Introduction to TOPOLOGY AND MODERN ANALYSIS GEORGE F. SIMMONS Associate Professor of Mathematics Colorado College © ROBERT E. KRIEGER PUBLISHING COMPANY MALABAR, FLORIDAFor Virgie May Hatcher and Elizabeth B. Blossom TO EACH OF WHOM | OWE MORE THAN | CAN POSSIBLY EXPRESS Original Edition 1963 Reprint Edition 1983 Printed and Published by ROBERT E. KRIEGER PUBLISHING COMPANY, INC. KRIEGER DRIVE MALABAR, FLORIDA 32950 Copyright © 1963 by McGraw-Hill, Inc. Reprinted by arrangement All rights reserved. No part of this book may be reproduced in any Sorm or by any electronic or mechanical means including information storage and retrieval systems without permission in writing from the publisher. Printed in the United States of America Library of Congress Cataloging in Publication Data Simmons, George Finlay, 1925- Introduction to topology and modern analysis. Reprint. Originally published: New York: McGraw- Hill, 1963 (International series in pure and applied mathematics) Bibliography: p. Includes index. 1, Topology. , 2. Mathematical analysis. 1. Title. IL. Title: Topology and modern analysis. IIL. Series: International series in pure and QAGIL.S49 1983 34 82-1484, ISBN 0-89874—551-9 098765Preface For some time now, topology has been firmly established as one of the basic disciplines of pure mathematics. Its ideas and methods have transformed large parts of geometry and analysis almost beyond recogni- tion. It has also greatly stimulated the growth of abstract algebra. As things stand today, much of modern pure mathematics must remain a closed book to the person who does not acquire a working knowledge of at least the elements of topology. There are many domains in the broad field of topology, of which the following are only a few: the homology and cohomology theory of com- plexes, and of more general spaces as well; dimension theory; the theory of differentiable and Riemannian manifolds and of Lie groups; the theory of continuous curves; the theory of Banach and Hilbert spaces and their operators, and of Banach algebras; and abstract harmonic analysis on locally compact groups. Each of these subjects starts from roughly the same body of fundamental knowledge and develops its own methods of dealing with its own characteristic problems. The purpose of Part 1 of this book is to make available to the student this “hard core” of funda- mental topology; specifically, to make it available in a form which is general enough to meet the needs of modern mathematics, and yet is unburdened by excess baggage best left in the research journals. A topological space can be thought of as a set from which has been swept away all structure irrelevant to the continuity of functions defined on it. Part 1 therefore begins with an informal (but quite extensive) treatment of sets and functions. Some writers deal with the theory of metric spaces as if it were merely a fragment of the general theory of topological spaces. This practice is no doubt logically correct, but it seems to me to violate the natural relation between these topics, in which metric spaces motivate the more general theory. Metric spaces are therefore discussed rather fully in Chapter 2, and topological spaces are introduced in Chapter 3. The remaining four chapters in Part 1 are concerned with various kinds of topological spaces of special importance in applications and with the continuous functions carried by them. It goes without saying that one aspect of this type of mathematics is its logical precision. Too many writers, however, are content with this, and make little effort to help the reader maintain his orientation in vilviii Preface the midst of mazes of detail. One of the main features of this book is the attention given to motivating the ideas under discussion. On every possible occasion I have tried to make clear the intuitive meaning of what is taking place, and diagrams are provided, whenever it seems feasible, to help the reader develop skill in using his imagination to visualize abstract ideas. Also, each chapter begins with a brief introduction which describes its main theme in general terms. Courses in topology are being taught more and more widely on the undergraduate level in our colleges and universities, and I hope that these features, which tend to soften the austere framework of definitions, theorems, and proofs, will make this book readable and easy to use as a text. Historically speaking, topology has followed two principal lines of development. In homology theory, dimension theory, and the study of manifolds, the basic motivation appears to have come from geometry. In these fields, topological spaces are looked upon as generalized geometric configurations, and the emphasis is placed on the structure of the spaces themselves. In the other direction, the main stimulus has been analysis. Continuous functions are the chief objects of interest here, and topological spaces are regarded primarily as carriers of such functions and as domains over which they can be integrated. These ideas lead naturally mto the theory of Banach and Hilbert spaces and Banach algebras, the modern theory of integration, and abstract harmonic analysis on locally compact groups. In Part 1 of this book, I have attempted an even balance between these two points of view. This part is suitable for a basic semester course, and most of the topics treated are indispensable for further study in almost any direction. If the instructor wishes to devote a second semester to some of the extensions and applications of the theory, many possibilities are open. If he prefers applications in modern analysis, he can continue with Part 2 of this book, supplemented, perhaps, with a brief treatment of measure and integration aimed at the general form of the Riesz representation theorem. Or if his tastes incline him toward the geometric aspects of topology, he can switch over to one of the many excellent books which deal with these matters. The instructor who intends to continue with Part 2 must face a question which only he can answer. Do his students know enough about algebra? This question is forced to the surface by the fact that Chapters 9 to 11 are as much about algebra as they are about topology and analy- sis. If his students know little or nothing about modern algebra, then a careful and detailed treatment of Chapter 8 should make it possible to proceed without difficulty. And if they know a good deal, then a quick survey of Chapter 8 should suffice. It is my own opinion that education in abstract mathematics ought to begin on the junior level with a course in modern algebra, and that topology should be offered only to studentsPreface ix who have acquired some familiarity, through such a course, with abstract methods. Part 3 is intended for individual study by exceptionally well-qualified students with a reasonable knowledge of complex analysis. Its principal purpose is to unify Parts 1 and 2 into a single body of thought, along the lines mapped out in the last section of Chapter 11. Taken as a whole, the present work stands at the threshold of the more advanced books by Rickart [34], Loomis (27], and Naimark [32]; and much of its subject matter can be found (in one form or another and with innumerable applications to analysis) in the encyclopedic treatises of Dunford and Schwartz (8] and Hille and Phillips [20].!_ This book is intended to be elementary, in the sense of being accessible to well-trained undergraduates, while those just mentioned are not, Its prerequisites are almost negligible. Several facts about determinants are used without proof in Chapter 11, and Chapter 12 leans heavily on Liouville’s theorem and the Laurent expansion from complex analysis. With these excep- tions, the book is essentially self-contained. It seems to me that a worthwhile distinction can be drawn between two types of pure mathematics. The first—which unfortunately is somewhat out of style at present—centers attention on particular func- tions and theorems which are rich in meaning and history, like the gamma function and the prime number theorem, or on juicy individual facts, like Euler’s wonderful formula 14M+wt--- = 9/6. The second is concerned primarily with form and structure. The present. book belongs to this camp; for its dominant theme can be expressed in just two words, continuity and linearity, and its purpose is to illuminate the meanings of these words and their relations to each other. Mathe- matics of this kind hardly ever yields great and memorable results like the prime number theorem and Euler’s formula, On the contrary, its theorems are generally small parts of a much larger whole and derive their main significance from the place they occupy in that whole. In my opinion, if a body of mathematics like this is to justify itself, it must possess aesthetic qualities akin to those of a good piece of architecture. It should have a solid foundation, its walls and beams should be firmly and truly placed, each part should bear a meaningful relation to every other part, and its towers and pinnacles should exalt the mind. It is my hope that this book can contribute to a wider appreciation of these mathe- matical values, George F, Simmons 1 The numbers in brackets refer to works listed in the Bibliography.A Note to the Reader ‘Two matters call for special comment: the problems and the proofs. The majority of the problems are corollaries and extensions of theorems proved in the text, and are freely drawn upon at all later stages of the book. In general, they serve as a bridge between ideas just treated and developments yet to come, and the reader is strongly urged to master them as he goes along. In the earlier chapters, proofs are given in considerable detail, in an effort to smooth the way for the beginner. As our subject unfolds through the successive chapters and the reader acquires experience in following abstract mathematical arguments, the proofs become briefer and minor details are more and more left for the reader to fill in for himself. The serious student. will train himself to look for gaps in proofs, and should regard them as tacit invitations to do a little thinking on his own. Phrases like “it is easy to see,” “‘one can easily show,” “evidently,” “clearly,” and so on, are always to be taken as warning signals which indicate the presence of gaps, and they should put the reader on his guard, It is @ basic principle in the study of mathematics, and one too seldom emphasized, that a proof is not really understood until the stage is reached at which one can grasp it as a whole and see it as a single idea, In achieving this end, much more is necessary than merely follow- ing the individual steps in the reasoning. This is only the beginning. A proof should be chewed, swallowed, and digested, and this process of assimilation should not be abandoned until it yields a full comprehension of the overall pattern of thought.Contents Preface vii A Note to the Reader xi PART ONE: TOPOLOGY Chapter One SETS AND FUNCTIONS 3 Sets and set inclusion 3 The algebra of sets 7 Functions 14 Products of sets 21 Partitions and equivalence relations 25 Countable sets 31 . Uncountable sets 36 Partially ordered sets and lattices 43 exes eee Chapter Two METRIC SPACES 49 9. The definition and some examples 51 10. Open sets 59 11, Closed sets 65 12, Convergence, completeness, and Baire’s theorem 70 13. Continuous mappings 75 14, Spaces of continuous functions 80 15, Euclidean and unitary spaces 85 Chapter Three TOPOLOGICAL SPACES 91 16. The definition and some examples 92 17. Elementary concepts 95 18. Open bases and open subbases 99 19. Weak topologies 104 20, The function algebras €(X,R) and e(X,C) 106 Chapter Four COMPACTNESS 110 21, Compact spaces 111 22, Products of spaces 115 xlxiv Contents 23, Tychonofi’s theorem and locally compact spaces 118 24, Compactness for metric spaces 120 25. Ascoli’s theorem 124 Chapter Five SEPARATION 26. Tr-spaces and Hausdorff spaces 130 27, Completely regular spaces and normal spaces 132 28. Urysohn’s lemma and the Tietze extension theorem 135 29. The Urysohn imbedding theorem 137 30. The Stone-Cech compactification 139 Chapter Sit CONNECTEDNESS 31. Connected spaces 143 32. The components of a space 146 33, Totally disconnected spaces 149 34, Locally connected spaces 150 Chapter Seven APPROXIMATION 35. The Weierstrass approximation theorem 153 36. The Stone-Weierstrass theorems 157 37. Locally compact Hausdorff spaces 162 38. The extended Stone-Weierstrass theorems 165 PART TWO: Chapter Hight ALGEBRAIC SYSTEMS 39, Groups 172 40. Rings 181 41. The structure of rings 184 42, Linear spaces 191 43. The dimension of a linear space 196 44, Linear transformations 203 45, Algebras 208 Chapter Nine BANACH SPACES 46, The definition and some examples 212 47, Continuous linear transformations 219 48, The Hahn-Banach theorem 224 49, The natural imbedding of Nin N** 231 50. The open mapping theorem 235 51. The conjugate of an operator 239 Chapter Ten HILBERT SPACES 52. The definition and some simple properties 244 53. Orthogonal complements 249 54, Orthonormal sets 251 129 142 153 OPERATORS 171 211 243Contents xv 55. The conjugate space H* 260 56. The adjoint of an operator 262 57. Self-adjoint operators 266 58, Normal and unitary operators 269 59. Projections 273 Chapter Eleven FINITE-DIMENSIONAL SPECTRAL THEORY 278 60. Matrices 280 61. Determinants and the spectrum of an operator 287 62. The spectral theorem 290 63. A survey of the situation 295 PART THREE: ALGEBRAS OF OPERATORS Chapter Twelve GENERAL PRELIMINARIES ON BANACH ALGEBRAS 301 64. The definition and some examples 302 65. Regular and singular elements 305 66. Topological divisors of zero 307 67. The spectrum 308 68, The formula for the spectral radiue 312 69. The radical and semi-simplicity 313 Chapter Thirteen THE STRUCTURE OF COMMUTATIVE BANACH ALGEBRAS 318 70. The Gelfand mapping 318 71. Applications of the formula r(z) = lim lz" 323 72, Involutions in Banach algebras 324 73. The Gelfand-Neumark theorem 325 Chapter Fourteen SOME SPECIAL COMMUTATIVE BANACH ALGEBRAS 327 74, Ideals in €(X) and the Banach-Stone theorem 327 75. The Stone-Cech compactification (continued) 330 76. Commutative C*-algebras 332 APPENDICES ONE Fixed point theorems and some applications to analysis 337 TWO Continuous curves and the Hahn-Mazurkiewics theorem 341 THREE Boolean algebras, Boolean rings, and Stone’s theorem 344 Bibliography 355 Index of Symbols 359 Subject Index 363PART ONE CopologyCHAPTER ONE Sets and Functions It is sometimes said that mathematics is the study of sets and func- tions. Naturally, this oversimplifies matters; but it does come as close to the truth as an aphorism can. The study of sets and functions leads two ways. One path goes down, into the abysses of logic, philosophy, and the foundations of mathematics. The other goes up, onto the highlands of mathematics itself, where these concepts are indispensable in almost all of pure mathe- matics as it is today. Needless to say, we follow the latter course. We regard sets and functions as tools of thought, and our purpose in this chapter is to develop these tools to the point where they are sufficiently powerful to serve our needs through the rest of this book. As the reader proceeds, he will come to understand that the words set and function are not as simple as they may seem. In a sense, they are simple; but they are potent words, and the quality of simplicity they possess is that which lies on the far side of complexity. They are like seeds, which are primitive in appearance but have the capacity for vast and intricate development. 1, SETS AND SET INCLUSION We adopt a naive point of view in our discussion of sets and assume that the concepts of an element and of a set of elements are intuitively clear. By an element we mean an object or entity of some sort, as, for example, a positive integer, a point on the real line (= a real number), 34 Topology or a point in the complex plane (= a complex number). A set is a collec- tion or aggregate of such elements, considered together or as a whole. Some examples are furnished by the set of all even positive integers, the set of all rational points on the real line, and the set; of all points in the complex plane whose distance from the origin is 1 (= the unit circle in the plane). We reserve the word class to refer to a set of sets. We might speak, for instance, of the class of all circles in a plane (thinking of each circle as a set of points). It will be useful in the work we do if we carry this hierarchy one step further and use the term family for a set of classes. One more remark: the words element, set, class, and family are not intended to be rigidly fixed in their usage; we use them fluidly, to express varying attitudes toward the mathematical objects and systems we study. It is entirely reasonable, for instance, to think of a circle not as a set of points, but as a single entity in itself, in which case we might justifiably speak of the set of all circles in a plane. There are two standard notations available for designating a par- ticular set. Whenever it is feasible to do so, we can list its elements between braces. Thus {1, 2, 3} signifies the set consisting of the first three positive integers, {1, 1, —1, —1} is the set of the four fourth roots of unity, and {+1, +3, +5, . . .} is the set of all odd integers. This manner of specifying a set, by listing its elements, is unworkable in many circumstances. We are then obliged to fall back on the second method, which is to use a property or attribute that characterizes the elements of the set in question. If P denotes a certain property of elements, then {z:P} stands for the set of all elements x for which the property P is meaningful and true. For example, the expression {x:a is real and irrational}, which we read the set of all x such that x is real and irrational, denotes the set of all real numbers which cannot be written as the quotient of two integers. The set under discussion contains all those elements (and no others) which possess the stated property. The three sets of numbers described at the beginning of this paragraph can be written either way: {1, 2, 3} = {n:n is an integer and 0 < n < 4}, {1, 4, -1, —7} is a complex number and z‘ = 1}, and {+1, £3, +5, ...