a 0-form on U . Define
:= dxi1 dxip ,
:= a,
and, for 1 k n, define
k := ak dxk .
a
(Here, ak = x
k .)
Now, do the following.
R1
(a) Show that g k (x) = 0 ak (tx)tp dt for 1 k n and x U .
k
(b) Show that = xk dxk for 1 k n.
(c) Compute h k for 1 k n.
P
(d) Show that d = nk=1 k .
(e) Compute h d.
(f) Show that d = p .
(g) Compute d(h).

d p
(h) Compute
t a(tx) .
dt
P
(i) Show that pg + nk=1 g k xk = a.
(j) Show that (dh + hd)() = .
10.6.3. Proposition. If is a 1-form on a nonempty convex open subset U of R3 with curl = 0,

then there exists a 0-form f on U such that = grad f .
10.6.4. Exercise. Use the proof of Poincares lemma to find a 0-form on R3 whose gradient is
(2xyz 3 y 2 z) dx + (x2 z 3 2xyz) dy + (3x2 yz 2 xy 2 ) dz.
10.6.5. Exercise. Consider the 1-form = ez dx + x dy in R3 and let = d. Use the proof of
Poincares lemma to find another 1-form = a dx + b dy + c dz such that = d. Explain carefully
a
what happens at z = 0. Find
(0, 0, 0) for every integer n 0.
z n
10.6.6. Proposition. If is a 1-form on a nonempty convex open subset U of R3 with div = 0,
then there exists a 1-form on U such that = curl .
10.6.7. Exercise. Let = 2xyz dx + x3 z 2 dy yz 2 dz. Check that div = 0. Use the proof of
Poincares lemma to find a 1-form whose curl is .
LEMMA
10.6. POINCARES
103
10.6.8. Proposition. Every smooth real valued function f on a nonempty convex open subset U
of R3 is div for some 1-form on U .
10.6.9. Exercise. In the proof of the preceding proposition, what needs to be changed if U lies
in R2 ?
10.6.10. Exercise. Use the proof of Poincares lemma to find a 1-form on R3 whose whose
divergence is the function
f : (x, y, z) 7 xy y 2 z + xz 3 .
CHAPTER 11
HOMOLOGY AND COHOMOLOGY

11.1. The de Rham Cohomology Group
11.1.1. Definition. Let M be an n-manifold. We denote by Z k (M ) (or just Z k ) the vector space
of all closed k-forms on M . The Z is for the German word Zyklus, which means cycle. So in
cohomological language closed forms are often called cocycles.
Also we denote by B k (M ) (or just B k ) the vector space of exact k-forms on M . Since there are
no differential forms of degree strictly less than 0, we take B 0 = B 0 (M ) = {0}. For convenience
we also take Z k = {0} and B k = {0} whenever k < 0 or k > n. The letter B refers to the word
boundary. So exact forms in the context of cohomology are often called coboundaries.
It is a trivial consequence of proposition 10.5.2 that B k (M ) is a vector subspace of Z k (M ).
Thus it makes sense to define
Z k (M )
.
H k = H k (M ) := k
B (M )
The quotient space H k (M ) is the k th de Rham cohomology group of M . (Yes, even though it
is a vector space, it is traditionally called a group.) The dimension of the vector space H k (M ) is
the k th Betti number of the manifold M .
Another (obviously equivalent) way of phrasing the definition of the k th de Rham cohomology
group is in terms of the maps
^k1
dk1 ^k
dk ^k+1
/
/
/
/
(M )
(M )
(M )
where dk1 and dk are exterior differentiation operators. Define
ker dk
H k (M ) :=
ran dk1
for all k.
It is an interesting fact, but one that we shall not prove, that these cohomology groups are
topological invariants. That is, if two manifolds M and N are homeomorphic, then H k (M ) and
H k (N ) are isomorphic.
11.1.2. Example. If M is a connected manifold, then H 0 (M )
= R.
11.1.3. Exercise. For U an open subset of Rn give a very clear description of H 0 (U ) and explain
why its dimension is the number of connected components of U . Hint. A function is said to be
locally constant if it is constant in some neighborhood of each point in its domain.
11.1.4. Definition. Let F : M N be a smooth function between smooth manifolds. For k 1
define
^k
^k
^k
^k
F:
(N )
(M ) : 7 ( F )()
where
^k

( F )() m (v 1 , . . . , v k ) = F (m) dFm (v 1 ), . . . , dFm (v k )
(11.1)

V
for every m M and v 1 , . . . v k Tm . Also define ( 0 F )() m = F (m) . We simplify the notation
in (11.1) slightly
^k
( F )()(v 1 , . . . , v k ) = (dF (v 1 ), . . . , dF (v k )).
(11.2)
105
106
Denote by F theVmap induced by the maps

Z-graded algebra (M ).
Vk
F which takes the Z-graded algebra
(N ) to the
V
V
11.1.5. Example. For each k Z+ the pair of maps M 7 k (M ) and F 7 k (F ) (as defined
in 10.3.1 and 11.1.4) is a contravariant functor from the category of smooth manifolds and smooth
maps to the category of vector spaces and linear maps.
V
11.1.6. Proposition. If F : M N is a smooth function between smooth manifolds, j (N ),
V
and k (N ), then
^k
^j
^j+k
F )( ) = ( F )() ( F )().
(
V
11.1.7.
Exercise. In example 11.1.5 you showed that k was a functor for each k. What about
V
itself? Is it a functor? Explain.
11.1.8. Proposition. If F : M N is a smooth function between smooth manifolds, then
d F = F d .
11.1.9. Exercise. Let V = {0} be the 0-dimensional Euclidean space. Compute the k th de Rham
cohomology group H k (V ) for all k Z.
11.1.10. Exercise. Compute H k (R) for all k Z.
11.1.11. Exercise. Let U be the union of m disjoint open intervals in R. Compute H k (U ) for all
k Z.
11.1.12. Exercise. Let U be an open subset of Rn . For [] H j (U ) and [] H k (U ) define
[][] = [ ] H j+k (U ).
Explain why proposition 10.5.3 is necessary for this definition to make sense. Prove also that
this definition does not depend on theLrepresentatives chosen from the equivalence classes. Show
that this definition makes H (U ) =
H k (U ) into a Z-graded algebra. This is the de Rham
kZ
cohomology algebra of U .
11.1.13. Definition. Let F : M N be a smooth function between smooth manifolds. For each
integer k define
^k
H k (F ) : H k (N ) H k (M )k : [] 7 [ (F )()].
Denote by H (F ) the induced map which takes the Z-graded algebra H (N ) into H (M ).
11.1.14. Example. With the definitions given in 11.1.12 and 11.1.13 H becomes a contravariant
functor from the category of open subsets of Rn and smooth maps to the category of Z-graded
algebras and their homomorphisms.
11.2. Cochain Complexes
11.2.1. Definition. A sequence
/ Vk1
dk1
/ Vk
dk
/ Vk+1
of vector spaces and linear maps is a cochain complex if dk dk1 = 0 for all k Z. Such a
sequence may be denoted by (V , d) or just by V .
11.2.2. Definition. We generalize definition 11.1.1 in the obvious fashion. If V is a cochain
complex, then the k th cohomology group H k (V ) is defined to be ker dk / ran dk1 . (As before,
this group is actually a vector space.) In this context the elements of Vk are often called kcochains, elements of ker dk are k-cocycles, elements of ran dk1 are k-coboundaries, and d
is the coboundary operator.
11.3. SIMPLICIAL HOMOLOGY
107
11.2.3. Definition. Let (V , d) and (W , ) be cochain complexes. A cochain map G : V W

is a sequence of linear maps Gk : Vk Wk satisfying
k Gk = Gk+1 dk
for every k Z. That is, the diagram
dk
/ Vk
...
Gk
/ Wk
...
/ Vk+1

/ ...
Gk+1
/ Wk+1
/ ...
commutes.
11.2.4. Proposition. Let G : V W be a cochain map between cochain complexes. For each
k Z define
Gk : H k (V ) H k (W ) : [v] 7 [Gk (v)]
whenever v is a cocycle in Vk . Then the maps Gk are well defined and linear.
Hint for proof . To prove that Gk is well-defined we need to show two things: that Gk (v) is a
cocycle in Wk and that the definition does not depend on the choice of representative v.
11.2.5. Definition. A sequence
/ U
/V
/ W
/0
of cochain complexes and cochain maps is (short) exact if for every k Z the sequence
Fk
/ Uk
Gk
/ Vk
/ Wk
/0
of vector spaces and linear maps is (short) exact.

11.2.6. Proposition. If 0
complexes, then
/ U
/ V
Fk
H k (U )
is exact at H k (V ) for every k Z.
/ W
/ H k (V )
/ 0 is a short exact sequence of cochain

Gk
/ H k (W )
11.2.7. Proposition. A short exact sequence

0
/ U
/V
/ W
/0
Gk
/ H k (W )
of cochain complexes induces a long exact sequence

/ H k1 (W )
k1
/ H k (U )
Fk
/ H k (V )
/ H k+1 (U )
Hint for proof . If w is a cocycle in Wk , then, since Gk is surjective, there exists v Vk such
that w = Gk (v). It follows that dv ker Gk+1 = ran Fk+1 so that dv = Fk+1 (u) for some u Uk+1 .
Let k ([w]) = [u].
11.3. Simplicial Homology
11.3.1. Definition. Let V be a vector space. Recall
that a linear combination of a finite set
P
{x1 , . . . , xn } of vectors in V is a vector of the form nk=1 k xk where 1 , . . . , n R. If 1 = 2 =
= n = 0, then the linear combination is trivial ; if at
Pleast one k is different from zero, the
linear combination is nontrivial. A linear combination nk=1 P
k xk of the vectors x1 , . . . , xn is a
convex combination if k 0 for each k (1 k n) and if nk=1 k = 1.
11.3.2. Definition. If a and b are vectors in the vector space V , then the closed segment
between a and b, denoted by [a, b], is {(1 t)a + tb : 0 t 1}.
108
11.3.3. CAUTION. Notice that there is a slight conflict between this notation, when applied to
the vector space R of real numbers, and the usual notation for closed intervals on the real line. In
R the closed segment [a, b] is the same as the closed interval [a, b] provided that a b. If a > b,
however, the closed segment [a, b] is the same as the segment [b, a], it contains all numbers c such
that b c a, whereas the closed interval [a, b] is empty.
11.3.4. Definition. A subset C of a vector space V is convex if the closed segment [a, b] is
contained in C whenever a, b C.
11.3.5. Definition. Let A be a subset of a vector space V . The convex hull of A is the smallest
convex subset of V which contain A.
11.3.6. Exercise. Show that definition 11.3.5 makes sense by showing that the intersection of
a family of convex subsets of a vector space is itself convex. Then show that a constructive
characterization is equivalent; that is, prove that the convex hull of A is the set of all convex
combinations of elements of A.
11.3.7. Definition. A set S = {v0 , v1 , . . . , vp } of p + 1 vectors in a vector space V is convex
independent if the set {v1 v0 , v2 v0 , . . . , vp v0 } is linearly independent in V .
11.3.8. Definition. An affine subspace of a vector space V is any translate of a linear subspace
of V .
11.3.9. Example. The line whose equation is y = 2x 5 is not a linear subspace of R2 . But it is
an affine subspace: it is the line determined by the equation y = 2x (which is a linear subspace of
R2 ) translated downwards parallel to the y-axis by 5 units.
11.3.10. Definition. Let p Z+ . The closed convex hull of a convex independent set S =
{v0 , . . . , vp } of p + 1 vectors in some vector space is a closed p -simplex. It is denoted by [s] or
by [v0 . . . . , vp ]. The integer p is the dimension of P
the simplex. The open p -simplex determined
by the set S is the set of all convex combinations pk=0 k vk of elements of S where each k > 0.
The open simplex will be denoted by (s) or by (v0 , . . . , vp ). We make the special convention that
a single vector {v} is both a closed and an open 0 -simplex.
If [s] is a simplex in Rn then the plane of [s] is the affine subspace of Rn having the least
dimension which contains [s]. It turns out that the open simplex (s) is the interior of [s] in the
plane of [s].
11.3.11. Definition. Let [s] = [v0 , . . . , vp ] be a closed p -simplex in Rn and {j0 , . . . , jq } be a
nonempty subset of {0, 1, . . . , p}. Then the closed q -simplex [t] = [vj0 , . . . , vjq ] is a closed q-face
of [s]. The corresponding open simplex (t) is an open q-face of [s]. The 0 -faces of a simplex are
called the vertices of the simplex.
Note that distinct open faces of a closed simplex [s] are disjoint and that the union of all the
open faces of [s] is [s] itself.
11.3.12. Definition. Let [s] = [v0 , . . . , vp ] be a closed p -simplex in Rn . We say that two orderings
(vi0 , . . . , vip ) and (vj0 , . . . , vjp ) of the vertices are equivalent if (j0 , . . . , jp ) is an even permutation
of (i0 , . . . , ip ). (This is an equivalence relation.) For p 1 there are exactly two equivalence
classes; these are the orientations of [s]. An oriented simplex is a simplex together with one
of these orientations. The oriented simplex determined by the ordering (v0 , . . . , vp ) will be denoted
by hv0 , . . . , vp i. If, as above, [s] is written as [v0 , . . . , vp ], then we may shorten hv0 , . . . , vp i to hsi.
Of course, none of the preceding makes sense for 0 -simplexes. We arbitrarily assign them two
orientations, which we denote by + and . Thus hsi and hsi have opposite orientations.
11.3.13. Definition. A finite collection K of open simplexes in Rn is a simplicial complex if
the following conditions are satisfied:
(1) if (s) K and (t) is an open face of [s], then (t) K; and
11.3. SIMPLICIAL HOMOLOGY
109
(2) if (s), (t) K and (s) 6= (t), then (s) (t) = .

