ir A e

-CARL i ALLENDOERFER
Ss

a
Bece me TES)
mii MU EUnein,.

COU
| ai
a
Se
ae
a
a
See
a
Sn
ee
eas
ana

/
Be
are the easy-impf rae
MCE rey us
ATMS Cte 00
LEU Gc See
| UT a
c help ap HUTT Rest) SUT
— ><“MATHEMATICS
“FOR
lL, PARENTS

CARL B. ALLENDOERFER
PROFESSOR OF MATHEMATICS
UNIVERSITY OF WASHINGTON

THE MACMILLAN COMPANY, NEW YORK COLLIER-MACMILLAN LIMITED, LONDON

© Copyright, Carl B. Allendoerfer, 1965

All rights reserved. No part of this book may be

reproduced or utilized in any form or by any means,
electronic or mechanical, including photocopying,
recording or by any information storage and
retrieval system, without permission in writing
from the Publisher.

Second Printing 1965

Preliminary edition © Carl B. Allendoerfer, 1963

Library of Congress catalog card number: 65-12006

THE MACMILLAN COMPANY, NEW YORK

COLLIER-MACMILLAN CANADA, LTD., TORONTO, ONTARIO

PRINTED IN THE UNITED STATES OF AMERICA

 PREFACE

This book has been written to answer the many questions about
the “new mathematics” which come to me as I move about in public.
When I have lunch in the Faculty Club my colleagues ask me
about this “base 7” business their children are being exposed to;
when I meet someone at a summer resort he is likely to turn out to
be a member of a school board who wants the real truth about
SMSG; and when I go out in the evening my friends blame me for
all the troubles their children are having in school with mathe-
matics that even father never saw before. Although I am a mathema-
tician by profession, I sometimes like to talk about other things.
So this little volume is meant to get me off the hook, for hereafter
when I see a mathematical gleam in the eye of an approaching
parent I can say: “I know what you are about to ask. I have it all
written up in my new book.” And with a reference to my publishers,
I shall turn on my heel and look for some nonmathematical non-
parent who is still willing to talk about the Yankees, taxes, the
weather, or the Russians.
The book is really divided into two parts. The first four chapters
describe the nature of the current revolution in mathematics and
make suggestions for parents who are interested in the best mathe-
matical education of their children. Very little technical mathe-
Vv
vi PREFACE

matics appears in these chapters, and they should be understandable

to even dear old dad who had trouble with algebra in the ninth
grade. If, however, you have more fortitude and actually want to
see the details, the last nine chapters are for you. Here I have written
an easygoing account of some of the most important technical
changes that have occurred recently. When you have read through
these, you should at a minimum recognize the words your children
are using when they talk about their school mathematics. You
may even be able to break the barrier and help them with their
homework, but for most of you this will require more study. If
you wish to go beyond this book, I strongly recommend my ele-
mentary text Principles of Mathematics (jointly written with C. O.
Oakley, 2nd ed., McGraw-Hill, New York, 1963) which seems to
be within the grasp of many of the current generation of high school
students.
Good luck!

Acknowledgments. My thanks are due to KCTS-TV, the educa-

tional television station for Seattle and King County, Washington,
who made it possible for me to deliver a preliminary version of
this material as a series of television broadcasts. Some of the
chapters were also delivered as lectures before the Monday Club of
Seattle, the Mathematical Association of America, and various
groups in Hawaii and Australia where I was a visiting lecturer.
My thanks to these audiences whose questions have led to many
improvements in the presentation.
Carl B. Allendoerfer
Seattle, Washington
TABLE OF CONTENTS

PREFACE

THE NATURE OF MATHEMATICS

HOW THINGS ARE CHANGING 10

WHAT YOUR SCHOOL CAN DO 19
YOUR CHILD AS A MATHEMATICIAN 26
NUMBERS 33
SETS, FUNCTIONS, AND RELATIONS 47
THE LAWS OF ALGEBRA 63
GH
DONA
RWHD
THE DISTRIBUTIVE LAW 76
INEQUALITIES 85
INEQUALITIES IN THE PLANE 98
GEOMETRY 113
TRIGONOMETRY 130
0
Oo
—
ON LOGIC 146
CONCLUSION 164
INDEX 165
vii
 Digitized by the Internet Archive
in 2020 with funding from
Kahle/Austin Foundation

https://archive.org/details/mathematicsforoa0000unse
CHAPTER | THE NATURE OF
= MATHEMATICS

Before you can make sense of the current recommendations for the
reform of mathematics teaching, you must have an understanding
of the nature of mathematics. To many of you mathematics probably
means ordinary arithmetic with its set rules for computation. Others
may remember algebra as a collection of special procedures which,
by some sort of magic, can be used to get the answer in the book.
And geometry is a mysterious subject full of theorems to memorize
and outrageous tricks by means of which we prove the truth of
perfectly obvious relationships in nature. It is no wonder that parents
ask me, ‘‘How can there be anything new in mathematics? Are the
old formulas not still true?” So that you can fit your present recol-
lection of mathematics into the whole picture as a mathematician
sees it, I shall begin with a description of the nature of our subject.
Later we shall discuss the details of its separate branches.
Let me begin by describing the structure of a mathematical
theory in a mature form. In other words, let me show you how a
piece of mathematics looks when it has been completed and polished
and written up (or is it ‘“embalmed”’?) in textbooks. As we shall see,
1
2 MATHEMATICS FOR PARENTS

this picture is quite different from that of a mathematical theory in

the process of development.
I begin our story with “nature”, a word which I wish to use in
the broadest possible sense (see Fig. 1). Under this heading I shall
include all physical and biological aspects of nature as well as human
disciplines such as economics, psychology, anthropology, business,
and warfare. The scholars in these various fields have initially gone
to very great trouble to describe their subjects as best they can in
words, and as their scholarship matures they begin to investigate
Undefined terms
Definitions Theorems
Logical deduction Rules
Axioms

Mathematical
Applications
model

Nature Nature

Fig. 1. Structure of mathematics—polished version.

the quantitative aspects of their domains. Then they join with math-
ematicians to build what is currently called a mathematical model of
their portion of nature. This model is incomplete in many ways, for
first it contains only the quantitative and geometric aspects of nature,
and, moreover, like all models it is only an imperfect copy and not
the real thing. Extant models differ in their degree of excellence;
they are remarkably exact for the physical sciences, but are just
emerging in hesitant form in the social sciences. But whatever the
value of the model, it is the beginning of a mathematical theory.
Let us examine the character of such a model. In the first place,
it is necessary to settle upon a technical vocabulary to describe the
observed phenomena. It is surprising to most nonmathematicians to
§
“
 The Nature of Mathematics 3

be told that the basic words in this vocabulary cannot be defined.