} = {n:nis an odd integer}. We often shorten our notation. For instance, the last. two sets mentioned might perfectly well be written {z:z4 = 1} and {n:nis odd}. Our pur- pose is to be clear and to avoid misunderstandings, and if this can be achieved with less notation, so much the better. In the same vein we canSets and Functions 5 write the unit circle = {z:|z| = 1}, the closed unit dise = {z:|z| < 1}, and the open unit disc = {z:|z| < 1}. We use a special system of notation for designating intervals of various kinds on the real line. If @ and 6} are real numbers such that a < 6, then the following symbols on the left are defined to be the indicated sets on the right: [a,b] = {za Sx <5}, (a,b] = {z:a A and BD A. It will often be convenient to have a symbol for logical implication, and = is the symbol we use. If p and q are statements, then p> q@ means that p implies q, or that if p is true, then q is also true. Similarly, « is our symbol for two-way implication or logical equivalence. It means that the statement on each side implies the statement on the other, and is usually read #f and only if, or is equivalent to. The main properties of set inclusion are obvious. They are the following: (1) ACA for every A; (2) ACBand BCAA =B; (3) ACBandBOCSACC. It is quite important to observe that (1) and (2) can be combined into the single statement that A = B@ACB and BCA. This remark contains a useful principle of proof, namely, that the only way to show that two sets are equal, apart from merely inspecting them, is to show that each is a subset of the other. Problems 1. Perhaps the most famous of the logical difficulties referred to in the text is Russell’s parador. To explain what this is, we begin by observing that a set can easily have elements which are themselves sets, e.g., {1, {2,3}, 4}. This raises the possibility that a set might well contain itself as one of its elements. We call such a set an abnormal set, and any set which does not contain itself as an element we call a normal set. Most sets are normal, and if we suspect that. abnormal sets are in some way undesirable, we might try to confine our attention to the set NV of all normal sets. Someone is now sure to ask, Is N itself normal or abnormal? It is evidently one or the other, and it cannot be both. Show that if N is normal, then it must be abnormal. Show also that if N is abnormal, then it must be normal. We see in this way that each of our two alternatives is self-contradictory, and it seems to be the assumption that N exists as a set which has brought us to this impasse. For further discussion of these matters, we refer the interested reader to Wilder [42, p. 55]Sets and Functions = 7 or Fraenkel and Bar-Hillel [10, p. 6]. Russell’s own account of the discovery of his paradox can be found in Russell [36, p. 75]. The symbol we have used for set inclusion is similar to that used for the familiar order relation on the real line: if z and y are real numbers, 2 2 <2. It also has an important additional property: (4) for any z andy, either z < yor y < z. Property (4’) says that any two real numbers are comparable with respect to the relation in question, and it leads us to call the order relation on the real line a total (or linear) order relation. Show by an example that this property is not possessed by set inclusion. It is for this reason that set inclusion is called a partial order relation. (a) Let U be the single-element set {1}. There are two subsets, the empty set § and {1} itself. If A and B are arbitrary subsets of U, there are four possible relations of the form A € B. Count the number of true relations among these. (b) Let U be the set {1,2}. There are four subsets. List them. If A and B are arbitrary subsets of U, there are 16 possible relations of the form A C B. Count the number of true ones. (c) Let U be the set {1, 2, 3}. There are 8 subsets. What are they? There are 64 possible relations of the form A C B. Count the number of true ones. (d) Let U be the set {1, 2,..., m} for an arbitrary positive integer n. How many subsets are there? How many possible relations of the form A C B are there? Can you make an informed guess as to how many of these are true? 2. THE ALGEBRA OF SETS In this section we consider several useful ways in which sets can be combined with one another, and we develop the chief properties of these operations of combination. As we emphasized above, all the sets we mention in this section are assumed to be subsets of our universal set U. U is the frame of reference, or the universe, for our present discussions. In our later work the frame of reference in a particular context will naturally depend on what ideas we happen to be considering. If we find ourselves studying sets of real8 Topology numbers, then U is the set R of all real numbers. If we wish to study sets of complex numbers, then we take U to be the set C of all complex numbers. We sometimes want to narrow the frame of reference and to consider (for instance) only subsets of the closed unit interval [0,1], or of the closed unit dise {z:|z| < 1}, and in these cases we choose U accordingly. Generally speaking, the universal set U is at our disposal, and we are free to select it to fit the needs of the moment. For the présent, however, U is to be regarded as a fixed but arbitrary set. This generality allows us to apply the ideas we develop below to any situation which arises in our later work. Fig. 1. Set inclusion. Fig. 2. The union of A and B. It is extremely helpful to the imagination to have a geometric picture available in terms of which we can visualize sets and operations on sets. A convenient way to accomplish this is to represent U by a rectangular area in a plane, and the elements which make up U by the points of this area. Sets can then be pictured by areas within this rectangle, and diagrams can be drawn which illustrate operations on sets and relations between them. For instance, if A and B are sets, then Fig. 1 represents the circumstance that A is a subset of B (we think of each set as consisting of all points within the corresponding closed curve). Diagrammatic thought of this kind is admittedly loose and imprecise; nevertheless, the reader will find it invaluable. Nomathematics, however abstract it may appear, is ever carried on without the help of mental images of some kind, and these are often nebulous, personal, and diffi- cult to describe. The first operation we discuss in the algebra of sets is that of forming unions. The union of two sets A and B, written A U B, is defined to be the set of all elements which are in either A or B (including those which are in both). A U B is formed by lumping together the elements of A and those of B and regarding them as constituting a single set. In Fig. 2, 4 UB is indicated by the shaded area. The aboveSets and Functions 9 definition can also be expressed symbolically: AVB = {z:z¢A or zé B}. The operation of forming unions is commutative and associative: AUB=BUA and AU(BUC)=(AUB)UC. It has the following additional properties: AVA=A,AUG=A,andAUU=U, We also note that AGCBoAUB=B, so set inclusion can be expressed in terms of this operation. Our next operation is that of forming intersections. The inter- section of two sets A and B, written A ( B, is the set of all elements which are in both A and B. In symbols, AB = {a:a¢ A and xe B}. A(\B is the common part of the sets A and B. In Fig. 3, AM Bis represented by the shaded area. If AQ Bis non-empty, we express this by saying that A intersects B. If, on the other hand, it happens that A and B have no common part, or equivalently that A B = G, then Fig. 3. The intersection of A and B. we say that A does not intersect B, or that A and B are disjoint; and a class of sets in which all pairs of distinct sets are disjoint is called a disjoint class of sets. The operation of form- ing intersections is also commutative and associative: ANB=BONA and AN(BNC)=(ANB)NC. It has the further properties that ANA=A,ANG6=§$,and ANU = A; and since ACBeANB=A, we see that set inclusion can also be expressed in terms of forming intersections. We have now defined two of the fundamental operations on sets, and we have seen how each is related to set inclusion. The next obvious step is to see how they are related to one another. The facts here are given by10 Topology the distributive laws: AN (BUC) = (ANB)U(ANO) and AU(BNC) = (AUB) N(AUC). ‘These properties depend only on simple logic applied to the meanings of ttf ily / Fig. 4. A U(BNC) = (A UB) N(A UC). the symbols involved. For instance, the first of the two distributive laws says that an element is in A and is in B or C precisely when it is in A and Bor isin A and C. We can convince ourselves intuitively of the validity of these laws by drawing pictures. The second distributive law is illustrated in Fig. 4, where A U (B 1 C) is formed on the left by shading and (A U B)(\ (A UC) on the right by cross-shading. A moment’s Stare ai baal a < grams ought to convi th di that one obtains the same set in each case. Fig. 5. The complement of A. those which make up JU, it goes with- out saying—but it ought to be said The last of our major operations on sets is the formation of comple- —that A’ consists of all those elements in U which are not in A. Symbolically, ments. The complement of a set A, A’ = {xing A}. Figure 5 (in which A’ is shaded) illustrates this operation. The operation of forming complements has the following obvious properties: (A'’ = 4, 8 = U, U' = AUVA'=U,and AN A' =.Sets and Functions 11 Further, it is related to set inclusion by ACBeBCA’ and to the formation of unions and intersections by (AUB) =A’ OB and (AM BY = A’UB’. (1) The first equation of (1) says that an element is not in either of two sets precisely when it is outside of both, and the second says that it is not in both precisely when it is outside of one or the other. The operations of forming unions and intersections are primarily binary operations; that is, each is a process which applies to a pair of sets and yields a third. We have emphasized this by our use of parentheses to indicate the order in which the operations are to be performed, as in (A, U As) U As, where the parentheses direct us first to unite A; and A, and then to unite the result of this with A3. Associativity makes it possible to dispense with parentheses in an expression like this and to write Ai\U A; U A;, where we understand that these sets are to be united in any order and that the order in which the operations are performed is irrelevant. Similar remarks apply to Ai A: As. Furthermore, if {Ai, 42, ..., An} is any finite class of sets, then we can form AL\UA21U+++UAn and Ai NAI1N +++ OAs in much the same way without any ambiguity of meaning whatever. In order to shorten the notation, we let J = {1, 2,... , n} be the set of subscripts which index the sets under consideration. J is called the index set. ‘We then compress the symbols for the union and intersection just mentioned to Usar Ai and Mr Ai. As long as it is quite clear what the index set is, we can write this union and intersection even more briefly, in the form U;A; and (;A,. For the sake of both brevity and clarity, these sets are often written UZ, A; and Mf, Ai These extensions of our ideas and notations don’t reach nearly far enough. It is often necessary to form unions and intersections of large (really large!) classes of sets. Let {A;} be an entirely arbitrary class of sets indexed by a set J of subscripts. Then User Ay = {u:2 ¢ A; for at least one ¢ ¢ 7} and Over Ag = {ata € A; for every i€ I} define their union and intersection. As above, we usually abbreviate these notations to U:A; and (,A,; and if the class {A,} consists of a sequence of sets, that is, if {As} = (Ai, As, As, ...}, then their union and intersection are often written in the form UZ, A;and (2, Ai Observe that we did not require the class {A,} to be non-empty. If it does12 Topology happen that this class is empty, then the above definitions give (remem- bering that all sets are subsets of U) U.A; = § and M4; = U. The second of these facts amounts to the following statement: if we require of an element that it belong to each set. in a given class, and if there are no sets present in the class, then every element satisfies this requirement. If we had not made the agreement that the only elements under consider- ation are those in U, we would not have been able to assign a meaning to the intersection of an empty class of sets. A moment’s consideration makes it clear that Eqs. (1) are valid for arbitrary unions and intersections: (UA)! = OA! and (NiAy’ = UsAY. (2) It is instructive to verify these equations for the case in which the class { A.} is empty. We conclude our treatment of the general theory of sets with a brief discussion of certain special classes of sets which are of consider- able importance in topology, logic, and measure theory. We usually denote classes of sets by capital letters in boldface. First, some general remarks which will be useful both now and later, especially in connection with topological spaces. We shall often have occasion to speak of finite unions and finite intersections, by which we mean unions and intersections of finite classes of sets, and by a finite class of sets we always mean one which is empty or consists of n sets for some positive integer n. If we say that a class A of sets is closed under the formation of finite unions, we mean that A contains the union of each of its finite subclasses; and since the empty subclass qualifies as a finite subclass of A, we see that its union, the empty set, is necessarily an element of A. In the same way, a class of sets which is closed under the formation of finite intersections necessarily contains the universal set. Now for the special classes of sets mentioned above. For the remainder of this section we specifically assume that the universal set U is non-empty. A Boolean algebra of sets is a non-empty class A of subsets of U which has the following properties: (@) AandBeAS>AUBEA; (2) AandBeA>AMBeEA; (3) AcAsd4’ed. Since A is assumed to be non-empty, it must contain at least one set A. Property (3) shows that A’ is in A along with A, and since A A’ = 6 and A U A’ = JU, (1) and (2) guarantee that A contains the empty set and the universal set. Since the class consisting only of the empty set and the universal set is clearly a Boolean algebra of sets, these two distinct sets are the only ones which every Boolean algebra of sets mustSets and Functions = 13 contain. It is equally clear that the class of all subsets of U is also a Boolean algebra of sets. There are many other less trivial kinds, and their applications are manifold in fields of study as diverse as statistics and electronics. Let A be a Boolean algebra of sets. It is obvious that if {A1, 42, ... , 4a} is a non-empty finite subclass of A, then AiVUA2U+++UAn, and Ai MA2N +++ OAs are both sets in A; and since A contains the empty set and the universal set, it is easy to see that A is a class of sets which is closed under the formation of finite unions, finite intersections, and complements. We now go in the other direction, and let A be a class of sets which is closed under the formation of finite unions, finite intersections, and comple- ments. By these assumptions, A automatically contains the empty set and the universal set, so it is non-empty and is easily seen to be a Boolean algebra of sets. We conclude from these remarks that Boolean algebras of sets can be described alternatively as classes of sets which are closed under the formation of finite unions, finite intersections, and comple- ments. It should be emphasized once again that when discussing Boolean algebras of sets we always assume that the universal set is non- empty. One final comment. We speak of Boolean algebras of sets because there are other kinds of Boolean algebras than those which consist of sets, and we wish to preserve the distinction. We explore this topic further in our Appendix on Boolean algebras. Problems 1. If {Ai} and {B;} are two classes of sets such that {Aj C {B;}, show that U.A. C U;B; and NB; © MAG. 2. The difference between two sets A and B, denoted by A — B, is the set of all elements in A and not in B; thus A — B = AM B’. Show the following: A-~B=4A-~(AMB)=(AUB)-B; (A - B)-C=A-(BUC); A-(B-C) (AUB)-C=(A-C)UB-O); A-(BUC)=(A-B)N(A-0C). 3. The symmetric difference of two sets A and B, denoted by AAB, is defined by A AB = (A — B) U (B — A); it is thus the union of14. Topology their differences in opposite orders. Show the following: AA(BAC) = (AAB)AC; AAG = A;ADA =; AAB=BAA; AM (BAC) = (ANB)A(ANC). 4. A ring of sets is a non-empty class A of sets such that if A and B are in A, then A AB and A‘) B arealsoin A. Show that A must also contain the empty set, AU B, and A — B. Show that if a non- empty class of sets contains the union and difference of any pair of its sets, then it is a ring of sets. Show that a Boolean algebra of sets is a ring of sets. 5. Show that the class of all finite subsets (including the empty set) of an infinite set is a ring of sets but is not a Boolean algebra of sets. 6. Show that the class of all finite unions of closed-open intervals on the real line is a ring of sets but is not 2 Boolean algebra of sets. 7. Assuming that the universal set U is non-empty, show that Boolean algebras of sets can be described as rings of sets which contain U. 3. FUNCTIONS Many kinds of functions occur in topology, in a great variety of situations. In our work we shall need the full power of the general con- cept of a function, and since its modern meaning is much broader and deeper than its elementary meaning, we discuss this concept in con- siderable detail and develop its main abstract properties. Let us begin with a brief inspection of some simple examples. Con- sider the elementary function yor of the real variable x. What do we have in mind when we call this a function and say that y is a function of z? In a nutshell, we are drawing attention to the fact that each real number z has linked to it a specific real number y, which can be calculated according to the rule (or law of correspondence) given by the formula. We have here a process which, applied to any real number z, does something to it (squares it) to produce another real number y (the square of x). Similarly, y=o?