The dimension of a simplicial complex K, denoted by dim K, is the maximum dimension of the
simplexes constituting K. If r dim K, then the r-skeleton of K, denoted by K r , is the set of
all open simplexes in K whose dimensions are no greater than r. The polyhedron, |K|, of the
complex K is the union of all the simplexes in K.
11.3.14. Definition. Let K be a simplicial complex in Rn . For 0 p dim K let Ap (K) (or just
Ap ) denote the free vector space generated by the set of all oriented p -simplexes belonging to K.
For 1 p dim K let Wp (K) (or just Wp ) be the subspace of Ap generated by all elements of the
form
hv0 , v1 , v2 , . . . , vp i + hv1 , v0 , v2 , . . . , vp i
and let Cp (K) (or just Cp ) be the resulting quotient space Ap /Wp .For p = 0 let Cp = Ap and for
p < 0 or p > dim K let Cp = {0}. The elements of Cp are the p -chains of K.
Notice that for any p we have
[hv0 , v1 , v2 , . . . , vp i] = [hv1 , v0 , v2 , . . . , vp i] .
To avoid cumbersome notation we will not distinguish between the p -chain [hv0 , v1 , v2 , . . . , vp i] and
its representative hv0 , v1 , v2 , . . . , vp i.
11.3.15. Definition. Let hsi = hv0 , v1 , . . . , vp+1 i be an oriented (p + 1) -simplex. We define the
boundary of hsi, denoted by hsi, by
p+1
X
hsi =
(1)k hv0 , . . . , vbk , . . . , vp+1 i .
k=0
(The caret above the vk indicates that that term is missing; so the boundary of a (p + 1) -simplex
is an alternating sum of p -simplexes.)
11.3.16. Definition. Let K be a simplicial complex in Rn . For 1 p dim K define
p = : Cp+1 (K) Cp (K) :
as follows. If
a(s)hsi is a p -chain in K, let

X
X
a(s)hsi =
a(s)hsi .
For all other p let p be the zero map. The maps p are called boundary maps. Notice that each
p is a linear map.
11.3.17. Proposition. If K is a simplicial complex in Rn , then 2 : Cp+1 (K) Cp1 (K) is
identically zero.
Hint for proof . It suffices to prove this for generators hv0 , . . . , vp+1 i.
11.3.18. Definition. Let K be a simplicial complex in Rn and 0 p dim K. Define Zp (K) = Zp
to be the kernel of p1 : Cp Cp1 and Bp (K) = Bp to be the range of p : Cp+1 Cp . The
members of Zp are p -cycles and the members of Bp are p -boundaries.
It is clear from proposition 11.3.17 that Bp is a subspace of the vector space Zp . Thus we may
define Hp (K) = Hp to be Zp /Bp . It is the pth simplicial homology group of K. (And, of
course, Zp , Bp , and Hp are the trivial vector space whenever p < 0 or p > dim K.)
11.3.19. Exercise. Let K be the topological boundary (that is, the 1 -skeleton) of an oriented
2 -simplex in R2 . Compute Cp (K), Zp (K), Bp (K), and Hp (K) for each p.
11.3.20. Exercise. What changes in exercise 11.3.19 if K is taken to be the oriented 2 -simplex
itself?
11.3.21. Exercise. Let K be the simplicial complex in R2 comprising two triangular regions
similarly oriented with a side in common. For all p compute Cp (K), Zp (K), Bp (K), and Hp (K).
110
11.3.22. Definition. Let K be a simplicial complex. The number p := dim Hp (K) is the pth
P K
p
Betti number of the complex K. And (K) := dim
p=0 (1) p is the Euler characteristic
of K.
11.3.23. Proposition. Let K be a simplicial complex. For 0 p dim K let p be the number
of p -simplexes in K. That is, p = dim Cp (K). Then
(K) =
dim
XK
(1)p p .
p=0
11.4. Simplicial Cohomology

11.4.1. Definition. Let K be a simplicial complex. For each p Z let C p (K) = Cp (K) . The
elements of C p (K) are (simplicial) p -cochains. Then the adjoint p of the boundary map
p : Cp+1 (K) Cp (K)
is the linear map
p = : C p (K) C p+1 (K) .
(Notice that = 0.)
Also define
(1) Z p (K) := ker p ;
(2) B p (K) := ran p1 ; and
(3) H p (K) := Z p (K)/B p (K).
Elements of Z p (K) are (simplicial) p -cocycles and elements of B p (K) are (simplicial) p coboundaries. The vector space H p (K) is the pth simplicial cohomology group of K.

11.4.2. Proposition. If K is a simplicial complex in Rn , then H p (K)
= Hp (K) for every
integer p.
11.4.3. Definition. Let F : N M be a smooth injection between smooth manifolds. The pair
(N, F ) is a smooth submanifold of M if dFn is injective for every n N .
11.4.4. Definition. Let M be a smooth manifold, K be a simplicial complex in Rn , and h : [K]
M be a homeomorphism. The triple (M, K, h) is a smoothly triangulated manifold if for
every open simplex (s) in K the map h : [s] M has an extension hs : U M to a neighborhood
[s]
U of [s] lying in the plane of [s] such that (U, hs ) is a smooth submanifold of M .
11.4.5. Theorem. A smooth manifold can be triangulated if and only if it is compact.
The proof of this theorem is tedious enough that very few textbook authors choose to include
it in their texts. You can find a simplified proof in [5].
11.4.6. Theorem (de Rhams theorem). If (M, K, ) is a smoothly triangulated manifold, then
H p (M )
= H p (K)
for every p Z.
Proof. See [17], chapter IV, theorem 3.1; [21], theorem 16.12; [30], pages 164173; and [33],
theorem 4.17.
11.4.7. Proposition (pullbacks of differential forms). Let F : M N be a Vsmooth mapping
V
between smooth manifolds. Then there exists an algebra homomorphism F : (N ) (M ),
called the pullback associated with F which satisfies the following conditions:
V
V
(1) F maps p (N ) into p (M ) for each p;
(2) F (g) = g F for each 0-form g on N ; and
11.4. SIMPLICIAL COHOMOLOGY
111
(3) (F )m (v) = F (m) (dFm (v)) for every 1-form on N , every m M , and every v Tm .
11.4.8. Proposition. If F : MV N is a smooth map between n-manifolds,
then F is a cochain
V
map from the cochain complex ( (N ), d ) to the cochain complex ( (M ), d ). That is, the diagram
Vp
V
d
/ p+1 (N )
(N )
Vp
(M )
commutes for every p Z.
Vp+1
(M )
CHAPTER 12
STOKES THEOREM
12.1. Integration of Differential Forms
12.1.1. Definition. Let hsi be an oriented p -simplex in Rn (where 1 p n) and be a p -form
defined on a set U which is open in the plane of hsi and which contains [s]. If hsi = hv0 , . . . , vp i
take (v1 v0 , . . . , vp v0 ) to be an ordered basis for the plane of hsi and P
let x1 , . . . , xp be the
coordinate projection functions relative to this ordered basis; that is, if a = pk=1 ak (vk v0 ) U ,
then xj (a) = aj for 1 j p. Then = (x1 , . . . , xp ) : U Rp is a chart on U ; so there exists a
smooth function g on U such that = g dx1 dxp . Define
Z
Z
= g dx1 . . . dxp
hsi
[s]
where the right hand side is an ordinary Riemann integral. If hv0 i is a 0 -simplex, we make a special
definition
Z
f = f (v0 )
hv0 i
for every 0 -form f .

Extend the preceding definition
to p -chains by requiring the integral to be linear as a function
P
of simplexes; that is, if c =
as hsi is a p -chain (in some simplicial complex) and is a p -form,
define
Z
Z
X
=
a(s) .
c
hsi
12.1.2. Definition. For a smoothly triangulated manifold (M, K, h) we define a map

Z ^
p
:
(M ) C p (K)
p
R
as follows. If is a p -form on M , then p is to be a linear functional on Cp (K); that is, a member

of C p (K) = Cp (K) . In order to define a linear functional on Cp (K) it suffices to specify its values
on the basis vectors of Cp (K); that is, on the oriented p -simplexes hsi which constitute Cp (K).
Let hs : U M be an extension of h to an open set U in the plane of hsi. Then hs pulls back
[s]
V
p -forms on M to p -forms on U so that hs () p (U ). Define
Z
Z
hsi := hs () .
p
hsi
12.1.3. Exercise. Let V be an open subset of Rn , F : V Rn , and c : [t0 , t1 ] V be a smooth

curve in V . Let C = ran c. It is Rconventional to define the integral of the tangential component
of F over C , often denoted by C FT , by the formula
Z
Z t1
Z t1
FT =
hF c, Dci =
hF (c(t)), c0 (t)i dt.
(12.1)
C
t0
t0
113
114
12. STOKES THEOREM
P
The tangential component of F , written FT may be regarded as the 1 -form nk=1 F k dxk .
Make sense of the preceding definition in terms of the definition of the integral of 1 -forms over
a smoothly triangulated manifold. For simplicity take n = 2. Hint. Suppose we have the following:
(1) ht0 , t1 i (with t0 < t1 ) is an oriented 1 -simplex in R;
(2) V is an open subset of R2 ;
(3) c : J V is an injective smooth curve in V , where J is an open interval containing [t0 , t1 ];
and
(4) = a dx + b dy is a smooth 1 -form on V .
First show that

c (dx) (t) = Dc1 (t)
for t0 t t1 . (We drop the notational distinction between c and its extension cs to J. Since
the tangent space Tt is one-dimensional for every t, we identify Tt with R. Choose v (in (3) of
proposition 11.4.7) to be the usual basis vector in R, the number 1.)
Show in a similar fashion that

c (dy) (t) = Dc2 (t) .