In order to appreciate this point, consider the plight of an American
who tries to read a French text with only the aid of a French lan-
guage dictionary. He can look up each word, but finds that the
definitions are in terms of other French words which have no mean-
ing for him. The case is hopeless unless in some other fashion he
learns the meaning of a basic collection of French words. In math-
ematics we have no way of assigning meanings to this basic collec-
tion of technical words, and must leave them undefined. Thus in
geometry the words point, line, and plane, are taken as undefined.
I agree that we may well have mental pictures of what these words
connote, but this is quite another matter from giving them satis-
factory definitions. The rest of the technical vocabulary is built up
from these words by the familiar process of definition.
Next we wish to use these words to say something about our
observations of nature, and so we form sentences which we believe
to be true. Actually sentences such as “Two points determine a
straight line” cannot be shown to be true by any form of reasoning,
for how can one establish the truth of a sentence containing two
words which are not defined? This sentence is in fact an abstraction
from the observation that there is a unique straight road between a
pair of towns, and thus it is part of our model of nature. Sentences
of this kind which describe basic observed facts of nature are called
axioms and by assumption they are true.
‘Our model, then, consists of a collection of abstractions: unde-
fined words, words defined in terms of these, and statements called
axioms which are assumed to be true. The construction of sucha
model is a creative act of the highest order. In many ways it com-
pares with the landscape of the artist or the novel of the writer, each
of whom is expressing in his own way aspects of nature which appear
/-important to him at the time. Let us examine a few models of this
kind.
' The first illustration of a mathematical model that a child meets
is that of the concept of number. In his early years he learns to count
and is able to understand phrases such as: ‘“‘three blocks’, “three
trees’, “three boys’, and the like. But what does “three” mean all
* by itself? Primitive tribes have difficulty answering this question,
and some of them use different words for “three”? when they are
counting different kinds of objects. As a matter of fact there is a
mes 4
4 MATHEMATICS FOR PARENTS

trace of this in our own language, for we use different words to

describe groups of various kinds of animals, such as: flock of sheep,
herd of cattle, pack of wolves, or covey of quail. There is no need for
all these separate names for the idea involved, and you can see the
confusion that would result if we extended this practice to our ordi-
nary counting operations.
But what is “three”? It is an abstraction derived from our ex-
perience in actual counting. When we abstract the other counting
numbers and formulate the rules for computing with these, we have
constructed the mathematical model which we call arithmetic. The
troubles which young children have in arithmetic have their origin
in this process of abstraction, for this is the first time the youngsters
have met anything abstract. Now that we understand the nature of
the difficulties, we can consider what needs to be done to help the
children over this hurdle.
» Another illustration of a mathematical model is the usual for-
mulation of plane geometry. As you will remember, geometry begins ~
logically with a set of axioms, or postulates, and a set of undefined
terms. These are obtained by abstraction from the physical process
of measuring parcels of land. A farmer’s field has corners, boundaries,
and the land area itself. The corners may be marked with posts and
the boundaries with fences which are only approximately straight.
In our model these posts become points, the fences become lines, and
the field itself becomes a plane. Point, line, and plane are our ab-
stractions. Now we look at the field and observe, for instance, that
there is a single fence between every pair of corner posts. In our
abstraction this becomes the axiom: there is a unique line passing
through any pair of distinct points. By this process we build up the
model which we use as our foundation of plane geometry.
A further example of the construction of a model is Newton’s
invention of the calculus. His concern was with an appropriate
theory to explain the observed motions of the planets in our solar
system. Quantitative measurements and elementary mathematical
descriptions of the planetary orbits were at hand, but there was no
systematic theory which accounted for them. Newton’s first task
was to develop an adequate vocabulary which included the defini-
tions of words such as velocity and acceleration. These definitions are
now the basic ideas of the differential calculus. He did not lay a solid
The Nature of Mathematics 5

foundation of undefined terms and axioms, and these had to be

supplied at a later date. Like all model builders, he painted with a
broad brush and left the details to be filled in by lesser men. Never-
theless, he did state the most essential axioms, which are currently
known as Newton’s laws of motion. On this foundation he built his
theory of gravity and planetary motion. One of Einstein’s great con-
tributions is the construction of a different model for this same por-
tion of nature.
A most exciting development at the present time is the con-
struction of similar models in the social and biological sciences. Al-
though these models are far from ideal, they are bold attempts to
put these subjects on a firm mathematical foundation and are the
forerunners of great things to come.
Once the model is constructed, the mathematician proceeds to
prove theorems. These are statements about his subject which can
be deduced logically from his assumed axioms. To say that a math-
ematical statement is true is to mean that this statement is a logical
consequence of a set of axioms, and nothing more. The notion of
absolute truth is entirely foreign to mathematics in spite of the
worship of mathematics by laymen who are seeking something that
is unchangeable and eternal. If mathematicians were to apply the
method of reasoning of their own subject to their personal philosophy
of ethics and values, they would be (and frequently are) pure
relativists.
The chain is now closed by the applied mathematician who
takes the general theorems so deduced, puts numbers in them, and
attempts to discover new relationships in nature. Of course, he can
prove nothing about nature in this way. Nevertheless, he can arrive
at statements about nature which can probably be verified by obser-
vation of nature if his original model is reasonably good. Of course,
his objective is to find new facts about nature which had not been
previously observed, or better, to predict what would be found if a
certain measurement or observation were to be made.
I fear that I have misled you into believing that a mathematical
theory is built by following these three steps in order; nothing could
be farther from the truth. Mathematicians often perform the mag-
ical feat of building the fifth storey without first establishing a foun-
dation or even thinking about the first four stories. We start at both
6 MATHEMATICS FOR PARENTS

ends and the middle and only after decades of effort bring forth the
beautiful, logical, and polished gems which we dangle before our
students.
More often we follow a different path. Beginning with nature
as before, we seek to find as many relationships within it as we can.
If we can systematize these we do so, but a lack of organization of
our material does not keep us from pushing forward. On the basis of
what we have observed, we guess theorems and use these to derive
other theorems. Immediately we rush to apply these back again to
nature and proceed headlong if our predictions are successful. Ax-
ioms, logic, and rigor are thrown to the winds, and we become in-
toxicated with our success and open to dreadful errors.
This process is called zntuztron, and its nature is in no way under-
stood. Several great mathematicians have written accounts of their
experiences in this phase of mathematical discovery, but the psy-
chologists are baffled by their testimony. The successful unraveling
of this process should be a major contribution to the understanding
of the human mind. But explained or not, it is by this means that
the great majority of mathematical theorems are first discovered.
One of my teachers in graduate school was good for about one hun-
dred such discoveries a week, with an average of only about one
which stood up under careful analysis. Yet those that were correct
made him one of the great mathematicians of the past generation.
The procedure is then as in Fig. 2. By means of intuition we
guess theorems, test them against applications, organize those that
survive, find axioms on the basis of which they can be proved and
thus build our mathematical model. A little-known trade secret is
that in this process theorems come first and axioms second. A good
example of this is the re-examination of Euclidean geometry in the
past century. It was discovered that many-of Euclid’s theorems
could not be proved from his axioms. Yet we all believed these
theorems to be true, and a number of different systems of axioms
were developed which had these theorems as their consequences.
This process of justifying a belief by trying to find premises from
which it can be deduced is shockingly similar to much reasoning in
our daily lives, and I am embarrassed to have to let you know that
mathematicians are experts at this art.
After the model has been constructed in this way we then prove
theorems by deduction and finally seek for applications to nature.
The Nature of Mathematics 7

There are, then, four key words in the process of constructing a math-
ematical theory:
Intuition
Organization
Deduction
Application
We shall see that this same pattern applies to the learning of math-
ematics, and that any good course in our subject must include all
four of these features.

Theorems
Axioms Logical deduction Rules

Organization

Intuition Applications

Nature Nature
Fig. 2. Structure of mathematics—as it is discovered.