—3c and y= (2? +1) are two other simple functions of the real variable z, and each is given by a rule in the form of an algebraic expression which specifies the exact, manner in which the value of y depends on the value of =.Sets and Functions 15 The rules for the functions we have just mentioned are expressed by formulas. In general, this is possible only for functions of a very simple kind or for those which are sufficiently important to deserve special symbols of their own. Consider, for instance, the function of the real variable z defined as follows: for each real number 2, write z as an infinite decimal (using the scheme of decimal expansion in which infinite chains of 9’s are avoided—in which, for example, 4 is represented by .25000 . . . rather than by .24999 . . .); then let y be the fifty-ninth digit after the decimal point. There is of course no standard formula for this, but nevertheless it is a perfectly respectable function whose rule is given by a verbal description. On the other hand, the function y = sin x of the real variable z is so important that its rule, though fully as compli- cated as the one just defined, is assigned the special symbol sin. When discussing functions in general, we want to allow for all sorts of rules and to talk about them all at once, so we usually employ noncommittal notations like y = f(x), y = g(x), and so on. Each of the functions mentioned above is defined for all real numbers a. The example y = 1/z shows that this restriction is much too severe, for this function is defined only for non-zero values of z. Similarly, y = log z is defined only for positive values of z, and y = sin! z only for values of x which lie in the interval [—1,1]. Whatever our conception of a function may be, it should certainly be broad enough to include examples like these, which are defined only for some values of the real variable z. In real analysis the notion of function is introduced in the following way. Let X be any non-empty set of real numbers. We say that a function y = f(z) is defined on X if the rule f associates a definite real number y with each real number z in X. The specific nature of the tule f is totally irrelevant to the concept of a function. The set X is called the domain of the given function, and the set Y of all the values it Assumes is called its range. If we speak of complex numbers here instead of real numbers, we have the notion of function as it is used in complex analysis. This point of view toward functions is actually a bit more general than is needed for the aims of analysis, but it isn’t nearly general enough for our purposes. The sets X and Y above were taken to be sets of numbers. If we now remove even this restriction and allow X and Y to be completely arbitrary non-empty sets, then we arrive at the most inclusive concept of a function. By way of illustration, suppose that X is the set of all squares in a plane and that Y is the set of all circles in the same plane. We can define a function y = f(x) by requiring that the rule f associate with each square z that circle y which is inscribed in it. In general, there is no need at all for either X or Y to be a set of16 Topology numbers. All that is really necessary for a function is two non-empty sets X and Y and a rule f which is meaningful and unambiguous in assigning to each element z in X a specific element y in Y. With these preliminary descriptive remarks, we now turn to the rather abstract but very precise ideas they are intended to motivate. A function consists of three objects: two non-empty sets X and Y (which may be equal, but need not be) and a rule f which assigns to each element x in X a single fully determined element y in Y. The y which corresponds in this way to a given z is usually written f(x), and is called the image of x under the rule f, or the value of f at the element x. This f fee eee aa Fig. 6. A way of visualizing mappings. notation is supposed to be suggestive of the idea that the rule f takes the element z and does something to it to produce the element y = f(z). The rule f is often called a mapping, or transformation, or operator, to amplify this concept of it. We then think of f as mapping 2’s to y’s, or transforming 2's into y’s, or operating on z’s to produce y’s. The sct X is called the domain of the function, and the set of all f(z)’s for all z’s in X is called its range. A function whose range consists of just one element is called a constant function. We often denote by f:X — Y the function with rule f, domain X, and range contained in Y. This notation is useful because the essential parts of the function are displayed in a manner which emphasizes that it is a composite object, the central thing being the rule or mapping f. Figure 6 gives a convenient way of picturing this function. On the left, X and Y are different sets, and on the right, they are equal—in which case we usually refer to f as a mapping of X into itself. If it is clear from the context what the sets X and Y are, or if there is no real need to specify them explicitly, it is common practice to identify the function f:X— Y with the rule f, and to speak of f alone as if it were the function under consideration (without mentioning the sets X and Y). It sometimes happens that two perfectly definite sets X and Y are under discussion and that a mapping of X into Y arises which has nu natural symbol attached to it. If there is no necessity to invent aSets and Functions 7 symbol for this mapping, and if it is quite clear what the mapping is, it is often convenient to designate it by x—» y. Accordingly, the function y = 2 mentioned at the beginning of this section can be written as 2— x? or x— y (where y is understood to be the square of x). A function f is called an extension of a function g (and g is called a restriction of f) if the domain of f contains the domain of g and f(z) = g(z) for each z in the domain of g. Most of mathematical analysis, both classical and modern, deals with functions whose values are real numbers or complex numbers. f emer esa Pepe es Fig. 7. The inverse of a mapping. This is also true of those parts of topology which are concerned with the foundations of analysis. If the range of a function consists of real numbers, we call it a real function; similarly, a complex function is one whose range consists of complex numbers. Obviously, every real function is also complex. We lay very heavy emphasis on real and complex func- tions throughout our work. As a matter of usage, we generally prefer to reserve the term function for real or complex functions and to speak of mappings when dealing with functions whose values are not necessarily numbers. Consider a mapping f:X — Y. When we call f a mapping of X into Y, we mean to suggest by this that the elements f(z)—as x varies over all the elements of X—need not fill up Y; but if it definitely does happen that the range of f equals Y, or if we specifically want to assume this, then we call f a mapping of X onto Y. If two different elements in X always have different images under f, then we call f a one-to-one mapping of X into Y. If f:X— Y is both onto and one-to-one, then we can define its inverse mapping f-!: ¥ + X as follows: for each y in Y, we find that unique element x in X such that f(z) = y (z exists and is unique since f is onto and one-to-one); we then define z to be f-!(y). The equation z = f-'(y) is the result of solving y = f(x) for x in just the same way as x = log y is the result of solving y = e* for x. Figure 7 illustrates the concept of the inverse of a mapping.18 Topology If f is a one-to-one mapping of X onto Y, it will sometimes be con- venient to subordinate the conception of f as a mapping sending 2’s over to y’s and to emphasize its role as a link between z’s and y’s. Each z has linked to it (or has corresponding to it) precisely one y = f(x); and, turning the situation around, each y has linked to it (or has corre- sponding to it) exactly one z = f-'(y). When we focus our attention on this aspect of a mapping which is one-to-one onto, we usually call it a one-to-one correspondence. Thus f is a one-to-one correspondence between X and Y, and f-1 is a one-to-one correspondence between Y and X. Now consider an arbitrary mapping f:X— Y. The mapping f, which sends each element of X over to an element of Y, induces the following two important set mappings. If A is a subset of X, then its image f(A) is the subset of Y defined by S(A) = {f(@):2€ A}, and our first set mapping is that which sends each A over to its corre- sponding f(A). Similarly, if B is a subset of Y, then its inverse image f-1(B) is the subset of X defined by S-(B) = {x:f(z) € Bh, and the second set mapping pulls each B back to its corresponding J-\(B). It is often essential for us to know how these set mappings behave with respect to set inclusion and operations on sets. We develop most of their significant features in the following two paragraphs. The main properties of the first set mapping are: fM=9 SMOCY; Arf Ar=f(Ai) € f(A); HUA) = UIAd; SOA) S Of (Ad The reader should convince himself of the truth of these statements. For instance, to prove (1) we would have to prove first that f(U;A,) isa subset of Uif(Ai), and second that U,f(A,) is a subset of f(UiA). A proof of the first of these set inclusions might run as follows: an element in f(U;A,) is the image of some element in Ai, therefore it is the image of an element in some Ay, therefore it is in some f(A,), and so finally it is in Usf(As). The irregularities and gaps which the reader will notice in the above statements are essential features of this set mapping. For exam- ple, the image of an intersection need not equal the intersection of the images, because two disjoint sets can easily have images which are not disjoint. Furthermore, without special assumptions (see Problem 6) nothing can be said about the relation between f(A)’ and f(A’). q)Sets and Functions 19 The second set mapping is much better behaved. Its properties are satisfyingly complete, and can be stated as follows: f7G) =9; fr) =X; B,C Br f(Bi) S f"(B2); SUB) = Uif-(B); (2) PMOB) = Of-(B)5 (3) SB) = f(By. (4) Again, the reader should verify each of these statements for himself. f g Bere EEE eae ec eer eer (Pane. f(z) (x), Fig. 8. Multiplication of mappings. We discuss one more concept in this section, that of the multiplication (or composition) of mappings. If y = f(z) = 2* +1 and z= gy) = sin y, then these two functions can be put together to form a single function defined by z = (gf)(x) = g(f(z)) = gz? + 1) = sin (2? +1). One of the most important tools of calculus (the chain rule) explains how to dif- ferentiate functions of this kind. This manner of multiplying functions together is of basic importance for us as well, and we formulate it in general as follows. Suppose that f:X — Y and g:Y— Z are any two mappings. We define the product of these mappings, denoted by of:X — Z, by (9f)(z) = g(f(x)). In words: an element z in X is taken by f to the element f(z) in Y, and then g mapsf(zx) to g(f(x))in Z. Figure 8 isa picture of this process. We observe that the two mappings involved here are not entirely arbitrary, for the set Y which contains the range of the first equals the domain of the second. More generally, the product of two mappings is meaningful whenever the range of the first is con- tained in the domain of the second. We have regarded f as the first mapping and g as the second, and in forming their product gf, their symbols have gotten turned around. This is a rather unpleasant phenomenon, for which we blame the occasional perversity of mathe- matical symbols. Perhaps it will help the reader to keep this straight20 Topology in his mind if he will remember to read the product gf from right to left: first apply f, then g. Problems 1, Two mappings f:X — Y and g:X — Y are said to be equal (and we write this f = g) if f(x) = g(x) for every x in X. Let f, 9, andh be any three mappings of a non-empty set X into itself, and show that multiplication of mappings is associative in the sense that S(gh) = (fg)h. 2. Let X be a non-empty set. The identity mapping ix on X is the mapping of X onto itself defined by ix(x) = x for every x. Thus tx sends each element of X to itself; that is, it leaves fixed each element of X. Show that fix = txf = f for any mapping f of X into itself. If f is one-to-one onto, so that its inverse f-' exists, show that Sf = f-f = ix. Show further that f- is the only mapping of X into itself which has this property; that is, show that if g is a mapping of X into itself such that fg = gf =ix, then g =f! (hint: 9 = gix = g(ff") = Off = xf! = ft, or 9 = ixg = (F'fg = F(fo) = fix =F). 3. Let X and Y be non-empty sets and f a mapping of X into Y. Show the following: (a) f is one-to-one = there exists a mapping g of Y into X such that gf = tx; (6) f is onto + there exists a mapping h of Y into X such that fh = ty. 4, Let X be a non-empty set and f a mapping of X into itself. Show that f is one-to-one onto = there exists a mapping g of X into itself such that fg = gf = ix. If there exists a mapping g with this property, then there is only one such mapping. Why? 5. Let X be a non-empty set, and let f and g be one-to-one mappings of X onto itself. Show that fg is also a one-to-one mapping of X onto itself and that (fg)-! = g-f-1. 6. Let X and Y be non-empty sets and f a mapping of X into Y. If A and B are, respectively, subsets of X and Y, show the following: (a) ff-(B) C B, and ff-\(B) = B is true for all B + f is onto; (b) A Cf-¥(A), and A = f-¥f(A) is true for all A = fis one-to-one; (ce) f(A1O Az) = f(A) A f(A2) is true for all A, and A; @f is one-to-one; (@) f(A)’ C f(A’) is true for all A =f is onto; (e) if f is onto—so that f(A)’ C f(A’) is true for all A—then (A) = f(A’) is true for all A =f is also one-to-one.Sets and Functions 21 4, PRODUCTS OF SETS We shall often have occasion to weld together the sets of a given class into a single new set called their product (or their Cartesian product). The ancestor of this concept is the coordinate plane of analytic geometry, that is, a plane equipped with the usual rectangular coordinate system. We give a brief description of this fundamental idea with a view to paving the way for our discussion of products of sets in general. First, a few preliminary comments about the real line. We have already used this term several times without any explanation, and of course what we mean by it is an ordinary geometric straight line (see Fig. 9) whose points have been identified with—or coordinatized by—the Ya Ye v2 er o 1 2 3 Fig. 9. The real line. set R of all real numbers. We use the letter R to denote the real line as well as the set of all real numbers, and we often speak of real numbers as if they were points on the real line, and of points on the real line as if they were real numbers. Let, no one be deceived into thinking that the real line is a simple thing, for its structure is exceedingly intricate. Our present view of it, however, is as naive and uncomplicated as the picture of it given in Fig. 9. Generally speaking, we assume that the reader is familiar with the simpler properties of the real line—those relating to inequalities (see Problem 1-2) and the basic algebraic operations of addition, subtraction, multiplication, and division. One of the most significant facts about the real number system is perhaps less well known. This is the so-called least upper bound property, which asserts that every non-empty set of real numbers which has an upper bound has a least upper bound. It is an easy consequence of this that every non- empty set of real numbers which has a lower bound has a greatest lower bound. All these matters can be developed rigorously on the basis of a small number of axioms, and detailed treatments can often be found in books on elementary abstract elgebra. To construct the coordinate plane, we now proceed as follows. We take two identical replicas of the real line, which we call the z axis and the y axis, and paste them on a plane at right angles to one another in such a way that they cross at the zero point on each. The usual picture is given in Fig. 10. Now let P be a point in the plane. We project P perpendicularly onto points P, and P, on the axes. If z andy are the coordinates of P; and P, on their respective axes, this process22 Topology leads us from the point P to the uniquely determined ordered pair (z,y) of real numbers, where x and y are called the x coordinate and y coordi- nate of P. We can reverse the process, and, starting with the ordered pair of real numbers, we can recapture the point. This is the manner in which we establish the familiar one-to-one correspondence between points P in the plane and ordered pairs (z,y) of real numbers. In fact, we think of a point in the plane (which is a geometric object) and its corresponding ordered pair of real numbers (which is an algebraic object) as being—to all intents and purposes—identical with one another. The essence of analytic geometry lies y axis in the possibility of exploiting this identification by using algebraic tools in geometric arguments and giving geometric interpretations to ay Palay) algebraic calculations. The conventional attitude to- ward the coordinate plane in ana- lytic geometry is that the geometry is the focus of interest and the alge- bra of ordered pairs is only a con- xexis venient tool. Here we reverse this point of view. For us, the coordinate Fig. 10. The coordinate plane. plane is defined to be the set of all ordered pairs (x,y) of real numbers. We can satisfy our desire for visual images by using Fig. 10 as a picture of this set and by calling such an ordered pair a point, but this geo- metric language is more a convenience than a necessity. Our notation for the coordinate plane is R XR, or R*. This symbolism reflects the idea that the coordinate plane is the result of “multiplying together” two replicas of the real line R. It is perhaps necessary to comment on one possible source of mis- understanding. When we speak of R? as a plane, we do so only to establish an intuitive bond with the reader’s previous experience in ana- lytic geometry. Our present attitude is that R? is a pure set and has no structure whatever, because no structure has yet been assigned to it. We remarked earlier (with deliberate vagueness) that a space is a set to which has been added some kind of algebraic or geometric structure, In Sec. 15 we shall convert the set 2? into the space of analytic geometry by defining the distance between any two points (x1,y1) and (#2,y2) to be VG ~ ay FH) This notion of distance endows the set R? with a certain “spatial” char- acter, which we shall recognize by calling the resulting space the Huclidean plane instead of the coordinate plane. Bey 1 | ' ! I 1 It 'Sets and Functions = 23. We