R
Then write an expression for c () (t). Finally conclude that 1 (ht0 , t1 i) is indeed equal to
R t1
t0 h(a, b) c, Dci as claimed in 12.1.
12.1.4. Exercise. Let S1 be the unit circle in R2 oriented counterclockwise and let F be the
3
3
3
3
vector field
R defined by F(x, y) = (2x y ) i + (x + y ) j. Use your work in exercise 12.1.3 to
calculate S1 FT . Hint. You may use without proof two facts: (1) the integral does not depend
on the parametrization (triangulation) of the curve, and (2) the results of exercise 12.1.3 hold also
for simple closed curves in R2 ; that is, for curves c : [t0 , t1 ] R2 which are injective on the open
interval (t0 , t1 ) but which satisfy c(t0 ) = c(t1 ).
12.1.5. Notation. Let Hn = {x Rn : xn 0}. This is the upper half-space of Rn .
12.1.6. Definition. A n -manifold with boundary is defined in the same way as an n -manifold
except that the range of a chart is assumed to be an open subset of Hn .
The interior of Hn , denoted by int Hn , is defined to be {x Rn : xn > 0}. (Notice that this
is the interior of Hn regarded as a subset of Rn not of Hn .) The boundary of Hn , denoted by
Hn , is defined to be {x Rn : xn = 0}.
If M is an n -manifold with boundary, a point m M belongs to the interior of M (denoted
by int M ) if (m) int Hn for some chart . And it belongs to the boundary of M (denoted by
M ) if (m) Hn for some chart .
12.1.7. Theorem. Let M and N be a smooth n -manifolds with boundary and F : M N be a
smooth diffeomorphism. Then both int M and M are smooth manifolds (without boundary). The
interior of M has dimension n and the boundary of M has dimension n 1. The mapping F
induces smooth diffeomorphisms int F : int M int N and F : M N .
Proof. Consult the marvelous text [2], proposition 7.2.6.
12.1.8. Exercise. Let V be an open subset of R3 , F : V R3 be a smooth vector field, and
(S, K, h) be a smoothly triangulated 2 -manifold suchRRthat S V . It is conventional to define the
normal component of F over S , often denoted by S FN , by the formula
ZZ
ZZ
FN =
hF h, ni
S
where n = h1 h2 . (Notation: hk is the k th partial derivative of h.)

Make sense of the preceding definition in terms of the definition of the integral of 2 -forms
over a smoothly triangulated manifold (with or without boundary). In particular, suppose that
12.2. GENERALIZED STOKES THEOREM
115
F = a i+b j+c k (where a, b, and c are smooth functions) and let = a dy dz +b dz dx+c dxdy.
This 2 -form is conventionally called the normal component of F and is denoted by FN . Notice
that FN is just where is the 1 -form associated with the vector field F. Hint. Proceed as
follows.
(a) Show that the vector n(u, v) is perpendicular to the surface S at h(u, v) for each (u, v) in
[K] by showing that it is perpendicular to D(h c)(0) whenever c is a smooth curve in [K]
such that c(0) = (u, v).
(b) Let u and v (in that order) be the coordinates in the plane of [K] and x, y, and z (in that
order) be the coordinates in R3 . Show that h (dx) = h11 du + h12 dv. Also compute h (dy)
and h (dz).
Remark. If at each point in [K] we identify the tangent plane to R2 with R2 itself and if
we use conventional notation, the v which appears in (3) of proposition 11.4.7 is just
not written. One keeps in mind that the components of h and all the differential forms
are functions on (a neighborhood of) [K].
(c) Now find h (). (Recall that = FN is defined above.)
(d) Show for each simplex (s) in K that
Z
ZZ
(hsi) =
hF h, ni .
2
[s]
(e) Finally show that if hs1 i, . . . , hsn i are the oriented 2 -simplexes of K and c =
then
Z
ZZ
(c) =
hF h, ni .
2
Pn
k=1 hsk i,
[K]
12.1.9. Exercise. Let F(x, y, z) = xz i RR

+ yz j and H be the hemisphere of x2 + y 2 + z 2 = 4 for
which z 0. Use exercise 12.1.8 to find
FN .
H
12.2. Generalized Stokes Theorem

12.2.1. Theorem (Generalized Stokes theorem). Suppose that (M,RK, h)Ris an oriented smoothly
triangulated manifold with boundary. Then the integration operator = p pZ is a cochain map
V
from the cochain complex ( (M ), d ) to the cochain complex (C (K), ).
Proof. This is an important and standard theorem, which appears in many versions and with
many different proofs. See, for example, [2], theorem 7.2.6; [20], chapter XVII, theorem 2.1; [26],
theorem 10.8; or [33], theorems 4.7 and 4.9.
Recall that when we say in Stokes theorem that the integration operator is a cochain map, we
are saying that the following diagram commutes.
V
V
d
d
d
/ p (M )
/ p+1 (M )
/

/ C p (K)

/ C p+1 (K)
Thus if is a p -form on M and hsi is an oriented (p + 1) -simplex belonging to K, then we must

have
Z

Z
d (hsi) =
(hsi).
(12.2)
p+1
116
12. STOKES THEOREM
This last equation (12.2) can be written in terms of integration over oriented simplexes:
Z
Z
hs .
d hs =
hsi
(12.3)
hsi
In more conventional notation all mention of the triangulating simplicial complex K and of the
map h is suppressed. This is justified by the fact that it can be shown that the value of the integral
is independent of the particular triangulation used. Then when the equations of the form (12.3) are
added over all the (p + 1) -simplexes comprising K we arrive at a particularly simple formulation
of (the conclusion of) Stokes theorem
Z
Z
d =
.
(12.4)
M
One particularly important topic that has been glossed over in the preceding is a discussion of
orientable manifolds (those which possess nowhere vanishing volume forms), their orientations, and
the manner in which an orientation of a manifold with boundary induces an orientation on its
boundary. One of many places where you can find a careful development of this material is in
sections 6.5 and 7.2 of [2].
12.2.2. Theorem.
Let be a 1 -form on a connected open subset U of R2 . Then is exact on U
R
if and only if C = 0 for every simple closed curve in U .
Proof. See [9], chapter 2, proposition 1.
12.2.3. Example. Let =
y dx
x dy
+ 2
. On the region R2 \ {(0, 0)} the 1-form is closed
2
+y
x + y2
x2
but not exact.

12.2.4. Exercise. What classical theorem do we get from the version of Stokes theorem given by
equation (12.4) in the special case that M is a flat 1 -manifold (with boundary) in R and is a
0 -form defined on some open set in R which contains M ? Explain.
equation (12.4) in the special case that M is a (not necessarily flat) 1 -manifold (with boundary)
in R3 and is a 0 -form defined on some open subset of R3 which contains M ? Explain.
12.2.6. Exercise. What classical theorem do we get from the version of Stokes theorem given
by equation (12.4) in the special case that M is a flat 2 -manifold (with boundary) in R2 and
is the 1 -form associated with a vector field F : U R2 defined on an open subset U of R2 which
contains M ? Explain.
R
12.2.7. Exercise. Use exercise 12.2.6 to compute S1 (2x3 y 3 ) dx + (x3 + y 3 ) dy (where S1 is the
unit circle oriented counterclockwise).
12.2.8. Exercise. Let F(x, y) = (y, x) and let Ca and Cb be the circles centered at the origin
with radii a and b, respectively, where a < b. Suppose that Ca is oriented clockwise and Cb is
oriented counterclockwise. Find
Z
Z
F dr +
Ca
F dr .
Cb
equation (12.4) in the special case that M is a (not necessarily flat) 2 -manifold (with boundary)
in R3 and is the 1 -form associated with a vector field F : U R3 defined on an open subset U
of R3 which contains M ? Explain.
12.2. GENERALIZED STOKES THEOREM
117
12.2.10. Exercise. What classical theorem do we get from the version of Stokes theorem given
by equation (12.4) in the special case that M is a (flat) 3 -manifold (with boundary) in R3 and
= where is the 1 -form associated with a vector field F : U R3 defined on an open subset
U of R3 which contains M ? Explain.
12.2.11. Exercise. Your good friend Fred R. Dimm calls you on his cell phone seeking help with
a math problem. He says that he wants to evaluate the integral of the normal component of the
vector field on R3 whose coordinate functions are x, y, and z (in that order) over the surface
of a cube whose edges have length 4. Fred is concerned that hes not sure of the coordinates of
the vertices of the cube. How would you explain to Fred (over the phone) that it doesnt matter
where the cube is located and that it is entirely obvious that the value of the surface integral he is
interested in is 192?
CHAPTER 13
GEOMETRIC ALGEBRA
13.1. Geometric Plane Algebra
In this section we look at an interesting algebraic approach to plane geometry, one that keeps
track of orientation of plane figures as well as such standard items as magnitudes,
V direction, perpendicularity, and angle. The object we start with is the Grassmann algebra (R2 ) on which we
define a new multiplication, called the Clifford product, under which it will become, in the language
of the next chapter, an example of a Clifford algebra.
13.1.1. Notation. We make a minor notational change. Heretofore the standard basis vectors for
Rn were denoted by e1 , e2 , . . . en . For our present purposes we will list them as e1 , e2 , . . . en . This
allows us to use superscripts for powers.
13.1.2. Definition. We start with the inner product space R2 . On the corresponding Grassmann
V
L
V
algebra (R2 ) = 2k=0 k (R2 ) we say that an element homogeneous of degree 0 (that is, belonging
V0 2
(R ) ) is a scalar (or real number). An element homogeneous of degree 1 (that is, belonging
to
V1 2
(R ) ) we will call a vector (or a 1-vector). And an element homogeneous of degree 2
to
V
(that is, belonging to 2 (R2 ) ) will be called a bivector (or a 2-blade, or a pseudoscalar).
We will shorten the name of the bivector e1 e2 to e12 . Similarly we write e21 for the
V bivector
e2 e1 . Notice that {1, e1 , e2 , e12 } is a basis for the 4-dimensional Grassmann algebra (R2 ), and
V
that e21 = e12 . Notice also that since 2 (R2 ) is 1-dimensional, every bivector is a scalar multiple
of e12 .
We now introduce V
a new multiplication, called Clifford multiplication
(or geometric
V
multiplication) into (R2 ) as follows: if v and w are 1-vectors in (R2 ), then their Clifford
product vw is defined by
vw := hv, wi + v w .
(13.1)
V 2
To make (R ) into a unital algebra, we will extend this new multiplication
to all elements of the
V 2
space. The Clifford product
of
a
scalar
and
an
arbitrary
element
of
(R
)
is
the
same as it is in the
V 2
Grassmann algebra (R ). We must also specify the Clifford product of a vector and a bivector.
Since we require the distributive law to hold for Clifford multiplication we need only specify the
product ei e12 and e12 ei for i = 1 and 2. If we wish Clifford multiplication to be associative there
is only one way to do this (see exercise 13.1.7): set e1 e12 = e2 , e2 e12 = e1 , e12 e1 = e2 , and
e12 e2 = e1 . Similarly, if we are to have associativity, the product of bivectors must be determined
by e12 e12 = 1.
13.1.3. CAUTION. Now we have three different multiplications to consider: the inner (or dot)
product hv, wi, the wedge (or exterior, or outer ) product v w, and the Clifford (or geometric)
V
product vw. Equation (13.1) holds only for vectors. DoV
not use it for arbitrary members of (R2 ).
(There is, of course, a possibility of confusion: Since (R2 ) is indeed a vector space, any of its
elements may in general usage
be called a vector. On the other hand, we have divided the
V
homogeneous elements of (R2 ) into three disjoint classes: scalars, vectors, and bivectors. When
we wish to emphasize that we are using the word vector in the latter sense, we call it a 1-vector.
13.1.4. Notation. The object we have just defined turns out to be a unital algebra and is an
example of a Clifford algebra.
V In the next chapter we will explain why we denote this particular
algebrathe vector space (R2 ) together with Clifford multiplicationby Cl(2, 0).
119
120
13. GEOMETRIC ALGEBRA
13.1.5. Proposition. If v is a 1-vector in Cl(2, 0), then v 2 = kvk2 . (Here, v 2 means vv.)
13.1.6. Corollary. Every nonzero 1-vector v in Cl(2, 0) has an inverse (with respect to Clifford
multiplication). As usual, the inverse of v is denoted by v 1 .
13.1.7. Exercise. Justify the claims made in the last two sentences of definition 13.1.2.
The magnitude (or norm) of a 1-vector is just its length. We will also assign a magnitude (or
norm) to bivectors. The motivation for the following definition is that we wish a bivector v w to
represent an equivalence class of directed regions in the plane, two such regions being equivalent
if they have the same area and the same orientation (positive = counterclockwise or negative =
clockwise). So we will take the magnitude of v w to be the area of the parallelogram generated
by v and w.
13.1.8. Definition. Let v and w be 1-vectors in Cl(2, 0). Define kv wk := kvkkwk sin where
is the angle between v and w (0 ).
13.1.9. Proposition. If v and w are 1-vectors in Cl(2, 0), then kv wk is the area of the parallelogram generated by v and w.
13.1.10. Proposition. If v and w are 1-vectors in Cl(2, 0), then v w = kv wke12 = det(v, w)e12 .
13.1.11. Exercise. Suppose you know how to find the Clifford product of any two elements of
Cl(2, 0). Explain how to use equation (13.1) to recapture formulas defining the inner product and
the wedge product of two 1-vectors in Cl(2, 0).
A vector in the plane is usually regarded as an equivalence class of directed segments. (Two
directed segments are taken to be equivalent if they lie on parallel lines, have the same length,
and point in the same direction.) Each such equivalence class of directed segments has a standard
representative, the one whose tail is at the origin. Two standard representatives are parallel if one
is a nonzero multiple of the other. By a common abuse of language, where we conflate the notions
of directed segments and vectors, we say that two nonzero vectors in the plane are parallel if one
is a scalar multiple of the other.
13.1.12. Proposition. Two nonzero vectors in Cl(2, 0) are parallel if and only if their Clifford
product commutes.
13.1.13. Definition. Two elements, a and b, in a semigroup (or ring, or algebra) anticommute
if ba = ab.
13.1.14. Example. The bivector e12 anticommutes with all vectors in Cl(2, 0).
13.1.15. Proposition. Two vectors in Cl(2, 0) are perpendicular if and only if their Clifford product anticommutes.
Let v and w be nonzero vectors in Cl(2, 0). We can write v as the sum of two vectors vk
and v ,where vk is parallel to w and v is perpendicular to w. The vector vk is the parallel
component of v and v is the perpendicular component of v (with respect to w).
13.1.16. Proposition. Let v and w be nonzero vectors in Cl(2, 0). Then vk = hv, wiw1 and
v = (v w)w1 .
13.1.17. Definition. If v and w are nonzero 1-vectors in Cl(0, 2), the reflection of v across w
is the map v = vk + v 7 vk v .
13.1.18. Proposition. If v and w are nonzero 1-vectors in Cl(0, 2), then the reflection v 0 = vk v
of v across w is given by v 0 = wvw1 .
Although Cl(2, 0) is a real algebra, it contains a field isomorphic to the complex numbers. We
start with the observation
e12 2 = 1. This prompts us to give a new name to e12 . We will
V0 that
V
2
call it
. In the space
(R ) 2 (R2 ), of scalars plus pseudoscalars, this element serves the same
role as the complex number i does in the complex plane C.
13.1. GEOMETRIC PLANE ALGEBRA
121
13.1.19. Definition. If := a + b
is the sum of a
scalar and a pseudoscalar, we define its
conjugate as a b
and its absolute value || as a2 + b2 .
V
V
13.1.20. Proposition. If 0 (R2 ) 2 (R2 ), then = ||2 and, if is not zero, it is invertible
2 .
with 1 = ||
V
V
13.1.21. Proposition. The space 0 (R2 ) 2 (R2 ) (with Clifford multiplication) is a field and is
isomorphic to the field C of complex numbers.
Complex numbers serve two purposes in the plane. They implement rotations via their representation in polar formVzei and, as points, they represent vectors.
V1 2 Recall that in the Clifford
1
2
2
algebra Cl(2, 0) we have
(R ) = R . So here vectors live in
(R ) while (the analogs of) comV
V
plex numbers live in 0 (R2 ) 2 (R2 ). Since these are both two-dimensional vector spaces, they are
isomorphic. It turns out that there exists a simple, natural, and very useful isomorphism between
these spaces: left multiplication by e1 .
V
V
V
13.1.22. Notation. In the remainder of this section we will shorten (R2 ) to
and k (R2 ) to
Vk
for k = 0, 1, and 2.
V
V
V
13.1.23. Proposition. The map from 1 to 0 2 defined by v 7 v := e1 v is a vector space
isomorphism. The mapping is also an isometry in the sense that |
v | = kvk. The mapping is its
own inverse; that is, e1 v = v.
13.1.24. Definition. For R define e by the usual power series
X
(
)k
e :=
= cos + (sin )
.
k!
k=0
13.1.25. Exercise. Discuss the convergence problems (or lack thereof) that arise in the preceding
definition.
Next we develop a formula for the positive (counterclockwise) rotation of a nonzero vector v
through angle of radians. Let v 0 be the vector in its rotated position. First move these 1-vectors
to their corresponding points in the (Cl(2, 0) analog of the) complex plane. Regarding these two
V
V
points, = e1 v, 0 = e1 v 0 0 2 as complex numbers (via proposition 13.1.21), we see that
V
0 = e . Then move and 0 back to 1 to obtain the formula in the following proposition.
13.1.26. Proposition. Let v be a nonzero 1-vector in Cl(2, 0). After being rotated by radians
in the positive direction it becomes the vector v 0 . Then
v 0 = e v = ve .
In definition 13.1.2 and exercise 13.1.7 we established that a necessary condition for Cl(2, 0)
under Clifford multiplication to be an algebra, is that certain relations involving the basis elements,
1, e1 , e2 , and e12 hold. This doesnt prove, however, that if these relations are satisfied, then Cl(2, 0)
under Clifford multiplication is an algebra. This lacuna is not hard to fill.
13.1.27. Proposition. The Clifford algebra Cl(2, 0) is, in fact, a unital algebra isomorphic to the
matrix algebra M2 (R).