This description of mathematics would have seemed reasonably

accurate until something more than one hundred years ago, but it
omits the most significant development in our subject during the
past century. The breakthrough which stimulated this new point of
view was the construction of non-Euclidean geometry. One of
Euclid’s axioms stated that there exists a unique line through a
given point which is parallel to a given line which does not pass
through this point (Fig. 3). This axiom was never fully accepted by
mathematicians on the ground that it did not have the ‘“‘self-evi-
dent” quality of Euclid’s other axioms, and many attempts were
made to prove it on the basis of the other axioms. Although no such
proofs were obtained, everyone believed that this axiom was true,
8 MATHEMATICS FOR PARENTS

and in fact Euclidean geometry was given the aura of absolute truth
commonly associated with the Bible.
In the early part of the nineteenth century attempts were made
to prove this axiom by the method of contradiction. In conformity
with this method mathematicians assumed that the axiom was false
and hoped to deduce conclusions which contradicted at least one of
the remaining axioms. In particular the two alternative axioms were
investigated: (1) there is no such parallel, and (2) there are at least
two such parallels. Although these mathematicians worked diligently
and imaginatively, they were frustrated in that they found no such
contradictions. It is reported that Gauss was the first to have arrived
at this awkward impasse and that he considered the possibility that

Fig. 3. Euclid’s parallel axiom.

alternatives to the parallel axiom were logically possible. So great

was the authority of Euclid that he felt it unwise to publish his
findings. Later others such as Lobachevski and Bolyai came to sim-
ilar conclusions, and gradually the secret was shared with the com-
munity of mathematicians. The astounding conclusion was that
there are three equally acceptable geometries which appear to be
satisfactory models of nature. Even today there are no compelling
reasons for choosing among them.
It required another fifty years for the major impact of this dis-
covery to transform the character of mathematics, but now we are
so deeply imbued with the new spirit that young mathematicians
cannot understand why anyone ever thought differently. The new
philosophy is that abstract systems of undefined words, axioms, etc.
(which we previously called mathematical models), need have noth-
ing whatever to do with nature. They are no longer models of any-
thing, but are merely structures built by mathematicians because
they are thought to be worth investigating. Since these abstract
 9
The Nature of Mathematics

mathematical systems did not spring from nature, there is no obliga-

tion for their advocates to apply the related theorems back to nature,
and consequently we have a mathematics which exists purely of and
for itself. A large part of contemporary mathematics is of this kind.
You may wonder how one goes about thinking of such an ab-
stract system, and the process is as difficult to describe as are most
creative acts. At the initial stages the abstract systems may be called
mutations of earlier models of nature. For example, geometry has
been extended from three dimensions to an arbitrarily large finite
number of dimensions, and finally to an infinite number of dimen-
sions. In algebra we are accustomed to assume that 3 X 4 = 4 X 3
and more generally that a X b = b X a. Inone of the newer algebras
the rule is that a X b = —b X a. Abstract spaces are considered in
which the notion of distance between two points has been replaced
by a much more general concept of nearness. And so on. At each
step the new abstraction becomes so familiar to the mathematician
that he regards it as concrete and makes it serve as the basis for
further abstractions. The length of a mathematical generation is
about ten years, and each generation regards with dismay the atti-
tude of the following generation, which treats the marvelous ab-
stractions of their predecessors as trivial or narrow.
Another source of abstraction is the unification of two older
theories into a single more general one. The two theories may well
have sprung from two different mathematical models of nature with-
out any regard for each other. Their combination into a single ab-
stract theory often illuminates each of the parent theories, and
always produces a great economy of thought. The growth of math-
ematical knowledge is so rapid that without such amalgamations no
one could find his way through the morass.
Although the appearance of these modern theories suggests that
they have no connection at all with nature, I cannot think of a single
one which does not owe its existence to aremote mathematical
model of nature out of which it has grown by the processes of muta-
tion and generalization. Perhaps the invention of a theory with no
remote roots in nature is beyond the powers of the mind, but more
probably such a construction will occur before long as the next major
breakthrough in the development of mathematics.
 os
CHAPTER 2 HOW THINGS ARE
a espe SESS (EAINIC

Let us return to Fig. 1 of the first chapter and look at it in the light
of the traditional curriculum in mathematics. Most of you will find
that your experience with mathematics was confined to the right
side of the figure. You were shown a set of rules and given practice
in applying these to a variety of more or less artificial situations.
Only rarely (especially in geometry) were you given any idea of the
reasons behind the rules or of how you might have invented them
yourself. An extreme version of a traditional mathematics textbook
might well look like that of Fig. 4.
On the left hand page we have a rule, usually printed in color,
and often imperfectly stated. Under this are several worked examples
illustrating the application of the rule. On the opposite right hand
page there are fifty or one hundred problems to which the rule is to
be applied. The student memorizes the rule, works a selection of the
problems, and then turns the page. There he finds exactly the same
thing, although with a different rule. Consecutive rules may have
some relationship, but this is infrequently explained in detail. After
a series of some five or ten episodes of this kind, the student is tested
10
How Things Are Changing 11

on his proficiency, and then proceeds to more rules. All too often the
earlier procedures are forgotten, and mathematics appears to the
youngster as a collection of rather stupid, isolated tricks which he is
supposed to learn by rote.

page 32 page 33

| Problems to which
Rule the rule is to be
(in color) applied

51.
52s
Dos
ILLUSTRATIVE
Examples

Fig. 4. Traditional textbook.

Teaching with Understanding

The first major change of the current mathematical revolution

is to put emphasis on teaching with understanding in place of rote
learning. Of course good teachers have done this all along, but too
frequently the teacher himself did not have the necessary back-
ground, and all that he could do was to drive his pupils through the
book. Teaching with understanding involves presenting the subject
as a whole, preferably following the sequence in Fig. 2, so that
all four steps are included: intuition, organization, deduction,
application.
First we must build intuition. This is accomplished by showing
the student a large number of elementary illustrations of the idea
which is to be presented later in a formal fashion. From these he gains
12 MATHEMATICS FOR PARENTS

a feeling for correct procedure, even though he cannot explain it in

any precise way. Then the teacher capitalizes on this intuition, and
helps the student organize his knowledge. Eventually axioms are
introduced and theorems are proved, thus filling in the portions of
Fig. 2 which were previously left out. It is most important that the
applications be introduced, and that mathematical skills be devel-
oped and maintained by periodic review. There is good evidence that
students taught in this fashion learn mathematics more easily and
remember it longer than if they were taught in the rote memory
tradition.
To implement this method of teaching, schools are making
changes in their curricula. For example, instruction in arithmetic
takes into account the laws of algebra which are to come later. If you
look ahead to Chapters 7 and 8 you will see that we base algebra on
the commutative, associative, and distributive laws. Since these ap-
ply equally well to arithmetic, we introduce them intuitively in the
elementary school, where they become a natural part of the child’s -
knowledge. Then he has little difficulty with algebra when he meets
it in high school. In the same way schools are teaching the facts of
plane and solid geometry in the elementary school and in junior high
school. The youngsters learn these by constructing figures with ruler
and compass, and thus develop their mechanical coordination simul-
taneously with their mathematical knowledge. Then they are pre-
pared to organize this geometrical knowledge when demonstrative
geometry is presented to them as an axiomatic system in the tenth
grade. The intuitive aspects of calculus are frequently taught in the
twelfth grade to build a foundation for the serious calculus the stu-
dents will meet in college.