assume that the reader is fully acquainted with the way in which the set C of all complex numbers can be identified (as a set) with the coordinate plane R®. If 2 is a complex number, and if z has the standard form x + iy where x and y are real numbers, then we identify z with the ordered pair (z,y), and thus with an element of R*. The complex numbers, however, are much more than merely a set. They constitute a number system, with operations of addition, multi- plication, conjugation, etc. When the coordinate plane R? is thought of as consisting of complex numbers and is enriched by the algebraic structure it acquires in this way, it is called the complex plane. The y,| letter C is used to denote either the set of all complex numbers or the complex plane. We shall make a space out of the complex plane in Sec. 9. Suppose now that Xi and X_ are any two non-empty sets. By analogy with our above discussion, their product X, X Xz is defined to aL x be the set of allordered pairs (xi,22), Fig. 11. A way of visualizing X1 X Xz. where z, is in X; and 22 is in Xo. In spite of the arbitrary nature of X, and X:, their product can be repre- sented by a picture (see Fig. 11) which is loosely similar to the usual picture of the coordinate plane. The term product is applied to this set, and it is thought of as the result of “multiplying together” Xi and X», for the following reason: if X, and X; are finite sets with m and n elements, then (clearly) X: X X2 has mn elements. If f:X:— X» is a mapping with domain X; and range in X2, its graph is that subset of X; X Xe which consists of all ordered pairs of the form (z1,f(z:)). We observe that this is an appropriate generalization of the concept of the graph of a function as it occurs in elementary mathematics. This definition of the product of two sets extends easily to the case of n sets for any positive integer n. If X1, X2,..., X, are non-empty sets, then their product Xi X X2 X +--+ X Xn is the set of all ordered n-tuples (21, Z2, . . . , Zn), where x; is in X; for each subscript 7. If the X¥s are all replicas of a single set X, that is, if Xoo =X,=4K, then their product is usually denoted by the symbol X*. These ideas specialize directly to yield the important sets R° and C*, R'is just R, the real line, and R? is the coordinate plane. R*—the set of all ordered triples of real numbers—is the set which underlies solid analytic geometry, and we assume that the reader is familiar with Re24 Topology the manner in which this set arises, through the introduction of a rec- tangular coordinate system into ordinary three-dimensional space. We can draw pictures here just as in the case of the coordinate plane, and we can use geometric language as much as we please, but it must be under- stood that the mathematics of this set is the mathematics of ordered triples of real numbers and that the pictures are merely an aid to the intuition. Once we fully grasp this point of view, there is no difficulty whatever in advancing at once to the study of the set R* of all ordered n-tuples (21, 22, . . . , Zn) of real numbers for any positive integer n. It is quite true that when n is greater than 3 it is no longer possible to draw the same kinds of intuitively rich pictures, but at worst this is merely an inconvenience. We can (and do) continue to use suggestive geometric language, so all is not lost. The set C” is defined similarly: it is the set of all ordered n-tuples (21, 22, . . . , 2n) of complex numbers. Each of the sets R” and C* plays a prominent part in our later work. We emphasized above that for the present the coordinate plane is to be considered as merely a set, and not a space. Similar remarks apply to R" and C*. In due course (in Sec. 15) we shall impart form and content to each of these sets by suitable definitions. We shall convert them into the Euclidean and unitary n-spaces which underlie and motivate so many developments in modern pure mathematics, and we shall explore some aspects of their algebraic and topological structure to the very last pages of this book. But as of now—and this is the point we insist on—neither one of these sets has any structure at all. As the réader doubtless suspects, it is not enough that we consider only products of finite classes of sets. The needs of topology compel us to extend these ideas to arbitrary classes of sets. We defined the product X: X X2X - ++ X X, to be the set of all ordered n-tuples (x1, t2,..., %») such that 2; is in X; for each subscript 7. To see how to extend this definition, we reformulate it as follows. We have an index set J, consisting of the integers from 1 to n, and corresponding to each index (or subscript) i we have a non-empty set X; The n-tuple (x1, zz, . . . , 2a) is simply a function (call it z) defined on the index set J, with the restriction that its value x(i) = xis an element of the set X; for each iin J. Our point of view here is that the function x is completely determined by, and is essentially equivalent to, the array (21, 2, . . . , tn) of its values. The way is now open for the definition of products in their full generality. Let {X,} be a non-empty class of non-empty sets, indexed by the elements i of an index set J. The sets X; need not be different from one another; indeed, it may happen that they are all identical replicas of a single set, distinguished only by different indices. The product of the sets X;,, written Pir X;, is defined to be the set of all functions z defined on I such that x(i) is an element of the set X; forSets and Functions 25 each index i. We call X; the tth coordinate set. When there can be no misunderstanding about the index set, the symbol P;,; X; isoften abbrevi- ated to P:X; The definition we have just given requires that each coordinate set be non-empty before the product car be formed. It will be useful if we extend this definition slightly by agreeing that if any of the X;’s are empty, then P;X; is also empty. This approach to the idea of the product of a class of sets, by means of functions defined on the index set, is useful mainly in giving the definition. In practice, it is much more convenient to use the subscript notation a; instead of the function notation z(i). We then interpret the product P.X, as made up of elements z, each of which is specified by the exhibited array {x;} of its values in the respective coordinate sets X;. We call a; the ith coordinate of the element x = {z;}. The mapping p; of the product P;X; onto its ith coordinate set Xi which is defined by p:(x) = x;—that is, the mapping whose value at an arbitrary element of the product is the ith coordinate of that element —is called the projection onto the ith coordinate set. The projection p; selects the ith coordinate of each element in its domain. There is clearly one projection for each element of the index set I, and the set of all projections plays an important role in the general theory of topological spaces. Problems 1. The graph of a mapping f:X — Y is a subset of the product X x Y. What properties characterize the graphs of mappings among all subsets of X X Y¥? 2. Let X and Y be non-empty sets. If A; and A are subsets of X, and B, and B; subsets of Y, show the following: (Ai X By) 1 (Ai X Br) = (Ar 0 Aa) X (Br Ba); (Ai X Bi) — (Aa X Bs) = (Air — Az) X (Br — Bz) U (ALO Az) X (Bi — Ba) U (Ai — A2) X (BLO B:). 3. Let X and Y be non-empty sets, and let A and B be rings of subsets of X and Y, respectively. Show that the class of all finite unions of sets of the form A X B with A € A and B ¢ Bisa ring of subsets of xXxXY. 5. PARTITIONS AND EQUIVALENCE RELATIONS In the first part of this section we consider a non-empty set X, and we study decompositions of X into non-empty subsets which fill it out26 Topology and have no elements in common with one another. We give special attention to the tools (equivalence relations) which are normally used to generate such decompositions. A partition of X is a disjoint class {X,} of non-empty subsets of X whose union is the full set X itself. The X,’s are called the partition sets. Expressed somewhat differently, a partition of X is the result of splitting it, or subdividing it, into non-empty subsets in such a way that each element of X belongs to one and only one of the given subsets. If X is the set {1, 2, 3, 4, 5}, then {1, 3, 5}, {2,4} and {1, 2, 3}, {4, 5} are two different partitions of X. If X is the set R of all real numbers, then we can partition X into the set of all rationals and the set of all irrationals, or into the infinitely many closed-open intervals of the form {n, n + 1) where n is an integer. If X is the set of all points in the coordinate plane, then we can partition X in such a way that each partition set, consists of all points with the same x coordinate (vertical lines), or so that each partition set consists of all points with the same y coordinate (horizontal lines). Other partitions of each of these sets will readily occur to the reader. In general, there are many different ways in which any given set can be partitioned. These manufactured examples are admittedly rather uninspiring and serve only to make our ideas more concrete. Later in this section we consider some others which are more germane to our present purposes. A binary relation in the set X is a mathematical symbol or verbal phrase, which we denote by R in this paragraph, such that for each ordered pair (z,y) of elements of X the statement z Ry is meaningful, in the sense that it can be classified definitely as true or false. For such a binary relation, z R y symbolizes the assertion that x is related by R to y, and x Wy the negation of this, namely, the assertion that x is not related by R to y Many examples of binary relations can be given, some familiar and others less so, some mathematical and others not. For instance, if X is the set of all integers and R is interpreted to mean “Gs less than,” which of course is usually denoted by the symbol <, then we clearly have 4 <7 and 5¢ 2. We have been speaking of binary relations, which are so named because they apply only to ordered pairs of elements, rather than to ordered triples, etc. In our work we drop the qualifying adjective and speak simply of a relation in X, since we shall have occasion to consider only relations of this kind.! We now assume that a partition of our non-empty set X is given, 1 Some writers prefer to regard a relation R in X asa subset Rof X X X. From this point of view, z Ry and x Hy are simply equivalent ways of writing (z,y)¢ R and (z,y)¢R. This definition has the advantage of being more tangible than ours, and the disadvantage that few people really think of a relation in this way.Sets and Functions 27 and we associate with this partition a relation in X. This relation is defined in the following way: we say that « is equivalent to y and write this z ~ y (the symbol ~ is pronounced “wiggle”), if z and y belong to the same partition set. It is obvious that the relation ~ has the follow- ing properties: (1) 2 ~ a for every x (reflexivity); (2) t~y=sy~ a (symmetry); (3) s~yandy~z=2 ~z (transitivity). This particular relation in X arose in a special way, in connection with a given partition of X, and its properties are immediate consequences of its definition. Any relation whatever in X which possesses these three properties is called an equivalence relation in X. We have just seen that each partition of X has associated with it a natural equivalence relation in X. We now reverse the situation and show that a given equivalence relation in X determines a natural partition of X. Let ~ be an equivalence relation in X; that is, assume that it is reflexive, symmetric, and transitive in the sense described above. If x is an element of X, the subset of X defined by [z] = {y:y ~ 2} is called the equivalence set of x. The equivalence set of z is thus the set of all elements which are equivalent to x. We show that the class of all distinct equivalence sets forms a partition of X. By reflexivity, x [z] for each element z in X, so each equivalence set is non-empty and their union is X. It remains to be shown that any two equivalence sets [x1] and [z2] are either disjoint or identical. We prove this by showing that if [z:] and [x2] are not disjoint, then they must be identical. Sup- pose that [21] and [zs] are not disjoint; that is, suppose that they have a common element z. Since z belongs to both equivalence sets, 2 ~ 21 and z~ 22, and by symmetry, %;~ 2. Let y be any element of [x], so that y~a:. Since y~ a: and x ~z, transitivity shows that y~z. By another application of transitivity, y ~~ z and z~ 2 imply that y ~ 22, so that yisin [z.]. Since y was chosen arbitrarily in [x1], we see by this that [z,] C [x]. The same reasoning shows that [x2] C [x], and from this we conclude (sce the last paragraph of Sec. 1) that [z:] = [zz]. The above discussion demonstrates that there is no real distinction (other than a difference in language) between partitions of a set and equivalence relations in the set. If we start with a partition, we get an equivalence relation by regarding elements as equivalent if they belong to the same partition set, and if we start with an equivalence relation, we get a partition by grouping together into subsets all elements which are equivalent to one another. We have here a single mathematical idea, which we have been considering from two different points of view, and the approach we choose in any particular application depends entirely28 = Topology on our own convenience. In practice, it is almost invariably the case that. we use equivalence relations (which are usually easy to define) to obtain partitions (which are sometimes difficult to describe fully). We now turn to several of the more important simple examples of equivalence relations. Let I be the set of all integers. If a and b are elements of this set, we write a = b (and say that a equals b) if a and b are the same integer. Thus 2 + 3 = 5 means that the expressions on the left and right are simply different ways of writing the same integer. It is apparent that = used in this sense is an equivalence relation in the set J: (1) a = a for every a; (2) b=b=a; (3) a=bandb=c>a=c. Clearly, each equivalence set consists of precisely one integer. Another familiar example is the relation of equality commonly used for fractions. We remind the reader that, strictly speaking, a fraction is merely a symbol of the form a/b, where a and b are integers and b is not zero. The fractions 24 and 46 are obviously not identical, but nevertheless we consider them to be equal. In general, we say that two fractions a/b and c/d are equal, written a/b = c/d, if ad and be are equal as integers in the usual sense (see the above paragraph). We leave it to the reader to show that this is an equivalence relation in the set of all fractions. An equivalence set of fractions is what we call a rational number. Everyday usage ignores the distinction between fractions and rational numbers, but it is important to recognize that from the strict point of view it is the rational numbers (and not the fractions) which form part of the real number system. Our final example has a deeper significance, for it provides us with the basic tool for our work of the next two sections. For the remainder of this section we consider a relation between pairs of non-empty sets, and each set mentioned (whether we say so explicitly or not) is assumed to be non-empty. If X and Y are two sets, we say that X is numerically equivalent to Y if there exists a one-to- one correspondence between X and Y, i.e., if there exists a one-to-one mapping of X onto Y. This relation is reflexive, since the identity mapping tx: X — X is one-to-one onto; it is symmetric, since if f:X > Y is one-to-one onto, then its inverse mapping f-': Y > X is also one-to-one onto; and it is transitive, since if f:X — Y and g:Y — Z are one-to-one onto, then gf: X — Z is also one-to-one onto. Numerical equivalence has all the properties of an equivalence relation, and if we consider it as an equivalence relation in the class of all non-empty subsets of some universal set U, it groups together into equivalence sets all those subsets of U which have the same number of elements. After we state and prove theSets and Functions 29 following very useful but rather technical theorem, we shall continue in Secs. 6 and 7 with an exploration of the implications of these ideas. The theorem we have in mind—the Schroeder-Bernstein theorem—is the following: if X and Y are two sets each of which is numerically equivalent to a subset of the other, then all of X is numerically equivalent to all of Y. There are several proofs of this classic theorem, some of which are quite difficult. The very elegant proof we give is essentially due to Birkhoff and MacLane. Now for the proof. We assume that f:X— Y is a one-to-one mapping of X into Y, and that g: Y > X is a one-to-one mapping of Y into X. Our task is to produce a mapping F': X — Y which is one-to-one onto. We may assume that neither f nor g is onto, since if f is, we can define F to be f, and if g is, we can define F to be g-'. Since both f and g are one-to-one, it is permissible to use the mappings f-' and g™! as long as we clearly understand that f—! is defined only on f(X) and g™ only on g(Y). We obtain the mapping F by splitting both X and Y into subsets which we characterize in terms of the ancestry of their elements. Let z be an element of X. We apply g~! to it (if we can) to get the element g(x) in Y. If g-*(z) exists, we call it the first ancestor of z. The ele- ment z itself we call the zeroth ancestor of z. We now apply f-! to g7\(z) if we can, and if (f-'g-*)(z) exists, we call it the second ancestor of x. We now apply g™ to (f-\g-1)(z) if we can, and if (g~'f-"g—)(z) exists, we call it the third ancestor of x. As we continue this process of tracing back the ancestry of x, it becomes apparent that there are three possibilities. (1) z has infinitely many ancestors. We denote by X; the subset of X which consists of all elements with infinitely many ancestors. (2) 2 has an even number of ancestors; this means that z has a last ancestor (that is, one which itself has no first ancestor) in X. We denote by X, the subset of X consisting of all elements with an even number of ancestors. (3) z has an odd number of ancestors; this means that x has a last ancestor in Y. We denote by X, the subset of X which consists of all elements with an odd number of ancestors. The three sets X;, X., X, form a disjoint class whose union is X. We decom- pose Y in just the same way into three subsets Y;, Y., Y.. It is easy to see that f maps X; onto Y; and X, onto Y,, and that g-! maps X, onto Y,; and we complete the proof by defining F in the following piecemeal manner: f(z) ifceX:UX, g(x) ifee X. We attempt to illustrate these ideas in Fig. 12. Here we present two replicas of the situation: on the left, X and Y are represented by the vertical lines, and f and g by the lines slanting down to the right and F(z) =30 Topology left; and on the right, we schematically trace the ancestry of three ele- ments in X, of which 2; has no first ancestor, z2 has a first and second ancestor, and 2; has a first, second, and third ancestor. x Y al) (X) Fig. 12, The proof of the Schroeder-Bernstein theorem. The Schroeder-Bernstein theorem has great theoretical and practical significance. Its main value for us lies in its role as a tool by means of which we can prove numerical equivalence with a minimum of effort for many specific sets. We put it to work in Sec. 7. Problems 1 Let f: X — Y be an arbitrary mapping. Define a relation in X as follows: 21 -~ 22 means that f(z:) = f(z2). Show that this is an equivalence relation and describe the equivalence sets. In the set F of all real numbers, let s ~~ y mean that x — y is an integer. Show that this is an equivalence relation and describe the equivalence sets. Let I be the set of all integers, and let m be a fixed positive integer. Two integers a and b are said to be congruent modulo m—symbolized by a = b (mod m)—if a — b is exactly divisible by m, i.e., if a — bis an integral multiple of m. Show that this is an equivalence relation, describe the equivalence sets, and state the number of distinct equivalence sets. Decide which ones of the three properties of reflexivity, symmetry, and transitivity are true for each of the following relations in the setSets and Functions 31 of all positive integers: m y ~ 2; a~yand y~z>2x~z; therefore x ~z for every z. In view of Problem 5f, this cannot be a valid proof. What is the flaw in the reasoning? 