1 0
0 1
Hint for proof . Try mapping e1 to
and e2 to
.
0 1
1 0
13.1.28. Example. The Clifford algebra Cl(2, 0) is not a Z+ -graded algebra.
13.1.29. Proposition. The Clifford algebra Cl(2, 0) does not have the cancellation property.
Hint for proof . Consider uv and uw where u = e2 e12 , v = e1 + e2 , and w = 1 + e2 .
122
13. GEOMETRIC ALGEBRA
13.2. Geometric Algebra in 3-Space

V
V
L
13.2.1. Definition. We now give the Grassmann algebra (R3 ) = 3k=0 k (R3 ) a new multiplication, called the Clifford product (or geometric product).
The resulting Clifford algebra will be
V 3
denoted by Cl(3, 0). Recall that the vector space (R ) is 8-dimensional. If we let {e1 , e2 , e3 } be
the standard basis for R3 , then
^0
(R3 ) = span{1},
^1
(R3 ) = span{e1 , e2 , e3 },
^2
(R3 ) = span{e2 e3 , e3 e1 , e1 e2 }, and
^3
(R3 ) = span{e1 e2 e3 }.
As in Cl(2, 0) we use the abbreviations e23 = e2 e3 , e31 = e3 e1 , and e12 = e1 e2 . Additionally
V
we let e123 = e1 e2 e3 . As before, members of 0 (R3 ) are called scalars. Element homogeneous
V1 2
of degree 1 (that is, belonging to
(R ) ) are called vectors (or a 1-vectors). Elements homoV
geneous of degree 2 (that is, belonging to 2 (R2 ) ) are called bivectors. Decomposable elements
V2 2
(R ) are called 2-blades. (Thus any bivector is a sum of 2-blades.) And elements homogeof
neous of degree 3 (that is, scalar multiples of e123 ) are called trivectors (or pseudoscalars, or
3-blades.
V
We start by taking scalar multiplication to be exactly as it is in the Grassmann algebra (R3 ).
Then we define the Clifford product or geometric product of two 1-vectors, v and w, exactly
as we did for Cl(2, 0):
vw := hv, wi + v w .
(13.2)
Additionally, we set e1 e2 e3 := e1 e2 e3 .
13.2.2. Exercise. Show that in order for definition 13.2.1 to produce an algebra we need the
following multiplication table to hold for the basis elements.
1
e1
e2
e3
e23
e31
e12
e123
1
e1
e2
e3
e23
e31
e12
1
e1
e2
e3
e23
e31
e12
e1
1
e12 e31 e123 e3
e2
e2 e12
1
e23
e3
e123 e1
e3
e31 e23
1
e2
e1
e123
e23 e123 e3
e2
1 e12 e31
e31
e3
e123 e1
e12
1 e23
e12 e2
e1
e123 e31 e23
1
e123 e23
e31
e12
e1 e2 e3
e123
e123
e23
e31
e12
e1
e2
e3
1
V
13.2.3. Notation. The object defined in 13.2.1, the vector space (R3 ) together with Clifford
multiplication, turns out to be a unital algebra and is a second example of a Clifford algebra. We
will see in the next chapter why this algebra is denoted by Cl(3, 0).
In the preceding section we commented on the fact that, geometrically speaking, we picture a
vector in the plane as an equivalence class of directed intervals (that is, directed line segments),
two intervals being equivalent if they lie on parallel lines, have the same length, and point in the
same direction. In a similar fashion it is helpful to have a geometrical interpretation of a 2-blade.
Let us say that an directed interval in 3-space is an oriented parallelogram generated by a pair of
vectors. Two such intervals will be said to be equivalent if they lie in parallel planes, have the same
area, and have the same orientation. We may choose as a standard representative of the 2-blade
v w the parallelogram formed by putting the tail of v at the origin and the tail of w at the head
13.2. GEOMETRIC ALGEBRA IN 3-SPACE
123
of v. (Complete the parallelogram with by adding, successively, the vectors v and w.) The
standard representative of the 2-blade w v = v w is the same parallelogram with the opposite
orientation: put the tail of w at the origin and the tail of v at the head of w.
13.2.4. Proposition. The Clifford algebra Cl(3, 0) is indeed a unital algebra.
Hint for proof . Consider the subalgebra of M4 (R) generated by the matrices
0 0 0 1
0 0 1 0
1 0 0
0
0 0 1 0
0
, 2 = 0 0 0 1 , and 3 = 0 1 0
1 =
0 1 0 0
1 0 0 0
0 0 1 0 .
1 0 0 0
0 1 0 0
0 0 0 1
13.2.5. Exercise. The proof of the preceding proposition provides a representation of the algebra
Cl(3, 0) as an 8-dimensional algebra of matrices of real numbers. Another, perhaps more interesting,
representation of the same algebra is by means of the Pauli spin matrices:

0 1
0 i
1 0
1 =
, 2 =
, and 3 =
.
1 0
i 0
0 1
Let C(2) be the real vector space of 2 2-matrices with complex entries
(a) Show that although the complex vector space M2 (C) is 4-dimensional, C(2) is 8-dimensional.
(b) Show that {I, 1 , 2 , 3 , 2 3 , 3 1 , 1 2 , 1 2 3 } is a linearly independent subset of C(2).
(Here, I is the 2 2 identity matrix.)
(c) Find an explicit isomorphism from Cl(3, 0) to C(2).
As was the case in the plane we wish to assign magnitudes to 2-blades. We adopt the same
convention: the magnitude of v w is the area of the parallelogram generated by v and w.
13.2.6. Definition. Let v and w be 1-vectors in Cl(3, 0). Define kv wk := kvkkwk sin where
is the angle between v and w (0 ).
13.2.7. Proposition. Let u be a unit vector in R3 and P be a plane through the origin perpendicular
to u. Then the reflection v of a vector v R3 with respect to P is given by
v = uvu .
Hint for proof . If vk and v are, respectively, the parallel and perpendicular components of v
with respect to u, then v = v vk .
13.2.8. Exercise. Using the usual vector notation, how would you express v in the preceding
proposition?
13.2.9. Exercise. Find the area of the triangle in R3 whose vertices are the origin, (5, 4, 2),
and (1, 4, 6).
13.2.10. Example. To rotate a vector in 3-space by 2 radians about an axis determined by a
nonzero vector a, choose unit vectors u1 and u2 perpendicular to a so that the angle between u1
and u2 is radians. Let P1 and P2 be the planes through the origin perpendicular to u1 and u2 ,
respectively. Then the map v 7 v = R1 vR where R = u1 u2 is the desired rotation.
13.2.11. Example. It is clear that a 90 rotation in the positive (counterclockwise) direction about
the z-axis in R3 will take the point (0, 2, 3) to the point (2, 0, 3). Show how the formula derived
in the previous example gives this result when we choose u1 = e1 , u2 = 12 e1 + 12 e2 , a = (0, 0, 1),
and v = 2e2 + 3e3 .
CHAPTER 14
CLIFFORD ALGEBRAS
In this last chapter we will barely scratch the surface of the fascinating subject of Clifford
algebras. For a more substantial introduction there are now, fortunately, many sources available.
Among my favorites are [1], [7], [10], [14], [15], [24], [25],and [31].
14.1. Quadratic Forms
14.1.1. Definition. A bilinear form B on a real vector space V is symmetric if B(u, v) = B(v, u)
for all u, v V .
14.1.2. Definition. Let V be a real vector space. A function Q : V R is a quadratic form if
(i) Q(v) = Q(v) for all v, and
(ii) the map B : V V R : (u, v) 7 Q(u + v) Q(u) Q(v) is a bilinear form.
In this case B is the bilinear form associated with the quadratic form Q. It is obviously symmetric. Note: In many texts B(u, v) is defined to be 21 [Q(u + v) Q(u) Q(v)].
14.1.3. Proposition. Let Q be a quadratic form on a real vector space. Then Q(0) = 0.
14.1.4. Example. Let B be a symmetric bilinear form on a real vector space V . If we define
Q : V R by Q(v) = B(v, v), then Q is a quadratic form on V .
14.1.5. Proposition. If Q is a quadratic form on a real vector space V , then
Q(u + v + w) Q(u + v) Q(u + w) Q(v + w) + Q(u) + Q(v) + Q(w) = 0
for all u, v, w V .
14.1.6. Corollary. If Q is a quadratic form on a real vector space V , then Q(2v) = 4Q(v) for
every v V .
Hint for proof . Look at Q(v + v v).
14.1.7. Proposition. If Q is a quadratic form on a real vector space V , then Q(v) = 2 Q(v) for
all R and v V .
14.2. Definition of Clifford Algebra
14.2.1. Definition. Let V be a real vector space. A pair (U, ), where U is a real unital algebra
and : V U is a linear map, is universal over V if for every real unital algebra A and every
linear map f : V A there exists a unique unital algebra homomorphism fe: U A such that
fe = f .
14.2.2. Proposition. Let V be a real vector space. If U and U 0 are unital algebras universal
over V , then they are isomorphic.
14.2.3. Example. For every real vector space V there is a unital algebra which is universal over V .
Hint for proof . See 8.2.2.
14.2.4. Definition. Let V be a real vector space with a quadratic form Q and A be a real unital
algebra. A map f : V A is a Clifford map if
125
126
(i) f is linear, and

2
(ii) f (v) = Q(v)1A for every v V .
14.2.5. Proposition. Condition (ii) in definition 14.2.4 is equivalent to
(ii 0 ) f (u)f (v) + f (v)f (u) = B(u, v)1A for all u, v V ,
where B is the bilinear form associated with Q.
14.2.6. Definition. Let V be a real vector space with a quadratic form Q. The Clifford
algebra over V is a real unital algebra Cl(V, Q), together with a Clifford map j : V Cl(V, Q),
which satisfies the following universal condition: for every real unital algebra A and every Clifford
map f : V A, there exists a unique unital algebra homomorphism fb: Cl(V, Q) A such that
fb j = f .
14.2.7. Proposition. Let V be a real finite dimensional vector space with a quadratic form Q. If
the Clifford algebra Cl(V, Q) exists, then it is unique up to isomorphism.
14.2.8. Example. For every real finite dimensional vector space V with a quadratic form Q the
Clifford algebra Cl(V, Q) exists.
Hint for proof . Try T (V )/J where J is the ideal in T (V ) generated by elements of the form
v v Q(v)1T (V ) where v V .
14.3. Orthogonality with Respect to Bilinear Forms
14.3.1. Definition. Let B be a symmetric bilinear form on a real vector space V . Vectors v and
w in V are orthogonal, in which case we write v w, if B(v, w) = 0. The kernel of B is the
set of all k V such that k v for every v V . The bilinear form is nondegenerate if its kernel
is {0}.
14.3.2. Exercise. One often sees the claim that the classification of Clifford algebras amounts to
classifying vector spaces with quadratic forms. Explain precisely what is meant by this assertion.
14.3.3. Proposition. Let B be a symmetric bilinear form on a real finite dimensional vector
space V . Suppose that V is an orthogonal direct sum V1 Vn of subspaces. Then B is
nondegenerate if and only if the restriction of B to Vk is nondegenerate for each k. In fact, if we
denote by Bj the restriction of B to Vj , then ker B = ker B1 ker Bn .
14.3.4. Proposition. Let Q be a quadratic form on a real finite dimensional vector space V and
let {e1 , . . . , en } be a basis for V which is orthogonal with respect to the bilinear form B associated
with Q. If Q(ek ) is nonzero for 1 k p and Q(ek ) = 0 for p < k n, then the kernel of B is
the span of {ep+1 , . . . , en }.
14.3.5. Proposition. Let Q be a quadratic form on a real finite dimensional vector space V . If
dim V = n, then dim Cl(V, Q) = 2n .
14.3.6. Exercise. Let V be a real finite dimensional vector space and let Q be the quadratic form
which is identically zero on V . Identify Cl(V, Q).
14.4. Examples of Clifford Algebras
14.4.1. Definition. Let V be a finite dimensional real vector space and B be a symmetric bilinear
form on V . An ordered basis E = (e1 , . . . , en ) for V is B-orthonormal if
(a) B(ei , ej ) = 0 whenever i 6= j and
(b) for each i Nn the number B(ei , ei ) is 1 or +1 or 0.
14.4.2. Theorem. If V is a finite dimensional real vector space and B is a symmetric bilinear
form on V , then V has a B-orthonormal basis.
14.4. EXAMPLES OF CLIFFORD ALGEBRAS
127
Proof. See [4], chapter 1, theorem 7.6.

14.4.3. Convention. Let V be a finite dimensional real vector space, let B be a symmetric
bilinear form on V , and let Q be the quadratic form associated with B. Let us agree that whenever
E = (e1 , . . . , en ) is an ordered B-orthonormal basis for V , we order the basis elements in such a
way that for some positive integers p and q
if 1 i p;
1,
1, if p + 1 i p + q;
Q(ei ) =
0,
if p + q + 1 i n.
14.4.4. Theorem. Let V be a finite dimensional real vector space, let B be a symmetric bilinear
form on V , and let Q be the quadratic form associated with B. P
Then there exist p, q Z+ such
that if E = (e1 , . . . , en ) is a B-orthonormal basis for V and v = vk ek , then
Q(v) =
p
X
k=1
vk
p+q
X
vk 2 .
k=p+1
Proof. See [4], chapter 1, theorem 7.11.

14.4.5. Notation. If (V, Q) is a finite dimensional real vector space V with a nondegenerate
quadratic form Q, we often denote the Clifford algebra Cl(V, Q) by Cl(p, q) where p and q are as
in theorem 14.4.4.
14.4.6. Proposition. Let f : (V, Q) (W, R) be a linear map between finite dimensional real
vector spaces with quadratic forms. If R(f (v)) = Q(v) for every v V we say that f is an
isometry.
(a) If f is such a linear isometry, then there exists a unique unital algebra homomorphism
Cl(f ) : Cl(V, Q) Cl(W, R) such that Cl(f )(v) = f (v) for every v V .
(b) The pair of mappings Cl described above is a covariant functor from the category of vector
spaces with quadratic forms and linear isometries to the category of Clifford algebras and
unital algebra homomorphisms.
(c) If a linear isometry f is an isomorphism, then Cl(f ) is an algebra isomorphism.
14.4.7. Proposition. Let V be a real finite dimensional vector space, Q be a quadratic form on V ,
and A = Cl(V, Q) be the associated Clifford algebra.
(a) The map f : V V : v 7 v is a linear isometry.
(b) If = Cl(f ), then 2 = id.
(c) Let A0 = {a A : (a) = a} and A1 = {a A : (a) = a}. Then A = A0 A1 .
(d) If i, j {0, 1}, then Ai Aj Ai+j (where i + j indicates addition modulo 2). This says
that a Clifford algebra is a Z2 -graded (or Z/2Z -graded) algebra.
Hint for proof of (c). If a A, let a0 = 12 (a + (a)).
14.4.8. Example. Let V = R and Q(v) = v 2 for every v V . Then the Clifford algebra Cl(1, 0)
associated with (R, Q) is isomorphic to R R.
Hint for proof . Let f : R Cl(1, 0) be the Clifford map associated with Q and let e := f (1).
Try the function : Cl(1, 0) R R under which 1 7 (1, 1) and e 7 (1, 1).
14.4.9. Example. Let V = R and Q(v) = v 2 for every v V .
(a) The algebra Cl(0, 1) is isomorphic to C.
(b) The algebra Cl(0, 1) can be represented as a subalgebra of M2 (R).
Hint for proof . For part
(a) keep in mind that as a real algebra C is 2-dimensional. For part
1 0
0 1
(b) consider the matrices
and
0 1
1 0
128
14.4.10. Example. Let V = R2 and Q(v) = v1 2 + v2 2 for every v V . Then Cl(2, 0)

= M2 (R).
(See 13.1.2 and 13.1.4.)
14.4.11. Example. Let V = R3 and Q(v) = v1 2 + v2 2 + v3 2 for every v V . Then Cl(3, 0) is
isomorphic to both a subalgebra of M4 and the algebra C(2) generated by the Pauli spin matrices.
(See 13.2.4 and 13.2.5.)
In 1833, when William Rowan Hamilton was still in his twenties, he presented a paper to the
Royal Irish Academy in which he explained how to represent complex numbers as pairs of real
numbers. He was fascinated by the connections this produced between complex arithmetic and
the geometry of the plane. For more than a decade he searched, sometimes quite obsessively, for
a 3-dimensional algebraic analog which would facilitate studies in the geometry of 3-space. (In
modern language, he was seeking a 3-dimensional normed division algebra. As it turns out, there
arent any!) On October 16, 1843, while walking with his wife along the Royal Canal in Dublin on
his way to chair a meeting of the Royal Irish Academy, he suddenly understood that to manage
triples of real numbers algebraically, he would need to add, in some sense, a fourth dimension. The
objects he could work with were of the form a + ib + jc + kd where a, b, c, and d are real numbers.
In this flash of inspiration he saw the necessary rules for multiplication of these objects. Although,
as far as we know, he had no previous history as a graffiti artist, he stopped with his wife to carve
into the stones of Brougham Bridge, which they were passing, the key to the multiplication of these
new objects, quaternions:
i2 = j2 = k2 = ijk = 1 .
14.4.12. Exercise. Derive from the preceding equations the multiplication table for quaternions.
14.4.13. Example. Let V = R2 and Q(v) = v12 v22 for every v V .
(a) The algebra Cl(0, 2) is isomorphic to the algebra H of quaternions.
(b) The algebra Cl(0, 2) can be represented as a subalgebra of M4 (R).
0 1 0 0
1 0 0 0
Hint for proof . For part (b) consider the matrices

0 0 0 1 and
0 0 1 0
0 0 1 0
0 0
0 1
.
1 0
0 0
0 1 0 0
14.4.14. Example. The Clifford algebra Cl(1, 1) is isomorphic to M2 (R).