Old and New Topics

Another of the changes in mathematics-teaching is the elimina-

tion of certain older topics and their replacement with others. There
is nothing very dramatic about this, for changes of this kind have
been carried on continuously for centuries. These changes, of course,
should be gradual and not subject to whims or fads. Let me give a
number of illustrations.
In my library I have an arithmetic book, published in 1808,
which carries the endorsements of the presidents of Dartmouth, Har-
How Things Are Changing 13

vard, and Yale universities. In this text the students studied pounds,
shillings, and pence—for we had only recently adopted the decimal
system in our currency. They were taught, for instance, to find one-
third of 4 pounds, 12 shillings, and 5 pence. Later in the course they
were taught not only how to take square root but also cube root,
fourth root, and fifth root. These topics have fortunately disappeared
from modern arithmetics, but undoubtedly there were head shakers
among the old-timers when these changes were made.
The most important trends of this kind at present have to do
with the techniques of computation and with solid geometry. Until
the 1920’s there were only two available methods for accurate com-
putation: longhand and logarithms. Early forms of the desk calcu-
lator were available, but these were not in widespread use; and some
forward-looking engineers were using slide rules. But most people
had to rely on pencils or logarithm tables for their work. Mathemat-
ical instruction was geared for this, and students were taught appro-
priate methods for handling their problems by these means.
Since then we have seen the widespread use of the desk calcula-
tor in business and technology, and the universal adoption of the
slide rule for approximate calculations by scientists and engineers.
Moreover, we are now in the midst of a major new development
resulting from the invention of the high-speed electronic computing
machine. Methods of handling problems which were designed for
efficient hand or logarithmic calculation are now replaced with those
which are most appropriate for the use of modern computing equip-
ment. For instance, the old, long method of taking square roots and
the logarithmic solution of triangles (in trigonometry) are now obso-
lete. There is no reason to teach such things to our children just
because they were essential tools for our own parents.
The situation concerning solid geometry is somewhat different.
This course used to require a semester, usually in the twelfth grade.
The purpose was twofold: (1) to reteach the axiomatic method previ-
ously taught in the tenth grade, and (2) to train the students in
three-dimensional perception. The first of these objectives can now
be met in other ways, particularly through the use of axioms in
algebra. Perception of relationships in space remains an important
objective in the total educational picture, but it is important that it
be taught at a much earlier age. For these reasons the formal course
in solid geometry has largely disappeared. Some schools have missed
14 MATHEMATICS FOR PARENTS

the point about space perception and have failed to reintroduce it in

other places—this is a most serious mistake. Three-dimensional ideas
can be taught to the youngest of children in the early grades, intuitive
space geometry can be presented in the junior high, and the ideas of
solid geometry (without all their proofs) can be taught hand in hand
with the parallel ideas of plane geometry in the tenth grade.
Something like a semester can be saved (for nonaccelerated
students) by these omissions, and a number of recommendations are
available for filling this time. I am specifically thinking of the last
semester of the twelfth grade. By all odds the most attractive subject
here, for me, is that of analytic geometry. This beautiful and prac-
tical piece of mathematics is absolutely essential for the applications
of mathematics to science and for the study of the calculus. It used
to be a standard college course, but the colleges, in their haste to
teach calculus to all freshmen (ready or not), have drastically short-
ened the time spent on analytic geometry. This unfortunate develop-
ment has made mathematical cripples of many of our young engi-
neers and scientists. An excellent scheme for remedying this omission
and for giving an exciting last course in high school is to make this
course one in analytic geometry.
Another popular suggestion is that this semester be spent on
probability with some application to statistics. One cannot escape
the applications of probability in any walk of life, for they are with
us at the race track, in public opinion polls, in industrial quality con-
trol, in all our economic data collection, and in many modern forms
of decision making. The suggestion, then, is that students should
start the study of probability as early in their lives as possible. A
very real problem with such a course, however, is the availability of
properly trained teachers. Probability is a special kind of mathe-
matics to which many good mathematicians have never been ex-
posed. Moreover, it can be best taught only by someone who has got
his hands dirty by using it in the field. Unless your school has a
teacher with these qualifications, it would be wise to discard this as a
possible twelfth grade course.
Principals and superintendents are especially fond of introduc-
ing calculus into the twelfth grade, for they take pride in this as a
sign of their progressive leadership. For unaccelerated students the
introduction of much calculus is a bad mistake. I do favor a month
or two of intuitive calculus for reasons mentioned above, but any
How Things Are Changing 15

more is likely to replace other essential parts of algebra, trigonom-

etry, or analytic geometry and hence leave the students unprepared
for their college work. My message to you as parents is: “Look most
critically at your school if it teaches more than eight weeks of cal-
culus to unaccelerated students.’

New Language

To many parents the most noticeable sign of change in math-

ematical education is the use of unfamiliar language. Your children
will tell you about sets and open sentences. They will read —4 as
‘negative four” instead of ‘minus four’’, and they will refer to the
commutative, associative, and distributive laws which I mentioned
before. Do not be dismayed by this language. I will explain it all to
you in later chapters. It is really quite simple, and is very helpful.
Although there are extreme mathematical revolutionaries who
would like to change the whole vocabulary, most teachers have
made only those changes which are essential for precise statements
of mathematical ideas. The old language had so many sources of
ambiguity that students had to learn the subject by osmosis rather
than in a logical way.

Acceleration

- Finally I should say something about acceleration. Since all

children do not learn with equal speed, the schools have recently
been separating the rapid learners from the others and ‘‘accelerating”’
their progress. It is certainly true that some eighth graders are ready
for algebra, and that it is a crime to waste a year of their lives giving
them more arithmetic when they know it already. Various forms and
degrees of acceleration are possible, and each school must adapt its
system to meet local conditions. Parents and teachers should remem-
ber, however, that the purpose of acceleration is to teach more mathe-
matics in the same number of years, rather than to enable the stu-
dents to drop their mathematics before they have finished high
school. As I shall develop in greater detail in Chapter 4, mathe-
matical education should be continuous from the first grade into
college. Acceleration which leads to fallow years with no course in
mathematics does more harm than good.
16 MATHEMATICS FOR PARENTS

If your school is introducing an accelerated program, be sure

that plans have been made for the continuous study of mathematics
through the twelfth grade for accelerated students. The normal
twelfth grade course for accelerated students is a serious course in
calculus at the college level. But students should not enter such a
course unless: (1) they have completed the full curriculum of college
preparatory mathematics including analytic geometry, and (2) the
school has a teacher who is as well equipped to handle the course
as a college professor. Students completing such a course should be
well prepared to attack the Advanced Placement Examination of
the College Entrance Examination Board. College credit is usually
given to those who obtain satisfactory scores on this examination.