7. Let X be a non-empty set. A relation ~ in X is called circular if ‘a~yandy ~2=2 ~ 2, and triangular ifz ~~ yandz ~z>y ~z. Prove that a relation in X is an equivalence relation + it is reflexive and circular + it is reflexive and triangular. a . COUNTABLE SETS ‘The subject of this section and the next—injintte cardinal numbers— lies at the very foundation of modern mathematics. It is a vital instru- ment in the day-to-day work of many mathematicians, and we shall make extensive use of it ourselves. This theory, which was created by the German mathematician Cantor, also has great aesthetic appeal, for it begins with ideas of extreme simplicity and develops through natural stages into an elaborate and beautiful structure of thought. In the course of our discussion we shall answer questions which no one before Cantor’s time thought to ask, and we shall ask a question which no one can answer to this day. Without further ado, we can say that cardinal numbers are those used in counting, such as the positive integers (or natural numbers) 1, 2, 3, . . . familiar to us all, But there is much more to the story than this. The act of counting is undoubtedly one of the oldest of human activities. Men probably learned to count in a crude way at about the same time as they began to develop articulate speech. The earliest men who lived in communities and domesticated animals must have found it necessary to record the number of goats in the village herd by means of a pile of stones or some similar device. If the herd was counted in each night by removing one stone from the pile for each goat accounted for, then stones left over would have indicated strays, and herdsmen would have gone out to search for them. Names for numbers and symbols for32 Topology them, like our 1, 2, 3, . . . , would have been superfluous. The simple and yet profound idea of a one-to-one correspondence between the stones and the goats would have fully met the needs of the situation. In a manner of speaking, we ourselves use the infinite set N = {1,2,3,...} of all positive integers as a “pile of stones.” We carry this set around with us as part of our intellectual equipment. Whenever we want to count a set, say, a stack of dollar bills, we start through the set N and tally off one bill against each positive integer as we come to it. The last number we reach, corresponding to the last bill, is what we call the number of bills in the stack. If this last number happens to be 10, then “10” is our symbol for the number of bills in the stack, as it also is for the number of our fingers, and for the number of our toes, and for the number of elements in any set which can be put into one-to-one correspondence with the finite set {1, 2, , 10}. Our procedure is slightly more sophisticated than that of the primitive savage. We have the symbols 1, 2, 8, . . . for the numbers which arise in counting; we can record them for future use, and communicate them to other people, and manipu- late them by the operations of arithmetic. But the underlying idea, that of the one-to-one correspondence, remains the same for us as it probably was for him. The positive integers are adequate for the purpose of counting any non-empty finite set, and since outside of mathematics all sets appear to be of this kind, they suffice for all non-mathematical counting. But in the world of mathematics we are obliged to consider many infinite sets, such as the set of all positive integers itself, the sct of all integers, the set of all rational numbers, the set of all real numbers, the set of all points in a plane, and so on. It is often important to be able to count such sets, and it was Cantor’s idea to do this, and to develop a theory of infinite cardinal numbers, by means of one-to-one correspondences. In comparing the sizes of two sets, the basic concept is that of numerical equivalence as defined in the previous section. We recall that two non-empty sets X and Y are said to be numerically equivalent if there exists a one-to-one mapping of one onto the other, or—and this amounts to the same thing—if there can be found a one-to-one corre- spondence between them. To say that two non-empty finite sets are numerically equivalent is of course to say that they have the same number of elements in the ordinary sense. If we count one of them, we simply establish a one-to-one correspondence between its elements and a set of positive integers of the form {1, 2, . . . , n}, and we then say that n is the number of elements possessed by both, or the cardinal number of both. The positive integers are the finite cardinal numbers. We encounterSets and Functions 33 many surprises as we follow Cantor and consider numerical equivalence for infinite sets. The set N = {1, 2, 3, .. .} of all positive integers is obviously “larger” than the set {2, 4, 6, . . .} of all even positive integers, for it contains this set as a proper subset. It appears on the surface that NV has “more” elements. But it is very important to avoid jumping to con- clusions when dealing with infinite sets, and we must remember that our criterion in these matters is whether there exists a one-to-one corre- spondence between the sets (not whether one set is or is not a proper subset of the other). Asa matter of fact, the pairing NaS pSE Het at eects 2,4,6,...,2n,... serves to establish a one-to-one correspondence between these sets, in which each positive integer in the upper row is matched with the even positive integer (its double) directly below it, and these two sets must therefore be regarded as having the same number of elements. This is a very remarkable circumstance, for it seems to contradict our intuition and yet is based only on solid common sense. We shall see below, in Problems 6 and 7-4, that every infinite set is numerically equivalent to a proper subset of itself. Since this property is clearly not possessed by any finite set, some writers even use it as the definition of an infinite set. In much the same way as above, we can show that N is numerically equivalent to the set of all even integers: OH g ratte He ttge 0,2, —2, 4, —4,6, -6,... Here our device is to start with 0 and follow each even positive integer as we come to it by its negative. Similarly, N is numerically equivalent to the set of all integers: 1,2, 3,4, 5,6, 7... 0,1, -1,2, -2,3, -3,... It is of considerable historical interest to note that Galileo observed in the early seventeenth century that there are precisely as many perfect squares (1, 4, 9, 16, 25, etc.) among the positive integers as there are positive integers altogether. This is clear from the pairing 1, 2, 3,4, 5... 1, 2, 34, 43,53... It struck him as very strange that this should be true, considering how34 ~— Topology sparsely strewn the squares are among all the positive integers. But. the time appears not to have been ripe for the exploration of this phenom- enon, or perhaps he had other things on his mind; in any case, he did not follow up his idea. These examples should make it clear that all that is really necessary in showing that an infinite set X is numerically equivalent to N is that we be able to list the elements of X, with a first, a second, a third, and so on, in such a way that it is completely exhausted by this counting off of its elements. It is for this reason that any infinite set which is numerically equivalent to N is said to be countably infinite. We say that a set is countable if it is non-empty and finite (in which case it can obviously be counted) or if it is countably infinite. One of Cantor’s earliest discoveries in his study of infinite sets was that the set of all positive rational numbers (which is very large: it contains N and a great many other numbers besides) is actually countable. We cannot list the positive rational numbers in order of size, as we can the positive integers, beginning with the smallest, then the next smallest, and so on, for there is no smallest, and between any two there are infi- nitely many others. We must find some other way of counting them, and following Cantor, we arrange them not in order of size, but according to the size of the sum of the numerator and denominator. We begin with all positive rationals whose numerator and denominator add up to 2: there is only one, }{ = 1. Next we list (with increasing numerators) all those for which this sum is 3:14, 2{ = 2. Next, all those for which this sum is 4:14, 34 = 1, 3 = 3. Next, all those for which this sum is 5:4, %, 34, = 4. Next, all those for which this sum is 6:14, 34 = 34, 3 = 1, 4% = 2,54 = 5. And soon. If we now list all these together from the beginning, omitting those already listed when we come to them, we get-a sequence 1, 34, 2, 18, 3, 14, 3%, 94, 4, 18,5,» which contains each positive rational number once and only once. Figure 13 gives a schematic representation of this manner of listing the positive rationals. In this figure the first row contains all positive rationals with numerator 1, the second all with numerator 2, etc.; and the first column contains all with denominator 1, the second all with denominator 2, and so on. Our listing amounts to traversing this array of numbers as the arrows indicate, where of course all those numbers already encountered are left out. It’s high time that we christened the infinite cardinal number we’ ve been discussing, and for this purpose we use the first letter of the Hebrew alphabet (N, pronounced “aleph”) with 0 as a subscript. We say that N> is the number of elements in any countably infinite set. OurSets and Functions 35 vomplete list of cardinal numbers so far is 1,2,3,...,%o We expand this list in the next section. Suppose now that m and n are two cardinal numbers (finite or infinite). The statement that m is less than n (written m < 7) is defined to mean the following: if X and Y are sets with m and n elements, then 2iadiaidk oo a 1 2 3 4 5 eee Heer HCeeeHeeeeH-Heee eerie a 1 2 3 4 6 ee ee eee eee eee eee aa-- 1 2 3 4 5 Eee Oe ee ee ee pees Ths tis gE Ee tee ree ree eee et ae 1 2 3 4 5 Fig. 13. A listing of the positive rationals. (1) there exists a one-to-one mapping of X into Y, and (2) there does not exist a one-to-one mapping of X onto Y. Using this concept, it is easy to relate our cardinal numbers to one another by means of 1<2<3< +++ and less than c? No one knows the answer to this question. Cantor himself thought that there is no such number, or in other words, that c is the next infinite cardinal number greater than Xo, and his guess has come to be known as Cantor’s continuum hypothesis. The continuum hypothesis can also be expressed by the assertion that every uncountable set of real numbers has ¢ as its cardinal number.' There is another question which arises naturally at this stage, and this one we are fortunately able to answer. Are there any infinite cardinal numbers greater than c? Yes, there are; for example, the cardinal number of the class of all subsets of R. This answer depends on the following fact: if X is any non-empty set, then the cardinal number of X is less than the cardinal number of the class of all subsets of X. We prove this statement as follows. In accordance with the defini- tion given in the last paragraph of the previous section, we must show 1 For further information about the continuum hypothesis, see Wilder (42, p. 125] and Gadel [12].«0 Topology (1) that there exists a one-to-one mapping of X into the class of all its subsets, and (2) that there does not exist such a mapping of X onto this class. To prove (1), we have only to point to the mapping z— {z}, which makes correspond to each element z that set {x} which consists of the element x alone. We prove (2) indirectly. Let us assume that there does exist a one-to-one mapping f of X onto the class of all its subsets. We now deduce a contradiction from the assumed existence of such a mapping. Let A be the subset of X defined by A = {a:2¢f(z)}. Since our mapping f is onto, there must exist an element ain X such that S(@ = A. Where is the element a? Jf ais in A, then by the definition of A we have a ¢ f(a), and since f(a) = A,ag¢A. Thisisa contradiction, soacannot belong to A. Butif ais notin A, then again by the definition of A we have aéf(a) or ae A, which is another contradiction. The situation is impossible, so our assumption that such a mapping exists must be false. This result guarantees that given any cardinal number, there always exists a greater one. If we start with a set X, = {1} containing one element, then there are two subsets, the empty set @ and the set {1} itself. If Xz: = {1,2} is a set containing two elements, then there are four subsets: 6, {1}, {2}, {1,2}. If Xs = {1, 2, 3} is a set containing three elements, then there are eight subsets: @, {1}, {2}, {3}, {1,2}, {1,3}, {2,3}, {1, 2,3}. In general, if X, is a set with n elements, where 7 is any finite cardinal number, then X, has 2" subsets. If we now take n to be any infinite cardinal number, the above facts suggest that we define 2" to be the number of subsets of any set with n elements. If n is the first infinite cardinal number, namely, No, then it can be shown that ok = 6, The simplest proof of this fact depends on the ideas developed in the following paragraph. Consider the closed-open unit interval [0,1) and a real number x in this set. Our concern is with the meaning of the decimal, binary, and ternary expansions of x. For the sake of clarity, let us take z to be }4. How do we arrive at the decimal expansion of }4? First, we split (0,1) into the 10 closed-open intervals (0,40), B40,3£0), - » - (Mos), and we use the 10 digits 0, 1, . . . , 9 to number them in order. Our number }4 belongs to exactly one of these intervals, namely, to [3{0,340). We have labeled this interval with the digit 2, so 2 is the first digit after the decimal point in the decimal expansion of 44: w= 2...Sets and Functions = 41 Next, we split the interval [2{0,3{0) into the 10 closed-open intervals (%{o,?4o0), [?4400,72f00), - - - » (?%00,340), and we use the 10 digits to number these in order. Our number }4 belongs to [25490,28{00), which is labeled with the digit 5, so 5 is the second number after the decimal point in the decimal expansion of 34: Wa 2... If we continue this process exactly as we started it, we can obtain the decimal expansion of }4 to as many places as we wish. As a matter of fact, if we do continue, we get 0 at each stage from this point on: 4 = .25000... The reader should notice that there is no ambiguity in this system as we have explained it: contrary to customary usage, .24999 . . . is not to be regarded as another decimal expansion of 14 which is “equivalent” to 25000... . In this system, each real number z in (0,1) has one and only one decimal expansion which cannot end in an infinite chain of 9’s. There is nothing magical about the role of the number 10 in the above discussion. If at each stage we split our closed-open interval into two equal closed-open intervals, and if we use the two digits 0 and 1 to number them, we obtain the binary expansion of any real number z in (0,1). The binary expansion of 14 is easily seen to be .01000.... The ternary expansion of z is found similarly: at each stage we split our closed-open interval into three equal closed-open intervals, and we use the three digits 0, 1, and 2 to number them. A moment’s thought should convince the reader that the ternary expansion of 14 is .020202 Just as (in our system) the decimal expansion of a number in (0,1) cannot end in an infinite chain of 9’s, so also its binary expansion cannot end in an infinite chain of 1’s, and its ternary expansion cannot end in an infinite chain of 2's. We now use this machinery to give a proof of the fact that ae = 6, Consider the two sets N = {1, 2,3, .. .