1 0
0 1
Hint for proof . Consider the matrices
and
.
0 1
1 0
14.4.15. Exercise. Take a look at the web page [23] written by Pertti Lounesto.
14.4.16. Exercise. The Clifford algebra Cl(3, 1) (Minkowski space-time algebra) is isomorphic to
M4 (R).
Hint for proof . Exercise 14.4.15.
Bibliography
1. Rafal Ablamowicz and Garret Sobczyk (eds.), Lectures on Clifford (Geometric) Algebras and Applications,
Birkh
auser, Boston, 2004. 125
2. Ralph Abraham, Jerrold E. Marsden, and Tudor Ratiu, Manifolds, Tensor Analysis, and Applications, AddisonWesley, Reading, MA, 1983. 114, 115, 116
3. Richard L. Bishop and Richard J. Crittenden, Geometry of Manifolds, Academic Press, New York, 1964. 83, 99
4. William C. Brown, A second Course in Linear Algebra, John Wiley, New York, 1988. 127
5. Stewart S. Cairns, A simple triangulation method for smooth manifolds, Bull. Amer. Math. Soc. 67 (1961),
389390. 110
6. P. M. Cohn, Basic Algebra: Groups, Rings and Fields, Springer, London, 2003. 9
7. John S. Denker, An Introduction to Clifford Algebra, 2006,
http://www.av8n.com/physics/clifford-intro.htm. 125
8. J. Dieudonne, Treatise on Analysis, Volumes I and II, Academic Press, New York, 1969, 1970. 67
9. Manfredo P. do Carmo, Differential Forms and Applications, Springer-Verlag, Berlin, 1994. 116
10. Chris Doran and Anthony Lasenby, Geometric Algebra for Physicists, Cambridge University Press, Cambridge,
2003. 125
11. John M. Erdman, A ProblemText in Advanced Calculus,
http://web.pdx.edu/~erdman/PTAC/PTAClicensepage.html. 67, 86, 90
12.
, A Companion to Real Analysis, 2007,
http://web.pdx.edu/~erdman/CRA/CRAlicensepage.html. 67
13. Douglas R. Farenick, Algebras of Linear Transformations, Springer-Verlag, New York, 2001. 48, 50
14. D. J. H. Garling, Clifford Algebras: An Introduction, Cambridge University Press, Cambridge, 2011. 125
15. David Hestenes and Garret Sobczyk, Clifford Algebra to Geometric Calculus, D. Reidel, Dordrecht, 1984. 125
16. Kenneth Hoffman and Ray Kunze, Linear Algebra, second ed., Prentice Hall, Englewood Cliffs,N.J., 1971. 52, 65
17. S. T. Hu, Differentiable Manifolds, Holt, Rinehart, and Winston, New York, 1969. 110
18. Thomas W. Hungerford, Algebra, Springer-Verlag, New York, 1974. 9
19. Saunders Mac Lane and Garrett Birkhoff, Algebra, Macmillan, New York, 1967. 9
20. Serge Lang, Fundamentals of Differential Geometry, Springer Verlag, New York, 1999. 115
21. John M. Lee, Introduction to Smooth Manifolds, Springer, New York, 2003. 83, 99, 110
22. Lynn H. Loomis and Shlomo Sternberg, Advanced calculus, Jones and Bartlett, Boston, 1990. 67
23. Pertti Lounesto, Counterexamples in Clifford Algebras, 1997/2002,
http://users.tkk.fi/~ppuska/mirror/Lounesto/counterexamples.htm. 128
24.
, Clifford Algebras and Spinors, second ed., Cambridge University Press, Cambridge, 2001. 125
25. Douglas Lundholm and Lars Svensson, Clifford algebra, geometric algebra, and applications, 2009,
http://arxiv.org/abs/0907.5356v1. 125
26. Ib Madsen and Jrgen Tornehave, From Calculus to Cohomology: de Rham cohomology and characteristic classes,
Cambridge University Press, Cambridge, 1997. 115
27. Theodore W. Palmer, Banach Algebras and the General Theory of -Algebras III, Cambridge University Press,
Cambridge, 1994/2001. 9
28. Charles C. Pinter, A Book of Abstract Algebra, second ed., McGraw-Hill, New York, 1990. 73
29. Steven Roman, Advanced Linear Algebra, second ed., Springer-Verlag, New York, 2005. 19, 65, 76
30. I. M. Singer and J. A. Thorpe, Lecture Notes on Elementary Topology and Geometry, Springer Verlag, New York,
1967. 110
31. John Snygg, A New Approach to Differential Geometry using Cliffords Geometric Algebra, Springer, New York,
2012. 125
32. Gerard Walschap, Metric Structures in Differential Geometry, Springer-Verlag, New York, 2004. 99
33. Frank W. Warner, Foundations of Differentiable Manifolds and Lie Groups, Springer Verlag, New York, 1983.
110, 115
129
Index
V (inverse of a morphism ), 30
(V ) (Grassmann or exterior algebra), 79
Vk
(V ) (homogeneous elements of degree k), 79
(curl of a 1-form), 100
f (m) (gradient of f at m), 100
(divergence of a 1-form), 100
(a) (gradient of at a), 70
hsi (boundary of a simplex), 109
p (boundary map), 109
A (characteristic function of A), 33
f |A (restriction of f to A), 7
-algebra (algebra with involution), 60
-homomorphism (star homomorphism), 60
T (B) (inverse image of B under T ), 22

T (A) (direct image of A under T ), 22
(s) = (v0 , . . . , vp ) (open p -simplex), 108
[s] = [v0 , . . . , vp ] (closed p -simplex), 108
hsi = hv0 , . . . , vp i (closed p -simplex), 108
kxk (norm of a vector), 56
kv wk (norm of a bivector), 120, 123
v (= e1 v), 121
x
b (= (x)), 26
f ' g (f is tangent to g), 86
' (tangency at 0), 68
vk (parallel component of v), 120
v (perpendicular component of v), 120
xs (alternative notation for x(s)), 9
F (pre-annihilator of F ), 27
A (orthogonal complement of a set), 58
M (annihilator of M ), 27
T (inner product space adjoint of T ), 61
T (vector space adjoint of T ), 31
T t (transpose of T ), 61
f (B) (preimage of B under f ), 31
f (A) (image of A under f ), 31
v , 25
(V , d), V (cochain complex), 106
M (boundary of a manifold), 114
Hn (boundary of a half-space), 114
int M (interior of a manifold with boundary), 114
int Hn (interior of a half-space), 114
[T ] (matrix representation of T ), 24
[x] (equivalence class containing x), 35
|A| (the forgetful functor acting on A), 31
/ T (morphism in a category), 29
S
At (transpose of A), 19
V
V
(M ), k (M ) (differential forms), 98
Vk
F , 105
hx, vi (alternative notation for v (x)), 25
f g (functions agree on some neighborhood), 89
(f, g) (function into a product), 7
S
T (tensor products of linear maps), 78
L
k=0 Vk (direct sum), 34, 80

BC (sums of products of elements in B and C), 40
M N (inner product space direct sum of M
and N ), 57
M N (vector space direct sum of M and N ), 14, 34
U V (tensor product), 76
V /M (quotient of V by M ), 35
[x] [y] = [x y] in T (V )/J, 80
(wedge product), 79
f g (product function), 7
p g q (supremum of projections), 64
p f q (infimum of projections), 64
H N , 64
M V (M is a subspace of V ), 12
V
= W (isomorphism of vector spaces), 23
p q (ordering of projections in a -algebra), 64
p q (orthogonality of projections), 63
x y (orthogonal vectors), 57
[a, b] (closed segment in a vector space), 107
|K| (polyhedron of a complex), 109
A[x]
of polynomials), 43
(algebra

A [x] (algebra of formal power series), 43
F[x] (polynomial algebra with coefficients in F), 44
hx, yi (inner product), 55
(notation for composition of morphisms), 29
1 (standard one-element set), 8
0-form, 98
1-form, 98
1-manifold, 83
1-vector, 119, 122
2-blade, 119, 122
2-manifold, 83
3-blade, 122
3-manifold, 83
Abelian group, 1
AbGp
the category, 29
absolute value
in Cl(2, 0), 121
absolutely
summable
131
132
sequence, 13
addition
pointwise, 5
additive
identity, 1
inverse, 1
inverses, 10
adjoint, 61
of the unilateral shift, 61
vector space, 31
adjointable, 61
affine subspace, 108
A(H, K), A(H) (adjointable maps), 61
algebra, 40
Z+ -graded, 79
Clifford, 126
commutative, 40
de Rham cohomology, 106
exterior, 79
Grassmann, 79
homomorphism, 40
unital, 40
polynomial, 44
quotient, 42
simple, 41
tensor, 80
unital, 40
with involution
L(V ) as a, 62
algebraic
tangent space, 89
tangent vector, 89
algebraically closed, 46, 48
ALG
bijective morphisms in, 40
alternating, 74
Altk (V ) (set of alternating k-linear maps), 81
annihilating polynomial, 45
annihilator, 9, 27
anticommute, 120
antisymmetric, 17
arithmetic
progression, 13
associative, 10
associativity, 1
atlas, 84
smooth, 84
atlases
compatible, 84
equivalent, 84
basis, 16
curves at a point, 87
dual, 26
for the geometric tangent space, 91
ordered, 82
orthonormal, 58
right-handed, 82
standard
for R3 , 16
for Rn , 16
INDEX
p (Betti number), 110

Betti number, 105, 110
bijection, 6
bijective, 6
bijective morphisms
are invertible
in ALG, 40
need not be invertible
in POSET, 30
bilinear
form
associated with a quadratic form, 125
nondegenerate, 126
symmetric, 125
map, 73
binary operation, 1
(cross-product of 1-forms), 100
bivector, 119, 122
norm of a, 120
B k (M ) (exact k-forms), 105
B-orthonormal, 126
bound
lower, 17
upper, 17
boundaries, 109
boundary
maps, 109
of a half-space, 114
of a manifold, 114
of a simplex, 109
bounded
function, 13
sequence, 13
B(S)
as a unital algebra, 41
B p (K) (space of simplicial p -coboundaries), 110
Bp (K) (space of simplicial p -boundaries), 109
bundle
cotangent, 98
tangent, 97
C (set of complex numbers), 2
C([a, b]) (continuous real valued functions), 21
C 1 ([a, b]) (continuously differentiable functions), 22
calculus
polynomial functional, 44
cancellable
left, 30
right, 30
cancellation property, 9
category, 29
AbGp as a, 29
POSET as a, 29
SET as a, 29
VEC as a, 29
concrete, 29
Cayley-Hamilton theorem, 50
centered at, 84
chain, 18
chain rule, 89, 92, 94
chains, 109
INDEX
change of variables, 91
characteristic
Euler, 110
function, 33
polynomial, 49
zero, 74
chart, 84
charts
compatible, 84
(K) (Euler characteristic), 110
C (smooth), 84
-manifold, 84
C (M ) (smooth real valued functions on M ), 97
C (U, Rn ), 84

V
V
C M, (M ) , C M, k (M ) (smooth differential
forms), 99
Cm
(M, R) (smooth functions defined in a
neighborhood of m), 89
Cl(2, 0), 119
Cl(3, 0), 122
Clifford
algebra, 126
map, 125
multiplication
in the plane, 119
product, 122
in the plane, 119
closed
algebraically, 48
differential form, 101
face of a simplex, 108
segment, 107
simplex, 108
Cl(V, Q) (Clifford algebra), 126
e
cm (tangency class of curves), 87
coboundaries, 105
coboundary
k-, 106
operator, 106
simplicial, 110
cochain, 110
k-, 106
complex, 106
map, 107
cocycle
k-, 106
simplicial, 110
cocycles, 105
codimension, 38
codomain, 6
coefficient
Fourier, 58
leading, 44
cohomology
de Rham, 106
group, 106
simplicial, 110
cohomology group
de Rham, 105
cokernel, 37
133
coker T (the cokernel of T ), 37

column vector, 18
combination
convex, 107
formal linear, 33
linear, 15
commutative, 10
algebra, 40
diagram, 6
ring, 1
commutativity, 1
comparable, 18
compatible
atlases, 84
charts, 84
complement, 14
orthogonal, 58
complementary subspace, 14
complete
order, 31
complex
cochain, 106
simplicial, 108
dimension of a, 109
vector space, 10
components, 7
composite, 83
composition
of morphisms, 29
concrete category, 29
conjugate
in Cl(2, 0), 121
linear, 55
connecting maps, 83
constant
locally, 105
polynomial, 44
continuous
differentiability, 22
continuously differentiable, 36
contravariant, 30
conventions
BC denotes sums of products of elements in B
and C, 40
V0
(V ) = F, 80
additive notation for Abelian groups, 5
all k-forms are smooth differential k-forms, 99
all categories are concrete, 29
all charts belong to the differential structure, 84
all differential forms are smooth, 99
all manifolds are smooth, 84
all one-forms are smooth differential one-forms, 98
all unital rings are nontrivial, 2
convolution a b written as a product ab, 43
e m = dF
b m , 94
dFm = dF
for homomorphisms write gf for g f , 6
for linear maps write T S for T S, 21
for multiplication write xy for x y, 1
ideals are two-sided, 41
134
identification of R with 1-dimensional tangent