The People Involved

I am often asked ‘‘Just who is behind all these changes?” There

seems to be a fear that they are another manifestation of progressive
education or some kind of nostrum sponsored by departments of
education. Nothing could be farther from the truth. These changes
are the outcome of the joint thinking of a very large number of the
nation’s best university mathematicians and secondary school
teachers. Recently elementary teachers have been taking part in
planning the changes in the grade schools. Let me say a few words
about how this all came about.
In the early years of this century there was relatively little
advanced mathematical competence in the United States. A few
universities had professors of outstanding quality, but an ambitious
young man was well advised to go to Europe for his advanced train-
ing if he planned to be a research mathematician. The teaching of
mathematics at the school and college level was apparently satis-
factory for the standards of the times, and some textbooks were
written that are better than many of their more recent counterparts.
About the time of World War I a conscious decision was made to
allow the small number of research mathematicians to put their
energy into the discovery of new theorems and the training of
Ph.D.’s, and to leave the problems of school and undergraduate
mathematics to others. This policy was highly successful on the
research side, for in less than forty years American mathematics
vaulted to a leading position in the world. At the same time school
How Things Are Changing 17

mathematics suffered a serious decline for lack of contact with the

exciting changes in the world of mathematical discovery.
The gap between school and research mathematics had become
a scandal in the 1930’s, and a reform movement was under way
when World War II snuffed it out. Because of the strains of the war
and the GI bulge, nothing further was accomplished until about
1950. At that time several individual mathematicians began trying
out new materials on their college freshmen. Some of these efforts
appeared as books, and schoolmen began to take notice. The College
Entrance Examination Board appointed its Commission on Math-
ematics in 1954, and work began in earnest. This commission in-
cluded both school and college teachers of mathematics, and its
formation is a historic event in that these groups had not met
together under proper auspices for at least 40 years. The commis-
sion’s final report was not issued until 1959, but preliminary drafts
had been widely circulated in previous years. The plan for reform
was ready; all that was needed was the money to support it. To our
very great good fortune the Russians launched Sputnik I in 1957,
and the public demanded action.
In response to this demand the School Mathematics Study
Group (SMSG) was established on the recommendation of the pres-
idents of the American Mathematical Society and the Mathematical
Association of America. Generous financial support was provided
by the National Science Foundation, and the movement was
launched in a big way. SMSG established writing groups composed
of university and school mathematicians and has published a
“five-foot shelf”? of experimental textbooks for elementary and
secondary school use.
Private foundations have also financed university writing
teams of which I shall only mention that at the University of
Illinois. These teams have also published experimental textbooks,
and now schools are able to choose among several alternative modern
presentations of mathematics. There is no “official line”’ and anyone
with ideas and the energy to write them up is welcome to the fray.

How Have the Changes Been Received?

The various experimental texts have been used for several
years in hundreds of schools throughout the country. Given properly
18 MATHEMATICS FOR PARENTS

trained teachers and able students, the results have been excellent.
Freshmen entering universities are far better prepared than their
counterparts of a few years ago. Teachers and students are enthu-
siastic, and there is much acclaim.
Some weaknesses, however, have appeared. The new materials
are too hard for the less than average student, and so new versions
are being written for these pupils. Some schools swung too far on
the pendulum and forgot to teach skills as well as ideas. They have
learned and are moving back toward a well-balanced position.
There has, however, been some outspoken opposition. Part of
this comes from congenital conservatives who oppose change of
any kind; part comes from those who are too lazy to change; part
comes from the uniformed (who have not read this book!); and
there is a small group of mathematicians who think that extremists
have taken over the movement and are leading it into dangerous
territory. Any revolution has its extreme fringe, and I agree that
our extremists have gone too far. The opposition, however, is badly
mistaken in its belief that these extremists represent the views of
the great body of mathematics teachers. Judged by the published
work of its responsible leaders, the revolution is on firm ground,
and its success will bring tremendous benefits to the American
people.
CHAPTER 3 WHAT YOUR
2 SCHOOL CAN DO

Those of us who are involved in this movement are frequently called

upon by somewhat bewildered school officials. Someone in the school
system has got the word about the ‘new mathematics’’, perhaps a
teacher who attended a summer institute, perhaps a curriculum
supervisor who attended a national meeting, or perhaps a parent
who is new to the district and knew what was happening in his
former home town. These officials are naturally nervous, for they
wish to do the right thing. But they are confused by the plethora
of programs and initials, by the aggressive claims of salesmen of
texts and equipment, and by their own inexperience with curriculum
reform of this kind. So they come to us and ask for help. This chap-
ter is a summary of the advice that I have given in many visits to
school districts. I hope that it will be helpful to parents who wish
to get action out of their local school people, or who wish to evaluate
those actions which are being taken.

Become Familiar with the New Programs

The first step is to learn what all the shouting is about. Reading
this book will help you, but you should also read the excellent
19
20 MATHEMATICS FOR PARENTS

pamphlets The Revolution in School Mathematics (1961) and An

Analysis of the New Mathematics Programs (1963), which are pub-
lished by the National Council of Teachers of Mathematics, 1201
Sixteenth Street N.W., Washington, D.C. 20036. Each pamphlet
costs $0.50, and quantity discounts are available. These give the
names, addresses, and points of view of the leading projects and
people in the field. You can also obtain information from the mathe-
matics department of your local university, and such departments
are usually delighted to provide speakers at PTA’s or the like to
whom they can explain what is happening. Your state department
of education can also be very helpful. A school district which does
not have this basic information is very much behind the times.
It is not enough, however, for a few administrators, teachers,
or parents to be familiar with the new alphabet soup; the mathe-
matics teachers must also know something of the details. Even
though they are completely new to the movement, teachers can get
a good start on their own by reading the Report of the Commission
on Mathematics of the College Entrance Examination Board together
with its detailed appendix. Copies may be obtained from the Edu-
cational Testing Service, Princeton, New Jersey ($1.00). Teachers
can also benefit greatly by working through one of the SMSG texts.
The one titled Mathematics for the Junior High School is an excellent
place to begin. These texts are available through the Yale Univer-
sity Press, 92A Yale Station, New Haven, Conn. (two parts, each
$3.00). A profitable method of studying such a text is to organize
an after-school faculty seminar in which teachers take turns present-
ing the new material to each other. When they have trouble, an
occasional visit from someone in a local college will usually straighten
them out. :

Get the Support of Teachers, Administrators,

School Board, and Parents
It is futile to try to begin one of the new programs without the
approval and knowledge of all the interested people in the commu-
nity. It is particularly dangerous for one school to change its cur-
riculum without full consultation with others that may be affected.
I have seen very serious difficulties, for example, when a junior high
school started teaching algebra in the eighth grade without consider-
What Your School Can Do 21

ing what would happen to these children when they entered a high
school which disapproved of acceleration. Decisions of this kind must
be system-wide and not piecemeal by schools.
The parents also need to be brought into the picture in advance.
For there can be a bad reaction from home when Johnny shows off
his algebra book, which is full of words and problems that father
never learned about in school. Some school people take the view
that telling parents anything is asking for trouble, but I strongly
disagree. Parents are entitled to know what is going on in their
schools, and if they are not informed they are quite unlikely to
approve the special taxes needed to support the schools.

Make Certain the Teachers Are Retrained

Once a decision has been made to install a new program, the

progress of events should not be too hasty. It is madness to hand
even the best teacher one of the new books on September 1 and
tell her to begin using it at once. The changes in approach, emphasis,
and terminology are too great to be overcome quickly without out-
side help. Fortunately there are many avenues by which this help
can be obtained. The school district should require that a teacher
about to begin a new program attend a summer institute (such as
those sponsored by the National Science Foundation) or a summer
school in a university which offers appropriate courses in the new
programs. School boards have even been known to pay the way of
teachers so that they could take such courses.
Those teachers who cannot get to such summer schools can
often find helpful courses or institutes offered by their local uni-
versity in the late afternoons, evenings, or on Saturdays. These
are also useful for those who have been to a summer institute, and
who wish to enlarge on the training received there. Some school
districts arrange for these extension courses to be held in regular
school buildings, and frequently sponsor courses for inexperienced
teachers which are taught by their more knowledgeable colleagues.