} and J = (0,1), the first with cardinal number No and the second with cardinal number c. If N denotes the class of all subsets of N, then by definition N has cardinal number 2** Our proof amounts to showing that there exists a one-to-one corre- spondence between N and J. We begin by establishing a one-to-one mapping f of Ninto J. If A is a subset of N, then f(A) is that real num- ber z in J whose decimal expansion z = .didads . . . is defined by the condition that d, is 3 or 5 according asnisorisnotin A. Any other two digits can be used here, as long as neither of them is 9. Next, we con-42 Topology struct a one-to-one mapping g of J intoN. If z isa real number in I, and if = .bibebs . . . is its binary expansion (so that each b, is either 0 or 1), then g(x) is that subset A of N defined by A = {n:b, = 1}. We con- clude the proof with an appeal to the Schroeder-Bernstein theorem, which guarantees that under these conditions N and J are numerically equivalent to one another. If we follow up the hint contained in the fact that 2** = c, and successively form 2¢, 2%, and so on, we get a chain of cardinal numbers 1<2 x. A non-empty set P in which there is defined a partial order relation is called a partially44 Topology ordered set. It is clear that any non-empty subset of a partially ordered set is a partially ordered set in its own right. Partially ordered sets are abundant in all branches of mathematics. Some are simple and easy to grasp, while others are complex and rather inaccessible. We give four examples which are quite different in nature but possess in common the virtues of being both important and easily described. Example 1. Let P be the set of all positive integers, and let m z= y = 2, that is, if no element other than z itself is greater than or equal toz. A maximal element in P is thus an element of P which is not less than or equal to any other element of P. Examples 1, 2, and 4 have no maximal elements. Example 3 has a single maximal element: the set U itself. Let A be a non-empty subset of a partially ordered set P. An element z in P is called a lower bound of A if z < a for each a€ A; anda lower bound of A is called a greatest lower bound of A if it is greater than orSets and Functions 45 equal to every lower bound of A. Similarly, an element y in P is said to be an upper bound of A if a < y for every ae A; and a least upper bound of A is an upper bound of A which is less than or equal to every upper bound of A. In general, A may have many lower bounds and many upper bounds, but it is easy to prove (see Problem 1) that a greatest lower bound (or least upper bound) is unique if it exists. It is therefore legitimate to speak of the greatest lower bound and the least upper bound if they exist. We illustrate these concepts in some of the partially ordered sets mentioned above. In Example 1, let the subset A consist of the integers 4 and 6. An upper bound of {4,6} is any positive integer divisible by both 4 and 6. 12, 24, 36, and so on, are all upper bounds of {4,6}. 12 is clearly its least upper bound, for it is less than or equal to (i.e., it divides) every upper bound. The greatest lower bound of any pair of integers in this example is their greatest common divisor, and their least upper bound is their least common multiple—both of which are familiar notions from elementary arithmetic. We now consider Example 2, the real line with its natural order relation. The reader will doubtless recall from his study of calculus that 3 is an upper bound of the set {(1 + 1/n)":n = 1, 2, 3, . . .J} and that its least upper bound is the fundamental constant e = 2.7182.... As we have stated before, it is a basic property of the real line that every non-empty subset of it which has a lower bound (or upper bound) has a greatest lower bound (or least upper bound). ‘There are several items of standard notation and terminology which must be mentioned in connec- tion with this example. Let A be any non-empty set of real numbers. If A has a lower bound, then its greatest lower bound is usually called its infimum and denoted by inf A. Correspondingly, if A has an upper bound, then its least upper bound is called its supremum and written sup A. If A happens to be finite, then inf A and sup A both exist and belong to A. In this case, they are often called the minimum and maximum of A and are denoted by min A and max A. If A consists of two real numbers a; and a», then min A is the smaller of a; and a2, and max A is the larger. Finally, consider Example 3, and let A be any non-empty class of subsets of U. A lower bound of A is any subset of U which is contained in every set in A, and the greatest lower bound of A is the intersection of all its sets. Similarly, the least upper bound of A is the union of all its sets. One of our main aims in this section is to state Zorn’s lemma, an exceedingly powerful tool of proof which is almost indispensable in many parts of modern pure mathematics. Zorn’s lemma asserts that4 Topology if P is a partially ordered set in which every chain has an upper bound, then P possesses a mazimal element. It is not possible to prove this in the usual sense of the word. However, it can be shown that Zorn’s lemma is logically equivalent to the aziom of choice, which states: given any non-empty class of disjoint non-empty sets, a set can be formed which contains precisely one element taken from each set in the given class. The axiom of choice may strike the reader as being intuitively obvious, and in fact, either this axiom itself or some other principle equivalent to Fig. 16. The geometric meaning of f Ag and f V9. it is usually postulated in the logic with which we operate. We therefore assume Zorn’s lemma as an axiom of logic. Any reader who is inter- ested in these matters is urged to explore them further in the literature.! A lattice is a partially ordered set L in which each pair of elements has a greatest lower bound and a least upper bound. If z and y are two elements in L, we denote their greatest lower bound and least upper bound by zany and zvy. These notations are analogous to (and are intended to suggest) the notations for the intersection and union of two sets. We pursue this analogy even further, and call ray and xv y the meet and join of x and y. It is tempting to assume that all properties of intersections and unions in the algebra of sets carry over to lattices, but this is not a valid assumption. Some properties do carry over (see Problem 5), but others, for instance the distributive laws, are false in some lattices. It is easy to see that all four of our examples are lattices. In Example 1, m an is the greatest common divisor of m and n, and m v nis their least common multiple; and in Example 3, AaB = AB and AvB=AUB. In Example 2, if z and y are any two real numbers, then xay is min {z,y} and vy is max {x,y}. In Example 4, fag is 48ee, for example, Wilder [42, pp. 129-132], Halmos [16, secs. 15-16], Birkho [4, p. 42], Sierpinski (37, chap. 6], or Fraenkel and Bar-Hillel [10, p. 44].Sets and Functions 47 the real function defined on X by (fag)(z) = min {f(z),g(z)}, and fvg is that defined by (fv g)(z) = max {f(x),g(z)}. Figure 16 illus- trates the geometric meaning of fag and fvg for two real functions f and g defined on the closed unit interval [0,1]. Let L be a lattice. A sublattice of L is a non-empty subset L, of L with the property that if z and y are in L,, then xay and xv y are also in L,. If L is the lattice of all real functions defined on the closed unit interval, and if L, is the set of all continuous functions in L, then L, is easily seen to be a sublattice of L. If a lattice has the additional property that every non-empty subset has a greatest lower bound and a least upper bound, then it is called a complete lattice. Example 3 is the only complete lattice in our list. There are many distinct types of lattices, and the theory of these systems has a wide variety of interesting and significant applications (see Birkhoff [4]). We discuss some of these types in our Appendix on Boolean algebras. Problems 1, Let A be a non-empty subset of a partially ordered set P. Show that A has at most one greatest lower bound and at most one least upper bound. 2. Consider the set {1, 2,3, 4,5}. What elements are maximal if it is ordered as Example 1? If it is ordered as Example 2? Under what circumstances is Example 4 @ chain? Give an example of a partially ordered set which is not a lattice. Let L be a lattice. If x, y, and z are elements of L, verify the following: raz =2,2V2=2,0AY=YAL,LZVY=YVA, Pre aa(yaz) = (ray) rz, av (yvz) = (evy)v2, (@ay)va=a,(avyrar=a. 6. Let A be a class of subsets of some non-empty universal set U. We say that, A has the finite intersection property if every finite subclass of A has non-empty intersection. Use Zorn’s lemma to prove that if A has the finite intersection property, then it is contained in some maximal class B with this property (to say that B is a maximal class with this property is to say that any class which properly contains B fails to have this property). (Hint: consider the family of all classes which contain A and have the finite intersection property, order this family by class inclusion, and show that any chain in the family has an upper bound in the family.) 7. Prove that if X and Y are any two non-empty sets, then there exists @ one-to-one mapping of one into the other. (Hint: choose an48 10. Topology element z in X and an element y in Y, and establish the obvious one-to-one correspondence between the two single-element sets {x} and {y}; define an extension to be a pair of subsets A of X and B of Y such that {x} C A and {y} C B, together with a one-to-one correspondence between them under which z and y correspond with one another; order the set of all extensions in the natural way; and apply Zorn’s lemma.) Let m and n be any two cardinal numbers (finite or infinite). The statement that m is less than or equal to n (written m < n) is defined to mean the following: if X and Y are sets with m and n elements, then there exists a one-to-one mapping of X into Y. Prove that any non-empty set of cardinal numbers forms a chain when it is ordered in this way. The fact that for any two cardinal numbers one is less than or equal to the other is usually called the compara- bility theorem for cardinal numbers. Let X and Y be non-empty sets, and show that the cardinal number of X is less than or equal to the cardinal number of Y + there exists @ mapping of Y onto X. Let {X;} be any infinite class of countable sets indexed by the ele- ments 7 of an index set 7, and show that the cardinal number of UX; is less than or equal to the cardinal number of J. (Hint: if I is only countably infinite, this follows from Problem 6-2, and if J is uncountable, Zorn’s lemma can be applied to represent it as the union of a disjoint class of countably infinite subsets.)CHAPTER TWO Metric Spaces Classical analysis can be described as that part of mathematics which begins with calculus and, in essentially the same spirit, develops similar subject matter much further in many directions. It is a great nation in the world of mathematics, with many provinces, a few of which are ordinary and partial differential equations, infinite series (especially power series and Fourier series), and analytic functions of a complex variable. Each of these has experienced enormous growth over a long history, and each is rich enough in content to merit a lifetime of study. In the course of its development, classical analysis became so complex and varied that even an expert could find his way around in it only with difficulty. Under these circumstances, some mathematicians became interested in trying to uncover the fundamental principles on which all analysis rests. This movement had associated with it many of the great names in mathematics of the last century: Riemann, Weierstrass, Cantor, Lebesgue, Hilbert, Riesz, and others. It played a large part in the rise to prominence of topology, modern algebra, and the theory of measure and integration; and when these new ideas began to percolate back through classical analysis, the brew which resulted was modern analysis. As modern analysis developed in the hands of its creators, many a major theorem was given a simpler proof in a more general setting, in an effort to lay bare its inner meaning. Much thought was devoted to analyzing the texture of the real and complex number systems, which are the context of analysis. It was hoped—and these hopes were well founded —that analysis could be clarified and simplified, and that stripping away "950 Topology superfluous underbrush would give new emphasis to what really mattered from the point of view of the underlying theory.! Analysis is primarily concerned with limit processes and con- tinuity, so it is not surprising that mathematicians thinking along these lines soon found themselves studying (and generalizing) two elementary concepts: that of a convergent sequence of real or complex numbers, and that of a continuous function of a real or complex variable. We remind the reader of the definitions. First, a sequence of real numbers is said to be convergent if there exists a real number x (called the limit of the sequence) such that, given e > 0, a positive integer no can be found with the property that {an} = {21,22 te - n> m= |t.— 2] 0 there exists 5 > 0 such that win X and |z — | < = |f(z) — fla) <6, and f is said to be continuous if it is continuous at each point of X. When X is an interval, this definition gives precise expression to the intuitive requirement that f have a graph without breaks or gaps. The corresponding definitions for sequences of complex numbers and complex functions of a complex variable are word for word the same. Our purpose in giving these definitions in detail here is a simple one. We wish to point out explicitly that each is dependent for its meaning on the concept of the absolute value of the difference between two real or complex numbers. We wish to observe also that this absolute value is the distance between the numbers when they are regarded as points on the real line or in the complex plane. In many branches of mathematics—in geometry as well as analysis— it has been found extremely convenient to have available a notion of distance which is applicable to the elements of abstract sets. A metric space (as we define it below) is nothing more than a non-empty set 1 We illustrate these points in Appendix 1, where one of the basic existence theorems in the theory of differential equations is given a brief and uncluttered proof which depends only on the ideas of this chapter.Metric Spaces 51 equipped with a concept of distance which is suitable for the treatment of convergent sequences in the set and continuous functions defined on the set. Our purpose in this chapter is to develop in a systematic manner the main elementary facts about metric spaces. These facts are impor- tant for their own sake, and also for the sake of the motivation they provide for our later work on topological spaces. 9, THE DEFINITION AND SOME EXAMPLES Let X be a non-empty set. A metric on X is a real function d of ordered pairs of elements of X which satisfies the following three conditions: (1) d(y) 2 0, and d(@z,y) = Oa = y; (2) d(zy) = d(y,2) (symmetry); (3) d(x,y) < d(z,z) + d(z,y) (the triangle inequality). The function d assigns to each pair (2,y) of elements of X a non-negative real number d(z,y), which by symmetry does not depend on the order of the elements; d(z,y) is called the distance between z and y. A metric space consists of two objects: a non-empty set X and a metric d on X. The elements of X are called the points of the metric space (X,d). When- ever it can be done without causing confusion, we denote the metric space (X,d) by the symbol X which is used for the underlying set of points. One should always keep in mind, however, that a metric space is not merely a non-empty set: it is a non-empty set together with a metric. It often happens that several different metrics can be defined on a single given non-empty set, and in this case distinct metrics make the set into distinct metric spaces. There are many different kinds of metric spaces, some of which play very significant roles in geometry and analysis. Our first example is rather trivial, but it is often useful in showing that certain statements we might wish to make are not true. It also shows that every non- empty set can be regarded as a metric space. Example t. Let X be an arbitrary non-empty set, and define d by 0 ife=y, aay) = 1 ife xy. The reader can easily see for himself that this definition yields a metric on X. Our next two examples are the fundamental number systems of mathematics.52 Topology Example 2. Consider the real line R and the real function |z| defined on R. Three elementary properties of this absolute value function are important for our purposes: @) |o| > 0, and |2| = 02 = 0; Gi) |=21 = Ia\; ii) le + yl < lel + byl. We now define a metric on R by (zy) = |x — 9}. This is called the usual metric on R, and the real line, as a metric space, is always understood to have this as its metric. The fact that d actually is a metric follows from the three properties stated above. This is a piece of reasoning which occurs frequently in our work, so we give the details. By (i), d(z,y) = |z — y| is a non-negative real number which equals02—y=Oerxr=y. By (ii), d(zy) = |z — yl = |-& — 2) = ly — 2] = dy,2). And by (ii), day) = |e -yl = |@-2+e@-Misle-a+le-yl = d(zz) + d(z,y). Example 3. Consider the complex plane C. We mentioned C briefly in Sec. 4, and we described the sense in which it can be identified as a set with the coordinate plane R?. We nowgive a somewhat fuller discussion. If z is a complex number, and if z = a + ib where a and 6 are real numbers, then a and b are called the real part and the imaginary part of z and are denoted by R(z) and I(z). Two complex numbers are said to be equal if their real and imaginary parts are equal: a+i=c+id@a=candb=d. We add (or subtract) two complex numbers by adding (or subtracting) their real and imaginary parts, and we multiply them by multiplying them out as in elementary algebra and replacing 7? by —1 wherever it appears: (a+) + (C+ id) = @ tc) +i +d), and (a + ib)(c + id) = ac + iad + tbe + bd = (ace — bd) + i(ad + be).Metric Spaces 53 Division is carried out in accordance with ati _ @+ tb)(c— td) _ (ac + bd) + tbe ~ ad) e+id (+ td)(c — id) oe + d? = wtbd , be ~ ad “ea eta where c* + d? is required to be non-zero. If z = a + tb is a complex number, then its negative —z and its conjugate 2 are defined by ~2= (-a) + i(-2) and = a+ %(—b), which are usually written more informally as —z= —a-— ibandz =a -— ib. It is easy to see that ene, 2 Re = 242 and I(2) The real line F is usually regarded as part of the complex plane: R= {z:I(z) = 0} = {2:2 =z}. Simple calculations show directly that atasatm, a=", and F=2, The origin, or zero, is the complex number 0 = 0 + 40. The ordinary distance from z = a + 4b to the origin is defined by le] = (ot +O, le| is called the absolute value of z, and it is easy to see that