spaces, 95
identification of a vector with a scalar valued
function, 25
in chapter 5 all vector spaces are real or complex,
55
in chapters 1011 all manifolds are smooth and
oriented, 97
in chapters 1011 all vector spaces are real, finite
dimensional, and oriented, 97
in inner product spaces projection means
orthogonal projection, 63
in section 7.3 all fields are of characteristic zero, 74
notation for composition
of algebra homomorphisms, 41
of morphisms, 29
notations associated with 1-dimensional tangent
spaces, 95
notations for a vector in a finite dimensional space,
18
on Cartesian products, 7
on ordering B-orthonormal bases, 127
open subsets of Rn are manifolds in a canonical
fashion, 84
spectrum of elements of nonunital algebras, 42
the geometric and algebraic tangent spaces are
the same, 93
when f is a 0-form, f means f , 99
write T x for T (x) when T is linear, 21
convergent
sequence, 13
convex
combination, 107
hull, 108
independent, 108
set, 108
convolution, 43
coordinate
projection, 84
coordinate chart, 84
coordinates, local, 84
coproduct
in a category, 33
uniqueness of, 34
cotangent
bundle, 98
projection, 98
cotangent space, 95
covariant, 30
Cp (K) (p -chains), 109
C (map from Tem to Rn ), 87
C p (K) (p -cochains), 110
cross-product, 100
curl, 100
curve, 83, 86
at a point, 86
parametrization of a, 83
smooth, 83
standard basis, 87
cycle, 73
INDEX
length of a, 73
cycles, 109
de Rham cohomology algebra, 106
de Rham cohomology group, 105
decomposable
element of a Grassmann algebra, 79
tensor, 76
decomposition
diagonalizable plus nilpotent, 52
primary, 52
decreasing
sequence, 13
deg p (degree of a polynomial), 44
degree
of a decomposable element, 79
of a homogeneous element, 79
of a polynomial, 44
fa (a translate of f ), 68
(diagonal mapping), 8
dependent, 15
derivation, 89
derivative
directional, 70
determinant, 75
function, 75
b m (differential of f at m), 94
dF
e m (differential of f at m), 88
dF
diagonal
mapping, 8
matrix, 51
diag(1 , . . . , n ) (diagonal matrix), 51
diagonalizable, 51
plus nilpotent decomposition, 52
unitarily, 64
diagram
commutative, 6
differentiable, 69
continuously, 22, 36
infinitely, 84
differential, 69
k-form, 98
form, 98
closed, 101
exact, 101
pullback of, 110
smooth, 99
G
ateaux, 70
manifold, 84
of a smooth map between manifolds, 88, 94
one-form, 98
structure, 84
dfa (the differential of f at a), 69
differentiation
exterior, 99
is linear, 22
dimension
of a simplex, 108
of a simplicial complex, 109
of a vector space, 19
dim K (dimension of a simplicial complex K), 109
INDEX
dim V (dimension of a vector space V ), 19

direct image, 22
direct sum
as a coproduct, 34
as a product, 34
external, 34, 80
external orthogonal, 57
internal
for vector spaces, 14
internal orthogonal, 57
projections associated with, 49
directed interval, 122
directed segment, 5
directional derivative, 70
disjoint
permutations, 73
distributivity, 1
divergence, 100
div (divergence of a 1-form), 100
divides, 45
division
ring, 1
division algorithm, 45
divisor
greatest common, 47
of zero, 9
domain, 6
dom f (domain of a function f ), 6
dual basis, 26
dual space, 25
duality functor
Dv f (directional derivative), 70
e12 (= e1 e2 ), 119
eigenspace, 50
eigenspaces
generalized, 52
eigenvalue, 49
eigenvector, 50
elementary tensor, 76
EN M (projection along N onto M ), 39
epic, 30
epimorphism, 29
S , (function from S into 1), 8
equivalent
atlases, 84
orderings of vertices, 108
unitarily, 64
essential
uniqueness, 34
Euclidean
locally, 84
Euler characteristic, 110
even
permutation, 73
even function, 14
eventually
zero, 13
exact
135
exact sequence
of cochain complexes, 107
of vector spaces, 35
short, 36
extension, 7
exterior
algebra, 79
differentiation operator, 99
product, 79
external
direct sum, 34, 57, 80
F(S), F(S, F), FS (F-valued functions on S), 9
F(S, T ) (functions from S to T ), 9
fb (germ containing f ), 89
F(S)
as a commutative unital algebra, 40
F(S, F), F(S)
F-valued functions on S, 5
as a vector space, 11
as an Abelian group, 5
face, 108
factorization, 47
field, 1
algebraically closed, 46
scalar, 70
vector, 97
finite
support, 9
finite dimensional, 19
first isomorphism theorem, 35
flat, 83
forgetful functor, 30
form
closed , 101
differential, 98
exact, 101
multilinear, 73
one-, 98
quadratic, 125
smooth differential, 99
formal linear combination, 33
formal power series, 43
Fourier
coefficient, 58
sum, 58
F , F (local representatives), 85
Fraktur fonts, 4
free
object, 32
vector space, 32
function
bounded, 13
characteristic, 33
diagonal, 8
integrable, 13
interchange, 8
product, 7
switching, 8
functional
calculus
136
polynomial, 44
linear, 25
multilinear, 73
Fa (V, W )
functions in a neighborhood of a, 67
functor
contravariant, 30
covariant, 30
forgetful, 30
power set, 31
vector space duality, 31
fundamental quotient theorem, 35
G
ateaux
differential, 70
variation, 70
(the map from V to V taking x to x
b), 26
generalized eigenspaces, 52
geometic
progression, 13
geometric
product, 119, 122
tangent space, 87
tangent vector, 87
germs, 89
Gm (M ) (germs of functions at m), 89
grade
of a decomposable element, 79
of a homogeneous element, 79
graded algebra, 79
gradient, 100
at a point, 70
on an open set, 70
grad f (m) (gradient of f at m), 100
graph, 6
Grassmann algebra, 79
greatest common divisor, 47
Greek letters, 3
group, 1
Abelian, 1
cohomology, 106
de Rham cohomology, 105
simplicial cohomology, 110
simplicial homology, 109
symmetric, 73
H(A) (self-adjoint elements of A), 61
half-space
boundary of a, 114
interior of a, 114
half-space, upper, 114
Hermitian, 61
H k (M ) (de Rham cohomology group), 105
Hn (upper half-space), 114
Hodge star operator, 82
Hom(G, H)
Hom(G, H), Hom(G) (group homomorphisms), 6
homogeneous
elements of a graded algebra, 79
tensor, 76
INDEX
homology, 109
homomorphism
of Abelian groups, 6
of algebras, 40
unital, 40
of rings, 10
unital ring, 10
H p (K) (the pth simplicial cohomology group), 110
Hp (K) (the pth simplicial homology group), 109
hull, convex, 108
ideal
in an algebra, 41
left, 41
principal, 41
proper, 41
right, 41
trivial, 41
idempotent, 39, 43
IV (identity operator on a vector space), 23
idS (identity function on S), 7
identity
additive, 1, 10
function on a set, 7
multiplicative, 1
operator on a vector space, 23
resolution of the, 49
image, 22, 31
direct, 22
inverse, 22
imaginary
part of a -algebra element, 61
inclusion map, 7
independent, 15
convex, 108
indeterminant, 43
inequality
Schwarz, 56
infimum
of projections in a -algebra, 64
infinite dimensional, 19
infinitely differentiable, 84
injection, 6
injective, 6
inner product, 55
space, 55
spectral theorem for, 64
integrable
function, 13
integral
of a p -form over a p -chain, 113
integration
is linear, 21
interchange operation, 8
interior
of a half-space, 114
of a manifold with boundary, 114
internal
orthogonal direct sum, 57
interpolation
Lagrange, 46
INDEX
invariant subspace, 47
trivial, 47
inverse
additive, 1, 10
image, 22
in a Clifford algebra, 120
left, 30
multiplicative, 1
of a morphism, 30
right, 30
invertible, 30
element of an algebra, 41
left, 23
linear map, 23
right, 23
involution, 60
algebra with, 60
irreducible, 46
isometry, 127
isomorphic, 10, 40
isomorphism, 23
in a category, 30
of algebras, 40
of rings, 10
k -coboundary, 106
k -cochain, 106
k -cocycle, 106
kernel, 22
of a bilinear form, 126
of a ring homomorphism, 10
of an algebra homomorphism, 40
ker T (the kernel of T ), 22
k-form, 98
smooth, 99
Lat T , Lat T (collection of invariant subspaces), 48
L(V )
as a unital -algebra, 62
L(V, W ), L(V ) (family of linear functions), 21
Ln (V1 , . . . , Vn ) (family of n-linear functions), 73
l(S, A) (functions from S into A), 43
l(S, F), l(S) (F-valued functions on S), 9
lc (S, A) (functions from S into A with finite
support), 43
lc (S, F), lc (S) (functions on S with finite support), 9
l2 (R), l2 (C) (square summable sequences), 56
Lagrange interpolation formula, 46
largest, 17
leading coefficient, 44
left
-handed basis, 82
cancellable, 30
ideal, 41
inverse
of a morphism, 30
invertible linear map, 23
Leibnizs rule, 89
length
of a cycle, 73
137
of a vector, 56
linear, 21
combination, 15
formal, 33
trivial, 15
conjugate, 55
functional, 25
ordering, 18
sesqui-, 55
transformation
invertible, 23
transformations, 21
tensor products of, 78
linearly
dependent, 15
independent, 15
linearly ordered set, 18
local
coordinates, 84
representative, 85
locally
Euclidean, 84
smooth k-form, 99
locally constant, 105
lower
bound, 17
Mn = Mn (F) = Mmn (F), 11
Mn
Mn (A) (n n matrices of members of A), 74
Mmn (F)
magnitude
of a bivector, 120
manifold, 83
boundary of a, 114
differential, 84
smooth, 84
smoothly triangulated, 110
topological, 84
with boundary, 114
interior of a, 114
mapping
universal, 32
matrix
diagonal, 51
Pauli spin, 123
representation of an linear map, 24
similarity, 50
symmetric, 19
transpose of a, 19
maximal, 17
minimal, 17
polynomial, 45
existence of, 45
monic, 30
monic polynomial, 44
monoid, 1
monomorphism, 29
morphism
138
map, 30
Mor(S, T ) (morphisms from S to T ), 29
morphisms, 29
composition of, 29
mT (minimal polynomial for T ), 45
multilinear
form, 73
functional, 73
map, 73
multiplication
Clifford, 119
pointwise, 40
multiplicative
identity, 1
inverse, 1
N (set of natural numbers), 2
Nn (first n natural numbers), 2
N (A) (normal elements of A), 61
nilpotent, 51
n-linear function, 73
n-manifold, 84
nondegenerate bilinear form, 126
norm, 56
of a bivector, 120
uniform, 57
normal, 61
normalizing a vector, 58
normed
linear space, 56
vector space, see normed linear space
nullity, 22
nullspace, 22
numbers
special sets of, 2
object
free, 32
map, 30
objects, 29
odd
permutation, 73
odd function, 14
O(V, W ) (big-oh functions), 67
o(V, W ) (little-oh functions), 67, 86
(v) (a one-form acting on a vector field), 98
one-form
differential, 98
one-to-one, 6
one-to-one correspondence, 6
onto, 6
open
face of a simplex, 108
simplex, 108
unit disc, 2
operation
binary, 1
operator, 21
diagonalizable, 51
exterior differentiation, 99
nilpotent, 51
INDEX
projection, 39
similarity, 50
opposite orientation, 82
order
complete, 31
preserving, 29
ordered
basis, 82
by inclusion, 17
partially, 17
pointwise, 17
ordering
linear, 18
partial
of projections, 64
total, 18
orientation
of a simplex, 108
of a vector space, 82
oriented
simplex, 108
vector space, 82
orthogonal, 57
complement, 58
direct sum
projections associated with, 64
element of a real -algebra, 61
inner product projections, 63
projection, 63
associated with an orthogonal direct sum
decomposition, 64
projections
in a -algebra, 63
resolution of the identity, 64
with respect to a bilinear form, 126
orthonormal, 58
basis, 58
orthonormal, B-, 126
overlap maps, 83
P(A) (projections on a -algebra), 63
P(S) (power set of S), 31
P(J) (polynomial functions on J), 16
Pn (J) (polynomial functions of degree strictly less
than n), 16
P2 (projective plane), 85
parallelogram law, 57
parametrization
of a curve, 83
of a solid, 83
of a surface, 83
partial
ordering, 17
of projections, 64
Pauli spin matrices, 123
p -boundaries, 109
p -chains, 109
p -coboundary
simplicial, 110
p -cochain, 110
p -cocycle
INDEX
simplicial, 110
p -cycles, 109
permutation, 73
cyclic, 73
even, 73
odd, 73
sign of a, 73
perpendicular, 57
plane
of a simplex, 108
Poincares lemma, 102
point spectrum, 49
pointwise
addition, 5
multiplication, 40
partial ordering, 17
scalar multiplication, 11
polarization identity, 57
polyhedron, 109
polynomial, 43
algebra, 44
annihilating, 45
characteristic, 49
constant, 44
degree of a, 44
function, 44
functional calculus, 44
irreducible, 46
minimal, 45
existence of, 45
monic, 44
prime, 46
reducible, 46
standard form of a, 44
POSET
bijective morphism in, 30
the category, 30
POSET
the category, 29
power series
formal, 43
power set, 31
functor, 31
(p, q)-shuffle, 81
pre-annihilator, 27
preimage, 31
primary decomposition theorem, 52
prime
relatively, 47
prime polynomial, 46
principal ideal, 41
product
Clifford, 119, 122
exterior, 79
geometric, 119, 122
in a category, 33
inner, 55
tensor, 76
uniqueness of, 34
wedge, 79
139
progression
arithmetic, 13
geometic, 13
projection
along one subspace onto another, 39
assocuated with direct sum decomposition, 49
coordinate, 84
cotangent bundle, 98
in a -algebra, 63
operator, 39
orthogonal, 63
tangent bundle, 97
projections
difference of, 64
infimum of, 64
ordering of, 64
orthogonality of, 63
product of, 63
sum of, 63
supremum of, 64
projective plane, 85
proper
ideal, 41
pseudoscalar, 80, 119, 122
p -simplex, 108
pullback of differential forms, 110
Pythagorean theorem, 57
Q (set of rational numbers), 2
quadratic form, 125
associated with a bilinear form, 125
quotient
algebra, 42
map, 34, 35
object, 34
vector space, 35
quotient map, the, 42
R (set of real numbers), 2
Rn (set of n-tuples of real numbers), 2
range, 22
ran T (the range of T ), 22
rank, 22
rank-plus-nullity theorem, 35
real
line
special subsets of, 2
part of a -algebra element, 61
vector space, 10
reducible, 46
reducing subspaces, 47
reflection, 120
reflexive, 17
relation, 17
relatively prime, 47
representation
matrix, 24
representative, local, 85
resolution of the identity
in vector spaces, 49
orthogonal, 64
140
restriction, 7
retraction, 23
reverse orientation, 82
M (cotangent bundle projection), 98
Riesz-Frechet theorem
for inner product spaces, 59
right
-handed basis, 82
cancellable, 30
ideal, 41
inverse
of a morphism, 30
invertible linear map, 23
R
ring, 1
commutative, 1
division, 1
homomorphism, 10
unital, 10
with identity, 1
Rn
root, 47
row vector, 18
r -skeleton, 109
S1 (unit circle), 2
S1 (unit circle), 84
Sn (n-sphere), 85
scalar, 10, 80, 119, 122
field, 70
multiplication
pointwise, 11
Schwarz inequality, 56
section, 23, 97
segment, 5
segment, closed, 107
self-adjoint, 61
semigroup, 1
separates points, 26
sequence
bounded, 13
convergent, 13
decreasing, 13
exact, 35
series
formal power, 43
sesquilinear, 55
SET
the category, 29
shift
unilateral, 22
short exact sequence, 36, 107
shuffle, 81
(interchange operation), 8
(a), A (a) (the spectrum of a), 42
p (A), p (T ) (point spectrum), 49
INDEX
sign
of a permutation, 73
similar, 50
simple
algebra, 41
simplex
closed, 108
dimension of a, 108
face of a, 108
open, 108
oriented, 108
plane of a, 108
vertex of a, 108
simplicial
coboundary, 110
cocycle, 110
cohomology group, 110
complex, 108
dimension of a, 109
simplicial homology group, 109
skeleton of a complex, 109
skew-Hermitian, 61
skew-symmetric, 61, 74
smallest, 17
subspace containing a set of vectors, 13
smooth, 84
k-form, 99
atlas, 84
curve, 83
differential one-form, 98
function, 83
locally, 99
manifold, 84
map between manifolds, 85
solid, 83
submanifold, 110
surface, 83
triangulation, 110
vector field, 97
smoothly compatible
atlases, 84
charts, 84
solid, 83
smooth, 83
space
algebraic tangent, 89
geometric tangent, 87
inner product, 55
normed linear, 56
vector, 10
span, 15
of the empty set, 15
span(A) (the span of A), 15
spectral theorem
for complex inner product spaces, 64
spectrum, 42
point, 49
INDEX
spin matrices, 123