Do Not Start a Crash Program

The most dangerous man in the whole business is the eager
beaver, hard-driving principal or superintendent who suddenly
22 MATHEMATICS FOR PARENTS

makes mathematics his special project for the year, and says, for
example: ‘“‘Put SMSG into every classroom in the schools, beginning
at 9 o’clock tomorrow.” Perhaps no such person exists, but I have
met some that come mighty close. A good school curriculum in
mathematics must be developed gradually; it should be watched
constantly and improved and extended every year. A crash program
which intends to do everything at once (and then get on with some-
thing “important” such as the new stadium) is sure to fail.

The Way to Start

Typical steps are these:

(1) Start with a few teachers and a few classes at strategic
points in the curriculum such as the seventh and ninth grades.
These teachers must have special training, and it is best if the classes
are composed of the more able students. .
(2) For the next year, train more teachers and expand the
programs to the next grades (eighth and tenth, for example, from
the seventh and ninth) and start a new group of students at the
earlier levels. Continue expansion in this way.
(3) Gradually introduce the new materials into classes for the
less able. At first only a few units may be inserted into a standard
course, but these can be expanded as the teachers learn how to
adapt them to the abilities of these children.
(4) As a final goal, have something new going on in every
classroom in your system. This can easily require a development of
five to ten years, but it must be done.

Distinguish Between New Mathematical: Ideas

and Commercial Mechanical Aids
Some schools have missed the point of the mathematical revo-
lution and have supposed that they can meet their obligations by
buying various mechanical teaching aids. It is no wonder that this
has happened, for NDEA (National Defense Education Act) money
is plentiful, persuasive salesmen are everywhere, and it is so much
easier to spend money than to change people. Beyond any doubt
teaching can be improved through the use of appropriate audiovisual
What Your School Can Do 23

devices, but these devices are no substitute for a good curriculum.

Let me discuss some of the more prominent of these devices.
1. BLocks. There is nothing new about using blocks in the teach-
ing of young children. These can be very effective in helping the
youngsters to develop intuition about numbers and in enlarging
their space perception. We have seen that the first step in learning
is at the level of intuition, and that even simple abstractions such
as the number “‘three’’ often come hard. It is here that blocks are
most valuable.
Among the varieties of blocks available as aids in the teaching
of arithmetic are some called Cuisenaire Rods. These are colored
sticks 1 centimeter square and with lengths varying from 1 centi-
meter to 10 centimeters. By using these as concrete examples of
nature, young children easily develop intuition about numbers, learn
how to add, subtract, multiply, and divide with whole numbers, and
later to handle the basic ideas of fractions. When these are used by
a skillful teacher, the children make rapid progress in learning arith-
metic with understanding. It is important, however, for the children
to realize that the rods are merely a crutch to be used as they master
arithmetic, and that the rods should be discarded as soon as they
are no longer necessary.
Another valuable kind of blocks are those invented by Dr.
Z. P. Dienes, called the Multibase Arithmetic Blocks. By means
of these children learn the meaning of the positional notation in
our number system (see Chapter 5) and rapidly acquire skills in
computing in number systems to various bases.
Many teachers supplement or replace these commercial blocks
with easily available counters such as milk bottle caps, pebbles,
leaves, and the like. It is probably desirable to have a considerable
variety of such materials on hand, for the main problem is to get
the children to abstract from concrete objects to numbers them-
selves. This abstraction is easier if the children have had numerous:
different kinds of practical experiences.
It should be realized that all these aids contribute solely to the
development of intuition. The additional steps of organization,
deduction, and application are still necessary. Thus blocks, rods,
pebbles, etc., no matter how useful, do not solve anything like all
the problems of teaching arithmetic. Some schools seem to think
24 MATHEMATICS FOR PARENTS

that, by the single act of buying rods, they have done all that is
needed to reform their arithmetic teaching. This is indeed an excel-
lent first step, but it must be followed up by a change in the method
of instruction as well.
2. TEACHING MACHINES. The past few years have seen the com-
mercial exploitatio n of Professor B. F. Skinner’s research on pro-
gramed learning and teaching machines. The basic idea is very
simple: students learn easily if they progress by very short steps,
and if they are told immediatel y after each step whether their action
was right or wrong. The questions which comprise these steps can
be presented to the student as “frames” in a specially designed
textbook, as films which he views in a teaching machine, or in
many other ways.
The advantages claimed for this method are that students can
proceed at their own pace independent of the rates of learning of
their classmates, that learning is more rapid than in an ordinary
class, that bad habits are stopped before they become ingrained,
and that teachers are relieved from routine paper grading and hence
are available to do more actual teaching. No one claims that the
method is cheaper than the usual one, or that teachers will be any
less important.
I am impressed with the probable usefulness of programed
learning in many teaching situations, but I deplore the excesses
of the commercial promoters of teaching machines. No matter how
good the basic idea of programed learning may be, it cannot be
proved in practice until some first-rate programs are written and
tested experimentally. The construction of a program requires the
collaboration of an experienced teacher and writer with a psycholo-
gist who is an expert in learning theory. Too often programs have
been thrown together by less competent persons so that they could
be available to promote the sale of teaching machines. The results
have been unfortunate in many cases, and the reaction of mathe-
matics teachers is quite mixed. Fortunately good programs follow-
ing modern curricula in mathematics are now becoming available,
and soon a better judgment can be made.
Moral: Do not rush into the purchase of teaching machines
until you are sure that well-tested, modern programs are available
for use in them. Consider programed textbooks quite carefully, but
What Your School Can Do 25

judge them by the same standards which you use in evaluating

ordinary textbooks.
3. OVERHEAD PROJECTORS. These are devices for projecting mate-
rial which is in front of the teacher onto a screen above his head.
These projectors can be very helpful in teaching visual subjects like
geometry, for transparencies of geometric figures can be purchased
(or prepared by the teacher) with much greater accuracy and
appeal than drawings hastily drawn on the blackboard. Their use
is especially recommended for the projection of drawings of solid
figures when the students are being trained in space perception.
Moral: Do not buy overhead projectors unless you have enough
money to buy transparencies and other supplies needed to make
them effective.
CHAPTER 4 YOUR CHILD AS A
MATHEMATICIAN

Apart from a knowledge of the English language, competence in

mathematics is the most useful skill which your child can acquire
in school. Originally one of the three R’s, its importance was played
down in the heyday of progressive education on the ground that
such a rigorous subject might interfere with the development of
the child’s personality. Events of the last ten years in the United
States have restored mathematics to its old position as a subject
which must be a major part of the education of every citizen, boy
or girl.
The most striking feature of our times is that we are in the
middle of the Scientific Revolution, which is rapidly transforming
our whole society much as did the Industrial Revolution of the last
century. No one can understand what is happening around him
without a knowledge of science, and in order to learn about science
one must first master mathematics. Indeed the situation is even
more serious than that. Automation is sweeping our factories and
offices, and jobs for the unskilled are rapidly disappearing. If your
child does not master mathematics in the schools of today, he may
26
Your Child as a Mathematician 27