and el? = 22. The usual metric on C is defined by (21,22) = |zr — aa. Exactly as in Example 2, the fact that this is a metric is a consequence of the following properties of the real function |z|: @) le] > 0, and [ze] = Oz = 0; Gi) |-2| = lal; Gii) ler + 2a] < las] + [eal Properties (i) and (ii) are obvious. Since (—1)z, property (ii) is also a special case of the fact that lerzel = lal leal,54 Topology which we prove by means of levee]? = eveszaza = zatazaza = |eil*leel* = (leil [2el)*. If we use the fact that |R(z)| < |e| for any z, property (iii) follows directly from la t+ al? = (@+a)G@iFa)=@Gtalit+e) _ eta + ante + arte + taza = leal® + leal® + (exe + 2122) = lel? + les? + 2220) S fal? + leak? + laseal_ = |ail? + leal? + 2lail lel = lea? + les|® + 2lzi] feal = (lel + leal)®. Whenever the complex plane C is mentioned as a metric space, its metric is always assumed to be the usual metric defined above. The remaining examples to be given in this section fit a common pattern, which we have tried to exhibit in our discussion of Examples 2 and 3. We now point out several major features of this pattern, so that the reader can see clearly how it applies in the slightly more complicated examples that follow. I, The elements of each space can be added and subtracted in a natural way, and every element has a negative. Each space contains a special element, denoted by 0 and called the origin, or zero element. II. In each space there is defined a notion of the distance from an arbitrary element to the origin, that is, a notion of the “size” of an arbitrary element. The size of an element z is a real number denoted below by ||z|| and called its norm. Our use of the double vertical bars is intended to emphasize that the norm is a generalization of the absolute value functions in Examples 2 and 3, in the sense that it satisfies the following three conditions: (i) |lz|| 20, and |x| =O02=0; (ii) all = [ells Gii) lle + yll < Mell + Hull III. Finally, each metric arises as the norm of the difference between two elements: d(z,y) = ||z — yl]. As in Example 2, the fact that this is a metric follows from the properties of the norm listed in II. This metric is called the metric induced by the norm. The knowledgeable reader will see at once that we are describing here (though incompletely and imprecisely) the concept of a normed linear space. Most of the metric spaces of major importance in analysis are of this type.Metric Spaces 55 Example 4. Let f be a real function defined on the closed unit.interval [0,1]. We say that f is a bounded function if there is a real number K such that |f(x)| < K for every z€[0,1]. This concept is familiar to the reader from elementary analysis, as is that of the continuity of f as defined in the introduction to this chapter. The underlying set of points in this example is the set of all bounded continuous real functions defined on the closed unit interval. Actually, the boundedness of such a function is a consequence of its other properties, but at this stage we assume it explicitly. If f and g are two such functions, we add and subtract them, and form negatives, pointwise: f+ 9)(@) = f@) + (2); GF - @) = f@) ~— 9@); (-A@) = —f@). The origin (denoted by 0) is the constant function which is identically zero; O(z) =0 for all x ¢ [0,1]. We define the norm of a function f by Il = fo Wel as, and the induced metric by a(f.0) = If — oll = fo Ise2) — o(@)l de. The integral involved in this definition is the Riemann integral of ele- mentary calculus. Properties (i) and (ii) of the norm are easy to prove, and (iii) follows from toll = fy Wie) + orl de < f (seal + Iota) ax = fp elas + ft lole)| de = lal + loll Example 5. The set of points in the preceding example—that is, the set of all bounded continuous real functions defined on the closed unit interval—has another metric which is far more important for our pur- poses. It is defined by means of Ill = sup {|f(@)]:2 € [0,1]}, which we usually write more briefly as fll = sup [f@)1, and d(fg) = If — all = sup [f(z) — 9).56 Topology Properties (i) and (ii) of the norm are obvious, and in Problem 5 we ask the reader to prove (iii) in a slightly more general form. This example is typical of a large class of metric spaces which will play a major role in all our work throughout the rest of this book. We denote this space by e(0,1). So much for the present for specific examples. We now turn to several fundamental principles relating to metric spaces in general. Let X be a metric space with metric d. Let Y be an arbitrary non- empty subset of X. If the function d is considered to be defined only for points in Y, then (Y,d) is evidently itself a metric space. Y, with d restricted in this way, is called a subspace of X. This technique of forming subspaces of a given metric space enables us to obtain an infinity of further examples from the handful described above. For instance, the closed unit interval [0,1] is a subspace of the real line, as is the set consisting of all the rational points; and the unit circle, the closed unit disc, and the open unit disc are subspaces of the complex plane. Also, the real line itself is a subspace of the complex plane. It is desirable at this stage to introduce the extended real number system, by which we mean the ordinary real number system FR with the symbols -o and +o adjoined. An extended real number is thus a real number or one of these symbols. We say (by definition) that -o <4; also, if z is any real number, then —ao itisa union of open spheres. Proor. We assume first that G is open, and we show that it is a union of open spheres. If G is empty, it is the union of the empty class of open spheres. If G@ is non-empty, then since it is open, each of its points is the center of an open sphere contained in it, and it is the union of all the open spheres contained in it. We now assume that G is the union of a class S of open spheres. We must show that G is open. If S$ is empty, then G is also empty, and by Theorem A, G is open. Suppose that S is non-empty. G is also non- empty. Let xz bea point inG. Since G is the union of the open spheres in S, x belongs to an open sphere S,(zo) in S. By Theorem B, z is the center of an open sphere S,,(z) C 8,(z0). Since S,(0) € G, S,,(z) G Gand we have an open sphere centered on z and contained in G. G is there- fore open. The fundamental properties of the open sets in a metric space are those stated in Theorem D. Let X be a metric space. Then (1) any union of open sets in X is open; and (2) any finite intersection of open sets in X is open. proor. To prove (1), let {G:} be an arbitrary class of open sets in X. We must show that G = U,G; is open. If {G;} is empty, then @ is62 Topology empty, and by Theorem A, Gis open. Suppose that {G;} is non-empty. By Theorem C, each G; (being an open set) is a union of open spheres; G is the union of all the open spheres which arise in this way; and by another application of Theorem C, @ is open. To prove (2), let {Gs} be @ finite class of open sets in X. We must show that G = OG; is open. If {G;} is empty, then @ = X; and by Theorem A, G is open. Suppose that {G,} is non-empty and that {Gs} = (Gi, Gs, . . . , Gn} for some positive integer n. If @ happens to be empty, then it is open by Theorem A, so we may assume that G is non-empty. Let z bea point in G. Since z is in each Gi, and each G; is open, for each ¢ there is a positive real number r; such that S,,(z) € Gi. Let r be the smallest number in the set. {r:, 72, . . . , 7a}. This number ris a positive real number such that S,(z) € S,,(z) for each ¢, so S,(z) © G; for each ¢, and therefore S.(z) © G. Since S,(z) is an open sphere centered on a and contained in G, G@ is open. The above theorem says that the class of all open sets in a metric space is closed under the formation of arbitrary unions and finite inter- sections. The reader should clearly understand that Theorem A is an immediate consequence of this statement, since the empty set is the union of the empty class of open sets and the full space is its intersection. The limitation to finite intersections in this theorem is essential. To see this, it suffices to consider the following sequence of open intervals on the real line: (-1,D, (—34,4), (-14,%), « - - The intersection of these open sets is the set {0} consisting of the single point 0, and this set is not open. In an arbitrary metric space, the structure of the open sets can be very complicated indeed. Theorem C contains the best information available in the general case: each open set is a union of open spheres. In the case of the real line, however, a description can be given of the open sets which is fairly explicit and reasonably satisfying to the intuition. Theorem E. Every non-empty open set on the real line is the union of a countable disjoint class of open intervals. proor. Let G be a non-empty open subset of the real line. Let 2 bea point of G. Since G is open, z is the center of a bounded open interval contained in G. Define J, to be the union of all the open intervals which contain z and are contained in G. The following three facts are easily proved: J, is an open interval (by Theorem D and Problem 9-6) which contains z and is contained inG; Jz contains every open interval which contains z and is contained in G; and if y is another point in Js,Metric Spaces 63 then I, = Iy. We next observe that if z and y are any two distinct points of G, then J, and J, are either disjoint or identical; for if they have a common point z, then J. = I, and I, = I,, 80 Jz = Iy. Consider the class I of all distinct sets of the form J. for points z in G. This is a disjoint class of open intervals, and G is obviously its union. It remains to be proved that | is countable. Let G, be the set of rational points in G. G, is clearly non-empty. We define a mapping f of G, onto | as follows: for each r in G,, let f(r) be that unique interval in | which con- tains r. G, is countable by Problem 6-7, and the fact that | is countable follows from Problem 6-8. A firm grasp of the ideas involved in the theory of metric spaces depends on one’s capacity to ‘“‘see” these spaces with the mind’s eye. The complex plane is perhaps the best metric space to use as a model from which to absorb this necessary intuitive understanding. When we consider an unspecified set A of com- plex numbers, we usually imagine it as a region bounded by a curve, as in Fig. 19. We think of the point a, which is completely surrounded by points of A, as being “inside” the set A, or in its “interior,” while zz ison the “boundary” of A. More pig. 19. A sot A of complex numbers precisely, z: is the center of some with interior point z; and boundary open sphere contained in A, and point zs. each open sphere centered on zz in- tersects both A and its complement A’. We formulate these ideas for a general metric space in the next paragraph and at the end of the next section. Let X be an arbitrary metric space, and let A be a subset of X. A point in A is called an interior point of A if it is the center of some open sphere contained in A; and the interior of A, denoted by Int(A), is the set of all its interior points. Symbolically, Int(A) = {z:2€ A and S,(z) C A for some r}. The basic properties of interiors are the following: (1) Int(A) is an open subset of A which contains every open subset of A (this is often expressed by saying that the interior of A is the largest open subset of A); (2) A is open = A = Int(A); (3) Int(A) equals the union of all open subsets of A.64 Topology The proofs of these facts are quite easy, and we ask the reader to fill in the details as an exercise (see Problem 8). Problems 1 10. Let X be a metric space, and show that any two distinct points of X can be separated by open spheres in the following sense: if x and y are distinct points in X, then there exists a disjoint pair of open spheres each of which is centered on one of the points. Let X be a metric space. If {2} is a subset of X consisting of a single point, show that its complement {z}’ is open. More gen- erally, show that A’ is open if A is any finite subset of X. Let X be a metric space and S,(z) the open sphere in X with center zand radius r, Let A be a subset of X with diameter less than r which intersects S,(z). Prove that A C S:,(z). Let X be a metric space. Show that every subset of X is open each subset of X which consists of a single point is open. Let X be a metric space with metric d, and let d, be the metric defined in Problem 9-1. Show that the two metric spaces (X,d) and (X,d,) have precisely the same open sets. (Hint: show that they have the same open spheres with one exception. What is this exception?) If X =X. X X: X ++: X X, is the product in Problem 9-4, and if d and d are the metrics on X defined in that problem, show that the two metric spaces (X,d) and (X,d) have precisely the same open sets. Observe that in this case the spaces do not have the same open spheres. Let Y be a subspace of a metric space X, and let A be a subset of the metric space Y. Show that A is open as a subset of Y © it is the intersection with Y of a set which is open in X. Prove the statements made in the text about interiors. Describe the interior of each of the following subsets of the real line: the set of all integers; the set of all rationals; the set of all irrationals; (0,1); [0,1]; (0,1) U {1,2}. Do the same for each of the following subsets of the complex plane: {z:|z| < 1}; {z:lz] < 1}; {z:I(z2) = 0}; {2:R@) is rational}. Let A and B be two subsets of a metric space X, and prove the following: (a) Int(A) U Int(B) C Int(A U B); (6) Int(A) M Int(B) = Int(A A B). Give an example of two subsets A and B of the real line such that Int(A) U Int(B) # Int(A U B).Metric Spaces 65 11. CLOSED SETS Let X be a metric space with metric d. If A is a subset of X, a point z in X is called a limit point of A if each open sphere centered on x contains at least one point of A different from x. The essential idea here is that the points of A different from z get “arbitrarily close” to 2, or “pile up” at x. The subset {1, 4, }4, . . .} of the real line has 0 as a limit point; in fact, 0 is its only limit point. The closed-open interval 0,1) has 0 as a limit point which is in the set and 1 as a limit point which is not in the set; further, every real number z such that 0 < x < 1 is also a limit point of this set. The set of all integral points on the real line has no limit points at all, whereas every real number is a limit point of the set of all rationals. In Example 9-1, every open sphere of radius less than 1 consists only of its center, so no subset of this space has any limit points. A subset F of the metric space X is called a closed set if it contains each of its limit points. In rough terms, a set is closed if its points do not get arbitrarily close to any point outside of it. Among the subsets of the real line mentioned in the preceding paragraph, only the set of integral points is closed. In Example 9-1, every subset is closed. Theorem A. In any metric space X, the empty set B and the full space X are closed sets. PRooF. The empty set has no limit points, so it contains them all and is therefore closed. Since the full space X contains all points, it auto- matically contains its own limit points and thus is closed. The following theorem characterizes closed sets in terms of open sets. We already know a good deal about open sets, so this characterization provides us with a useful tool for establishing properties of closed sets. Theorem B. Let X be a metric space. A subset F of X is closed + tts complement F’ is open. proor. Assume first that F is closed. We show that F’ is open. If F’ is empty, it is open by Theorem 10-A, so we may suppose that F” is non-empty. Let xbeapointin F’. Since F is closed and zis not in F, 2 is not a limit point of F. Since x is not in F and is not a limit point of F, there exists an open sphere S,(z) which is disjoint from F. S,(x) is an open sphere centered on z and contained in F”’, and since x was taken to be any point of F’, F’ is open. We now assume that ¥” is open and show that F is closed. The only way F can fail to be closed is to have a limit point in F’. This cannot66 Topology happen, for since F’ is open, each of its points is the center of an open sphere disjoint from F, and no such point can be a limit point of F. If zo is a point in our metric space X, and r is a non-negative real number, the closed sphere S,{2o] with center x and radius r is the subset of X defined by S[zo] = {x:d(z,ao) r, 7: = d(z,20) — r is a positive real number. We take r, as the radius of an open sphere S,,(z) centered on z, and we show that. S,[x0]’ is open by showing that S,,(x) C S,[{xo]’. Let y be a point in S,,(z), so that d(y,x) <1; On the basis of this and the fact that A(xo,2) < d(xo,y) + d(y,x), we see that d(y,a0) 2 d(x,r0) — dYy,x) > d(a,a0) — 1 = (a0) — (dear) — r] = 7, so that y is in S,[x9]’. The main general facts about closed sets are those given in our next theorem. Theorem D. Let X be a metric space. Then (1) any intersection of closed sets in X is closed; and (2) any finite union of closed sets in X is closed. prooF. By virtue of Eqs. 2-(2) and Theorem B above, this theorem is an immediate consequence of Theorem 10-D. We prove (1) as follows. If {F;} is an arbitrary class of closed subsets of X and F = Fi, then by Theorem B, F is closed if F’ is open; but F’ = U;F'/ isopen by Theorem 10-D, since by Theorem B each F;’ is open. The second statement is proved similarly. In Theorem E of the previous section, we gave an explicit charac- terization of the open sets on the real line. We now consider the struc- ture of its closed sets. Among the simplest closed sets on the real line are the closed intervals (which are the closed spheres) and finite unionsMetric Spaces 67 of closed intervals. Finite sets are included among these, since a set consisting of a single point is a closed interval with equal end-points. What is the character of the most general closed set on the real line? Since closed sets are the complements of open sets, Theorem 10-E gives a complete answer to this question: the most general proper closed subset. of the real line is obtained by remov- ing a countable disjoint class of open intervals. This process sounds inno- cent enough, but in fact it leads to some rather curious and complicated examples. One of these examples is of particular importance. It was studied by Cantor and is usually —.