S(p, q) (shuffle permutations), 81
square summable sequence, 56
standard
basis curves at a point, 87
form of a polynomial, 44
representative
of a 2-blade, 122
of a vector, 120
standard basis
for Pn (J), 16
for Mmn , 17
for R3 , 16
for Rn , 16
star operator, 82
star-algebra, 60
star-homomorphism, 60
stereographic projection, 85
Stokes theorem, 115
structure
differential, 84
subalgebra, 41
unital, 41
submanifold, 110
subprojection, 64
subspace, 12
affine, 108
complementary, 14
invariant, 47
reducing, 47
sum
direct, 34, 80
Fourier, 58
internal direct, 14
summable
absolutely, 13
square, 56
supp(f ) (support of f ), 9
support, 9
finite, 9
supremum
of projections in a -algebra, 64
surface, 83
smooth, 83
surjection, 6
surjective, 6
switching operation, 8
symmetric
bilinear form, 125
element of a real -algebra, 61
group, 73
matrix, 19
skew-, 74
tangent
at 0, 86
at m, 86
bundle, 97
projection, 97
space
141
algebraic, 89
geometric, 87
vector
algebraic, 89
geometric, 87
tangent (at zero), 68
M (tangent bundle projection), 97
tensor
algebra, 80
decomposable, 76
elementary, 76
homogeneous, 76
product, 76
of linear maps, 78
T -invariant subspace, 47
T M (tangent bundle), 97
T M (cotangent bundle), 98
Tm (tangent space at m), 93
Tbm (M ) (algebraic tangent space at m), 89
Tem (M ) (geometric tangent space at m), 87
topological
manifold, 84
total
ordering, 18
transition maps, 83
transitive
algebra, 48
propertty of a partial ordering, 17
transpose, 61, 75
of a matrix, 19
transposition, 73
triangulated manifold, 110
trilinear map, 73
trivector, 122
trivial
Lat T or Lat T, 48
ideal, 41
invariant subspaces, 47
linear combination, 15
vector space, 10
T (V ) (the tensor algebra of V ), 80
U(A) (unitary elements of A), 61
uniform
norm, 57
unilateral shift, 22
is adjointable on l2 , 61
unique factorization theorem, 47
uniqueness
essential, 34
of products, coproducts, 34
unit, 1
vector, 56
unital
algebra, 40
algebra homomorphism, 40
ring, 1
ring homomorphism, 10
subalgebra, 41
unitarily
142
diagonalizable, 64
equivalent, 64
unitary, 61
unitization, 42
universal, 125
mapping
diagram, 32
property, 32
upper
bound, 17
half-space, 114
V (dual space of V ), 25
v (= e1 v), 121
vk (parallel component of v), 120
v (perpendicular component of v), 120
variation
G
ateaux, 70
vector, 10, 119, 122
algebraic tangent, 89
column, 18
decomposition theorem, 14
field, 97
geometric tangent, 87
norm of a, 56
row, 18
space, 10
adjoint map, 31
complex, 10
dimension of a, 19
free, 32
normed, 56
orientation of a, 82
oriented, 82
real, 10
spectral theorem for, 51
trivial, 10
unit, 56
VEC
the category, 29
vertex
of a simplex, 108
vf (action of a vector field on a function), 97
vol (volume element or form), 82
volume
element, 82
wedge product, 79
Z (set of integers), 2
zero
divisor, 9
form, 98
Z+ -graded algebra, 79
Z k (M ) (closed k-forms), 105
Zorns lemma, 18
Z p (K) (space of simplicial p -cocycles), 110
Zp (K) (space of simplicial p -cycles), 109
INDEX

Multilinear Algebra PDF

Uploaded by

Copyright:

Available Formats

Multilinear Algebra PDF

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Multilinear Algebra PDF

Uploaded by

Copyright:

Available Formats

ELEMENTS OF LINEAR AND MULTILINEAR

E-mail address: erdman@pdx.edu

NOTATION AND TERMINOLOGY

Chapter 1. VECTOR SPACES

Chapter 2. LINEAR TRANSFORMATIONS

Chapter 3. THE LANGUAGE OF CATEGORIES

Chapter 4. The Spectral Theorem for Vector Spaces

The Spectral Theorem - Vector Space Version

Chapter 5. THE SPECTRAL THEOREM FOR INNER PRODUCT SPACES

Chapter 6. A BRIEF REVIEW OF DIFFERENTIAL CALCULUS

Chapter 7. MULTILINEAR MAPS AND DETERMINANTS

Chapter 8. TENSOR ALGEBRAS

Chapter 9. DIFFERENTIAL MANIFOLDS

10. DIFFERENTIAL FORMS ON MANIFOLDS

11. HOMOLOGY AND COHOMOLOGY

Chapter 12. STOKES THEOREM

Chapter 13. GEOMETRIC ALGEBRA

14. CLIFFORD ALGEBRAS

NOTATION AND TERMINOLOGY

(S, +) is a semigroup if it satisfies axioms (1)(2).

NOTATION AND TERMINOLOGY

English name (approximate pronunciation)

NOTATION AND TERMINOLOGY

1.1.11. Definition. Let G and H be Abelian groups. A map f : G H is a homomorphism if

1.2. FUNCTIONS AND DIAGRAMS

commute, then the square also commutes.

(Here f and g are the obvious bijections.) What is (S, a, )?

commutes. What can be said about the function f ?

then f is a unital (ring) homomorphism.

1.4. VECTOR SPACES

for all R and x V . Prove or disprove: under the operations and

the set V is a vector

for all R and x V . Prove or disprove: under the operations and

the set V is a vector

the set V is a vector

k=1 |xk | < .)

P = P[a, b] = {p : p is a polynomial function on [a, b]}

P4 = P4 [a, b] = {p P : the degree of p is less than 4}

Q4 = Q4 [a, b] = {p P : the degree of p is equal to 4}

K = K[a, b] = {f F : f is a constant function}

1.6. LINEAR COMBINATIONS AND LINEAR INDEPENDENCE

1.6.1. Definition. A vector y P

1.7. BASES FOR VECTOR SPACES

(1) Elements x and y in S are comparable if either x y or y x.

The notation can be unambiguously simplified by writing

1.7. BASES FOR VECTOR SPACES

commutes for every F. What can be said about the homomorphism T ?

C and x [a, b].

2.2. INVERTIBLE LINEAR MAPS

2.2. Invertible Linear Maps

2.3. Matrix Representations

(c) Independently of part (b) calculate the matrix product [T ]

2.5. DUAL SPACES

2.4.4. Proposition. Let T : V W be an injective linear map between vector spaces. If B is a

Why do you think things are done this way?

terms of the standard dual basis for R2 .)

THE LANGUAGE OF CATEGORIES

3. THE LANGUAGE OF CATEGORIES

/ C is left cancellable if whenever morphisms A

3.7.3. Proposition. Let W be a vector space and M V W . Then

4.6.12. Example. Let T be the operator on the complex