well be among the unemployed of tomorrow. It is, therefore, the

obligation of the schools to provide the best possible education in
mathematics, and the duty of you as parents to see that your child
is enrolled in mathematics throughout his entire school career.
Let us examine the mathematical preparation which is needed
for a number of popular careers. It is perhaps commonplace that
mathematics is required for a career as a physicist, chemist, or
engineer, but few people outside the colleges realize that the mathe-
matical requirements of these professions have doubled within the
past decade. School counselors and parents will still do well to urge
mathematics upon every prospective physical scientist or engineer,
but they are not properly informed if they recommend less mathe-
matics to prospective economists, psychologists, biologists, physi-
cians, and business executives. Even the poet and the musician are
not untouched by this cultural revolution.
World War II marks the turning point in the mathematical
needs of workers in the behavioral sciences and in business. Between
1940 and 1943 the Nazi threat was so great that the military leaders
of Great Britian and the United States turned to scientists (includ-
ing mathematicians) to help plan the strategy of the war. This
was the beginning of the discipline called Operations Research which
in one form or another has become an essential feature of our busi-
ness and industrial life. The basic method is that which I described
in Chapter I: the scientists built a mathematical model of warfare
and proceded to draw conclusions from this as to the proper conduct
of the war. Since this was total war, we had to learn as much as
possible about the economies of enemy countries. The purpose then
was to destroy these economies, but we soon discovered how little
economists knew about the interrelationship of the various portions
of an industrial complex. We wanted to know what industries to
bomb for maximum effect, but nobody could provide well-reasoned
answers. After the war economists started to build mathematical
models of a nation’s economy in order to find such interrelationships
for peaceful use and before they realized it they were head over
heels into more’ mathematics than they had ever known to exist.
The war also spawned the mathematical subjects known as
Game Theory and Linear Programing. These are serious efforts to
help executives make decisions in a rational fashion rather than
28 MATHEMATICS FOR PARENTS

on the basis of customs, intuition, or the advice of “experts” from

out of town. There has grown up a large and influential group who
are practicing these arts under the general heading of Management
Science.
A simultaneous development has been the perfection of the
high-speed digital electronic computing machine. From their early
beginnings during the war these machines have been refined and
improved to an amazing extent. They are now essential equipment
in scientific research organizations, industrial plants, and large
businesses. Technical and management decisions are now made on
the basis of data which are analyzed on these machines, and much
guesswork has disappeared from our industrial and business opera-
tions. The business executive who cannot cope with this new type
of precise decision making is likely to find himself bypassed for
promotion or even out of a job.
I do not mean to say that every business man must know how
to operate these machines, for with luck he can find technicians to
do this. He must, however, understand what machines can do and
what they cannot do. He must have a feeling for the quality of
information which is fed into the machine, and be able to judge
the reliability of its output. There is a popular worship of numbers
which is far from justified; for there is much truth in the old saying:
‘Figures don’t lie, but liars can figure.” So that our executive will
be able to avoid being misled by accidental or intentional errors
of this kind, he needs to have facility with numbers and logic. He
can obtain this by studying mathematics.
The future biologist or physician is faced with a revolution
of major proportions in these subjects. The important discoveries
in biology are now closely intertwined with modern chemistry and
physics, and these require a deep knowledge of mathematics. The
future physician will wish to build upon these discoveries in biology,
and will also wish to understand the uses of high-speed computers
in medical research and diagnosis. He cannot afford to be a mathe-
matical cripple.
The literary man spends his life interpreting people and society
through his essays, verse, stories, plays, and novels. When so much
of our society is involved in the Scientific Revolution, the writer
cannot make sound interpretations of society without understanding
the fundamental ideas of science and mathematics. He need not
Your Child as a Mathematician 29

acquire a large amount of skill, but he must have first-hand knowl-

edge of the chief ideas and methods of thought.
Since the school boy or girl can hardly know how his later life
will be spent, he should not avoid a subject which is as fundamental
to so many careers as is mathematics.

What Mathematics Courses in High School?

Since all children in elementary schools are taught arithmetic,

I need not argue for mathematics at this level. The real problem
begins in the high school, where students have a choice in their
academic programs. The basic principle is that mathematics should
be studied every year throughout the entire educational process.
We seem to have the idea that, once a course has been completed,
the knowledge so acquired is permanent. Yet we all know how
rapidly we forget anything that we do not practice continually,
whether it is golf, piano playing, or book learning. Since your child
will need his mathematics in his job or in college, he should maintain
and expand his knowledge of it throughout the high school and on
into college. :
I know, of course, that large numbers of students drop mathe-
matics after one or two years in high school; and it is to their parents
that these remarks are directed. The usual reasons given for this
practice are: (1) mathematics is hard, (2) they see no use for it,
(3) the colleges do not require more mathematics for entrance, and
(4) there are other subjects of greater interest or importance. I
cannot deny that mathematics is hard, but so are most worth while
intellectual activities. Students do not learn at the same rate, and
those who are slower with mathematics are the earliest dropouts
in our subject. The schools can help prevent this by offering mathe-
matics at a variety of speeds and difficulty, so that every child can
be placed in a class that suits his needs. I do not blame a youngster
for being uninterested in a subject if he sees no relationship between
it and his later life. If our children do not see this relationship in
mathematics, our teaching needs reform, school counselors need
to be indoctrinated, and parents must be educated.
The colleges have been negligent in setting their entrance
standards to attract students rather than insisting on proper prepara-
tion for those who are admitted. These entrance requirements are
30 MATHEMATICS FOR PARENTS

rapidly being raised, and in no case should they be interpreted as

the maximum preparation needed for college work. The competition
for time from other subjects does raise difficult problems. Mathe-
matics is not the only subject a high school student should study,
and there is heavy pressure for more high school work in foreign
languages, natural science, and social science—not to mention fine
arts, driver education, and physical fitness. With careful planning,
however, there is still time to meet these pressures within a four-
year high school career. Normal students can carry four academic
subjects without difficulty, and the honors students can handle
five without strain. A typical program is the following:

English 4 years
Mathematics 4 years
Foreign language 3-4 years
Social science 3 years
Natural science 2 years
Nonacademic subjects 4 years
Total 20-21 years

This outline may be criticized in that only two years are

allocated to natural science. In my opinion it is wiser for a student
to take four years of mathematics than to expand his high school
work in science. For then he is prepared to understand fully the
more rigorous science courses which he will meet in college. In the
long run this program will put him farther ahead.
I should note, also, that the foreign language needs to be chosen
carefully. The future scientist must know French, German, and
Russian, and he should begin one of these in, high school. Spanish
has many attractions, but it is a waste of time for the scientist.
I know that there will be static from parents and teachers
concerning the recommendations as they apply to the less mathe-
matically gifted. Let it be understood that I do not propose to
drag these students unwillingly through the four-year mathematical
curriculum established for the mathematically able. Special pro-
grams adapted to their speeds and needs should be made available.
Indeed, it is these students for whom four years of work is most
important, for it takes them longer to acquire the skills needed for
survival in our age of automation. Moreover, it is not mandatory
Your Child as a Mathematician 31

that they study mathematics every day in the week. Tuesday-

Thursday courses designed to maintain their skills can be dove-
tailed with Monday-Wednesday-Friday courses in other academic
or nonacademic subjects. But again, their instruction in mathe-
matics must be continuous.
Some of the difficulty with the mathematically less able arises
from the misconception that a child is stupid if he cannot learn
arithmetic. This leads to unpleasant scenes at home, and not infre-
quently to physical violence. Naturally such treatment prevents
the child from learning mathematics; what he needs is help and
encouragement rather than a sense of failure. The schools are learn-
ing how to deal with these individual differences, but the problem
is far from solved.