----——--—--- 2-0 called the Cantor set. To construct the Cantor set, we proceed as follows (see Fig. 20). es-es---ss-0.—--————ee-se---ee-s0 First, denote the closed unit interval Fig. 20. The Cantor set. [0,1] by Fi. Next, delete from Fi the open interval (34,3) which is its middle third, and denote the remaining closed set by F2. Clearly, Fr = (0,)4] U [34,1]. Next, delete from F, the open intervals (14,36) and (74,36), which are the middle thirds of its two pieces, and denote the remaining closed set by Fy. It is easy to see that Fs = (0,36) U [96,24] U [36,76] U [86,1]. If we continue this process, at each stage deleting the open middle third of each closed interval remaining from the previous stage, we obtain a sequence of closed sets F’,, each of which contains all its successors. The Cantor set F is defined by F=ORi Fa, and it is closed by Theorem D. F consists of those points in the closed unit interval [0,1] which “ultimately remain” after the removal of all the open intervals (14,24), (14,36), (14,86), - .. . What points do remain? F clearly contains the end-points of the closed intervals which make up each set Fn: 0, 1, 24, 36, 38, 3S, 26, 36, - - - Does F contain any other points? We leave it to the reader to verify that }4 isin F and is not an end-point. Actually, F contains a multitude of points other than the above end-points, for the set of these end-points68 = Topology is clearly countable, while the cardinal number of F itself is c, the cardinal number of the continuum. To prove this, it suffices to exhibit a one-to- one mapping f of [0,1) into F. We construct such a mapping as follows. Let x be a point in (0,1), and let z = .bibeb; . . . be its binary expansion (see Sec. 7). Eachb,iseitherOorl. Lett, = 2bn, and regard -titsts . . . as the ternary expansion of a real number f(z) in (0,1). The reader will easily convince himself that f(z) is in the Cantor set F: since ¢; is 0 or 2, S(@) is not in (34,34); since t is 0 or 2, f(z) is not in (34,36) or [74,36); ete. Also, it is easy to see that the mapping f: [0,1) > F is one-to-one. According to this, F contains exactly as many points as the entire closed unit interval [0,1]. It is interesting to compare this conclusion with the fact that the sum of the lengths of all the open intervals removed is precisely 1, since M+R tS to a1 Tt is also interesting to observe (by doing a little arithmetic) that Fes is the union of 16,777,216 disjoint closed intervals of the same length which are rather irregularly distributed along [0,1]. These facts may suffice to indicate that the Cantor set is a very intricate mathematical object and is just the sort of thing mathematicians delight in. We shall encounter this set again from time to time, for its properties illustrate several phenomena discussed in later sections. We conclude this section by defining two additional concepts which are often useful. Let X be an arbitrary metric space, and let A be a subset of X. The closure of A, denoted by A, is the union of A and the set of all its limit points. Intuitively, A is A itself together with all other points in X which are arbitrarily close to A. As an example, if A is the open unit disc {z:|z| <1} in the complex plane, then A is the closed unit disc {z:|z| <1}. The main facts about closures are the following: (1) A is a closed superset of A which is contained in every closed superset of A (we express this by saying that A is the smallest closed superset of A); (2) A is closed + A = A; (3) A equals the intersection of all closed supersets of A. It is a routine exercise to prove these statements, and we leave this task to the reader (in Problem 6). Our second concept relates to the discussion of Fig. 19 given at the end of the previous section. Again, let X be a metric space and A a subset of X. A point in X is called a boundary point of A if each open sphere centered on the point intersects both A and A’, and the boundary of A is the set of all its boundary points. This concept possesses the following properties;Metric Spaces 69 (1) the boundary of A equals A 1 A’; (2) the boundary of A is a closed set; (8) A is closed + it contains its boundary. We ask the reader to give the proofs in Problem 11. Problems 1, NO Let X be a metric space, and extend Problem 10-1 by proving the following statements: (a) any point and disjoint closed set in X can be separated by open sets, in the sense that if z is a point and F a closed set which does not contain z, then there exists a disjoint pair of open sets G, and G, such that x¢G, and F C G,; (b) any disjoint pair of closed sets in X can be separated by open sets, in the sense that if F; and F; are disjoint closed sets, then there exists a disjoint pair of open sets G; and G; such that F, CG, and F, CG. Let X be a metric space, and let A be a subset of X. If zis a limit point of A, show that each open sphere centered on z contains an infinite number of distinct points of A. Use this result to show that a finite subset of X is closed. Show that a subset of a metric space is bounded © it is non-empty and is contained in some closed sphere. Give an example of an infinite class of closed sets whose union is not, closed. Give an example of a set which (a) is both open and closed; (6) is neither open nor closed; (c) contains a point which is not a limit point of the set; and (d) contains no point which is not a limit point of the set. Describe the interior of the Cantor set. Prove the statements made in the text about closures. Let X be a metric space and A a subset of X. Prove the following facts: (a) A’ = Int(A); (0) A = {x:d(a,A) = 0}. Describe the closure of each of the following subsets of the real line: the integers; the rationals; the Cantor set; (0,+ ©); (—1,0) U (0,1). Do the same for each of the following subsets of the complex plane: {z:|z| is rational}; {z:1/R(z) is an integer}; {z:|z| <1 and I(z) <0}. Let X be a metric space, let z be a point of X, and let r be a positive real number. One is inclined to believe that the closure of S,(z) must equal S,{z]. Give an example to show that this is not neces- sarily true. (Hint: see Example 9-1.)70 Topology 10. Let X be a metric space, and let G be an open set. in X. Prove that G is disjoint from a set A & G is disjoint from A. 11, Prove the facts about boundaries stated in the text. 12. Describe the boundary of each of the following subsets of the real line: the integers; the rationals; [0,1]; (0,1). Do the same for each of the following subsets of the complex plane: {z:|z| < 1}; {z:|z| < 1}; {e:Z(z) > O}. 13. Let X be a metric space and A a subset of X. A is said to be dense (or everywhere dense) if A = X. Prove that A is dense = the only closed superset of A is X < the only open set disjoint from A is $+ A intersects every non-empty open set + A intersects every open sphere. 12, CONVERGENCE, COMPLETENESS, AND BAIRE’S THEOREM As we emphasized in the introduction to this chapter, one of our main aims in considering metric spaces is to study convergent sequences in @ context more general than that of classical analysis. The fruits of this study are many, and among them is the added insight gained into ordinary convergence as it is used in analysis. Let X be a metric space with metric d, and let {aa} = {ay B20) Tay so} be a sequence of points in X. We say that {z,} is convergent if there exists a point x in X such that either (1) for each «> 0, there exists a positive integer no such that n > 1 = d(xn,2) < €; or equivalently, (2) for each open sphere S,(z) centered on 2, there exists a positive integer mo» such that 2, is in S.(x) for all n > mo. The reader should observe that the first condition is a direct generaliza- tion of convergence for sequences of numbers as defined in the introduc- tion, and that the second can be thought of as saying that each open sphere centered on z contains all points of the sequence from some place on. If we rely on our knowledge of what is meant by a convergent sequence of real numbers, the statement that {z,} is convergent can equally well be defined as follows: there exists a point z in X such that d(z,,2) > 0. We usually symbolize this by writing In 2, and we express it verbally by saying that xz, approaches x, or that ta converges tox. It is easily seen from condition (2) and Problem 10-1 that the point x in this discussion is unique, that is, that 2, > y with y #Metric Spaces 7” is impossible. The point z is called the limit of the sequence {z,}, and we sometimes write tz, — 2 in the form lim tq = 2. The statements az and limz,=2 mean exactly the same thing, namely, that {z,} is a convergent sequence with limit z. Every convergent sequence {z,} has the following property: for each «> 0, there exists a positive integer m» such that m,n > 1p => d(tm,tn) < « For if ,— 2, then there exists a positive integer mo such that n > 1 = d(tn,z) < ¢/2, and from this we see that mn > 9 => d(tmtn) < d(m,z) + d(a,tn) < ¢/2 + ¢/2 = 6 A sequence with this property is called a Cauchy sequence, and we have just shown that every convergent sequence is a Cauchy sequence. Loosely speaking, this amounts to the statement that if the terms of a sequence approach a limit, then they get close to one another. It is of basic importance to understand that the converse of this need not be true, that is, that a Cauchy sequence is not necessarily convergent. As an example, consider the subspace X = (0,1] of the real line. The sequence defined by z, = 1/n is easily seen to be a Cauchy sequence in this space, but it is not convergent, since the point 0 (which it wants to converge to) is not a point of the space. The difficulty which arises in this example stems from the fact that the notion of a convergent sequence is not intrinsic to the sequence itself, but also depends on the structure of the space in which it lies. A convergent sequence is not convergent “on its own”; it must converge to some point in the space. Some writers emphasize the distinction between convergent sequences and Cauchy sequences by calling the latter ‘intrinsically convergent” sequences. A complete metric space is a metric space in which every Cauchy sequence is convergent. In rough terms, a metric space is complete if every sequence in it which tries to converge is successful, in the sense that, it finds a point in the space to converge to. The space (0,1] mentioned above is not complete, but it evidently can be made so by adjoining the point 0 to it to form the slightly larger space [0,1]. As a matter of fact, any metric space, if it isn’t already complete, can be made so by suitably adjoining additional points. We outline this process in a problem at the end of Sec. 14. It is a fundamental fact of elementary analysis that the real line is a complete metric space. The complex plane is also complete, as we see from the following argument. Let {z,}, where 2, = Gn + tba, be a Cauchy sequence of complex numbers. Then {a,} and {b,} are them-72 Topology selves Cauchy sequences of real numbers, since [dm — Gn] < [2m — 2nl and lbm — Bal < zm — 2a]. By the completeness of the real line, there exist real numbers a and b such that a, a and b,—b. If we now put 2 = a + ib, then we see that 2x — 2 by means of len — 2] = [(@n + thn) — (a + 10)| = |(a, — a) + i(b, — b)| S lan — al + |b, — | and the fact that both final terms on the right approach 0. The com- pleteness of the complex plane thus depends directly on the completeness of the realline. The metric space defined in Example 9-1 is also complete; for in this space a Cauchy sequence must be constant (i.e., it must consist, of a single point repeated) from some place on, and it converges with that point as its limit. The first three of the five metric spaces given as examples in Sec. 9 are therefore complete. What about the last two? We ask the reader to show in Problem 5 that Example 9-4 is not. complete. The problem of completing this space leads to the modern theory of Lebesgue integration, and it would carry us too far afield to pursue this matter to its natural conclusion. On the other hand, the space €[0,1] defined in Example 9-5 is com- plete. We prove this in a more general form in Sec. 14. The complete- ness of this space, and of others similar to it, is one of the major focal points of topology and modern analysis. The terms limit and limit point are often a source of confusion for people not thoroughly accustomed to them. On the real line, for instance, the constant sequence {1, 1,..., 1, .. .} is convergent with limit 1; but the set of points of this sequence is the set consisting of the single element 1, and by Problem 11-2, the point 1 is not a limit point of this set. The essence of the matter is that a sequence of points in a set is not a subset of the set: it is a function defined on the positive integers with values in the set, and is usually specified by listing its values, as in {an} = {a1 t ... 2m... -}, where z, is the value of the function at the integer n. A sequence may have a limit, but cannot have a limit point; and the set of points of a sequence may have a limit point, but cannot have a limit. The following theorem relates these concepts to one another and is a useful tool for some of our later work. Theorem A. If a convergent sequence in a metric space has infinitely many distinct points, then tts limit is a limit point of the set of points of the sequence.Metric Spaces 73 proor. Let X be a metric space, and let {z,} be a convergent sequence in X with limit z. We assume that « is not a limit point of the set of points of the sequence, and we show that it follows from this that the sequence has only finitely many distinct points. Our assumption implies that there exists an open sphere S,(x) centered on z which con- tains no point of the sequence different fromz. However, since z is the limit of the sequence, all x,’s from some place on must lie in S,(x), hence must coincide with z. From this we see that there are only finitely many distinct points in the sequence. Our next theorem guarantees the completeness of many metric spaces which arise as subspaces of complete metric spaces. Theorem B. Let X be a complete metric space, and let ¥ be a subspace of X. Then Y is complete = tt ts closed. PROOF. We assume first that Y is complete as a subspace of X, and we show that it is closed. Let y be a limit point of Y. For each positive integer n, Sij.(y) contains a point y, in Y. It is clear that {y,} con- verges to y in X and is a Cauchy sequence in Y, and since Y is complete, yisin Y. Y is therefore closed. We now assume that Y is closed, and we show that it is complete. Let {y,} be a Cauchy sequence in Y. It is also a Cauchy sequence in X, and since X is complete, {y.} converges to a point zin X. We show that. awisin Y. If {yn} has only finitely many distinct points, then z is that point infinitely repeated and is thusin Y. On the other hand, if {y,} has infinitely many distinct points, then, by Theorem A, z is a limit point of the set of points of the sequence; it is therefore also a limit point of Y, and since Y is closed, z is in Y. A sequence {A,} of subsets of a metric space is called a decreasing sequence if 4,.2D4:24;2°°°.- The following theorem gives conditions under which the intersection of such a sequence is non-empty. Theorem C (Cantor's Intersection Theorem). Let X be a complete metric space, and let {F,,} be a decreasing sequence of non-empty closed subsets of X such that d(F,) > 0. Then F = (\%_, F, contains exactly one point. pRooF. It is first of all evident from the assumption d(F,) — 0 that F cannot contain more than one point, so it suffices to show that F is non- empty. Let 2, be a point in F,. Since d(F,) 0, {z,} is a Cauchy sequence. Since X is complete, {z,} has a limit x. We show that z isin F, and for this it suffices to show that z is in F,, for a fixed but arbitrary no If {an} has only finitely many distinct points, then z is that point74 Topology infinitely repeated, and is therefore in F,,. If {za} has infinitely many distinct points, then z is a limit point of the set of points of the sequence, it is a limit point of the subset {z,: 7 > no} of the set of points of the sequence, it is a limit point of F,,, and thus (since F,, is closed) it is in Fy,. A subset A of a metric space is said to be nowhere dense if its closure has empty interior. It is easy to see that A is nowhere dense <= A does not contain any non-empty open set < each non-empty open set has a non-empty open subset disjoint from A + each non-empty open set has a non-empty open subset disjoint from A + each non-empty open set con- tains an open sphere disjoint from A. Ifa nowhere dense set is thought of as a set which doesn’t cover very much of the space, then our next theorem says that a complete metric space cannot be covered by any sequence of such sets. Theorem D. If {Aq} is a sequence of nowhere dense sets in a complete metric space X, then there exists a point in X which is not in any of the A,’s. proor. For the duration of this proof, we abandon our usual notations for open spheres and closed spheres. Since X is open and A, is nowhere dense, there is an open sphere S, of radius less than 1 which is disjoint from A;. Let F, be the concentric closed sphere whose radius is one-half that of §;, and consider its interior. Since A is nowhere dense, Int(F;) contains an open sphere S; of radius less than }4 which is disjoint from A:. Let Fz be the concentric closed sphere whose radius is one-half that of S2, and consider its interior. Since Az is nowhere dense, Int(F2) contains an open sphere S; of radius less than 14 which is disjoint from A;. Let F3 be the concentric closed sphere whose radius is one-half that of S;. Con- tinuing in this way, we get a decreasing sequence {F,} of non-empty closed subsets of X such that d(F,) > 0. Since X is complete, Theorem C guarantees that there exists a point z in X which is in all the F,’s. This point is clearly in all the S,’s, and therefore (since S, is disjoint from A,) it is not in any of the A,’s. For our purposes, the following equivalent form of Theorem D is often more convenient. Theorem E. If a complete metric space is the union of a sequence of its subsets, then the closure of at least one set in the sequence must have non- empty interior. Theorems D and E are really one theorem expressed in two different ways. We refer to both (or either) as Baire’s theorem. This theorem is admittedly rather technical in nature, and the reader can hardly be expected to appreciate its significance at the present stage of our work.