Careers as a Mathematician

Let me turn now to the problems of the child who is really

good at mathematics. Parents often ask me whether there is any
future for them as professional mathematicians. The answer to
this is a resounding ‘‘yes” for both boys and girls. Until recently
the only way for a mathematician to make a living was for him
to be a teacher at a school or college, but since the war the situation
has changed completely. Industry, business, and government are
now eager to find young mathematicians, and the colleges are
bedeviled with recruiters from these companies. The supply is only
a small fraction of the demand.
The most exciting jobs in industry are open only to Ph.D.’s,
and so I urge young people to stay in graduate school as long as
they are making the grade. Salaries at this level are very attractive,
and working conditions are generally good. Mathematicians with
bachelor’s and master’s degrees are also in heavy demand, espe-
cially in connection with high-speed computing machines. Girls
compete equally with boys in such work. There are also opportuni-
ties for mathematics B.A.’s in insurance companies as actuaries.
Employment as an actuary is an excellent introduction to the
insurance business and often leads to promotion in the manage-
ment of the company. Details of and current salaries in mathemati-
cal employment are contained in the pamphlet Professional Oppor-
tunities in Mathematics, available for $0.25 from the Mathematical
32 MATHEMATICS FOR PARENTS

Association of America, University of Buffalo, Buffalo, New York,

14214.
I close with a word of caution about industrial mathematics.
The mathematician in industry must work closely with other
employees of the company and needs to be able to communicate
with them. He must, therefore, have a substantial knowledge of
what these other people are doing, and it would be wise for him to
acquire this knowledge while he is in school. By this I mean, for
example, that a prospective industrial mathematician should be
well grounded in physics and portions of engineering, and that a
prospective mathematician in business should know economics,
accounting, etc. The pure mathematician without outside knowledge
of these kinds is not likely to be useful or happy except in an
academic environment.
 aX
CHAPTER © NUMBERS
Oo

=
a

With this chapter we begin an elementary survey of some of the

important ideas in mathematics as it is now being taught in the
schools. You will observe that most of this is familiar, and I doubt
that you will have any trouble with it. The best place to begin the
story is with numbers, for these are the foundation of our subject.
The most elementary numbers are those which are used in
counting, namely: 1, 2, 3, 4, 5, .... These are called counting
numbers or natural numbers. To these we should add 0 (zero), even
though historically it was invented much later than the others.
As we shall see, all other kinds of numbers are constructed by
various devices out of these.
Properly speaking, the symbols 0, 1, 2, 3, 4,5, . . . are names
for numbers, rather than numbers themselves, and we prefer to
call them numerals, a word which means a name for a number.
Other civilizations have used different names for these numbers
and hence had different systems of numerals. For example, the
Romans used I for one, V for five, X for ten, XI for eleven, and so
on. We owe our system of numerals to the Arabs.

33
 MATHEMATICS FOR PARENTS
34

ways of repre-
Even in our system of numerals, we have many
can write “‘five’ as
senting the same number. For instance we
5,1 +4,2+3,7 — 2, 1%, V/28, ete.

Decimal Notation
that it is based
A striking feature of our system of numerals is
of us are born with
upon the number 10, apparently because most
notation is called the
ten fingers, which we use for counting. This
have made many
decimal system, and we shall see that we could
m is devis ed to use only the symbols
other choices. The decimal syste
retati ons depen d upon the place
Op le2 se Oo, DULY their interp
give this inter preta tion we choose
in which they are used. In order to
choos e ten. We define 101 =10,
a base for our system, and in this case
etc., and now are ready to
102 = 10 X 10, 10? = 10 X 10 X 10,
as 724. By defini tion:
define what is meant by a numeral such
724 = (7 X 10°) + (2 X 101) + 4,
8216 = (8 X 10%) + (2 X 10?) + (1 xX 10') + 6,
il
57.3 I 6 x10) +) + (8x z),
1 1 1
2.458 24 (+xit)+(5x ih) + (8xa)

this notation that

and so forth. As adults you are so familiar with
the child, however,
you use it automatically without thinking. For
fine points of the
this use is not so automatic, and he must learn the
he knows enough
notation if he is to understand his arithmetic: Still,
point of careful
of the notation so that it is hard for him to see the
use an unfami liar base, say seven, to
analysis. For this reason we
teach him the ideas of the place values of numera ls.
, 6 and
In base seven, we use only the symbols 0, 1,...
define, for example:

Al3seven = (4 x ) al ql x 7) + 3,

where 413.even Means the number represented by the numeral 413

te a table from
to the base seven. Using this notation we can compu
and vice versa:
which we can convert from base ten to base seven
 Numbers 35

Base Ten Base Seven

en iperen
ten Marcos
Sten eeven
Aten qeeven
5Sten Oneven
Gren Opaven
es tO earer
ten leleeren
Oren loeeren

10ten etc. WS seven Cte

In a similar fashion,

Al 8seven= (4 X 77)
+ (1X7) +3 = 196 +743
==> 206ten-

Also,
2164 seven = (2 X 75) + (1 X 72) + (6 X 7) +4
= 686 + 49 + 42 + 4
= 78 lten.

There is a convenient scheme for doing this sort of conversion

in a compact form, known as synthetic division. Let me illustrate
by redoing the illustrations above.

lol 3 «7 2 1 6 4 7
a 14, 105 777
4 29 206 2eegloe Liesl
The steps are these: Write down the base seven digits leaving
spaces between them. Bring down the first of these to the bottom
line unchanged. Multiply this by 7, and add the product to the
second digit. Continue, moving to the right. The final sum is the
appropriate numeral to base ten.
In order to proceed in the opposite direction, it is convenient
to have a table of powers of seven available:

ete
7? = 49,
73 = 348,
74 = 2401.
 MATHEMATICS FOR PARENTS
36

l
Now let us find the expression to base seven of the numera
power of seven which goes
1954 ten. We divide 1954 by the largest
into it, namely, 7* = 343, and obtain
1954°= (5% 7) = 239:

We repeat the process by dividing 239 by 72 = 49, and obtain

239 = (4 77) 4 40.
Finally
AS le (OMe) tebe

Assembling all these results we have

1954... = (0 xX 7°) + 4X 7?)+ (6X7) +1
= 5461 seven:

In a similar fashion we find that

932ten = OQx7) 16 x 7) 4 (0 XK
ll 250 Leven:

can
We can do arithmetic to the base seven just as easily as we
to base ten, provided that we rememb er what we are doing. Addi-
tion depends on the table:

ADDITION TABLE FOR BASE SEVEN

(All Numerals Are to Base Seven)

Now let us add the following where all numerals are to the
base seven:
45
+26

Adding in the usual way we find that 5 + 6 = 14, which means

Numbers Bu

one seven plus four. We write the four and carry the one into the
second column:
1
45
+26
4
We add the second column and get 4 + 2 = 6 and 6+ 1 = 10.
So we have, finally,
45
+26
104
It is a good exercise to convert all these numerals to base ten and
check this answer. [Result: 33 + 20 = 53ten.]
Similarly, we can do multiplication to base seven. The multi-
plication table is as shown.

MULTIPLICATION TABLE FOR BASE SEVEN

(All Numerals Are to Base Seven)

0
6
15
24
33
42
51

We can multiply in the usual fashion using this table; for

example, let us compute

Daperen x a onevont

The process reads as follows:

34
25
236
101
1246
38 MATHEMATICS FOR PARENTS

Here we multiplied 5 times 4 and got 26. We wrote the 6 and

carried the 2. Then 5 times 3 = 21, to which we added the carried
2 to obtain 23. In the next line 2 times 4 = 11; we write 1 and
carry 1. Then 2 times 3 = 6, to which we add the carried 1 to
write 10. Now add the two rows using the addition table. [Check:
In base ten, this problem becomes 25 X 19 = 475.|
High-speed computers commonly do their internal arithmetic
in base two notation. Here we use only the numerals 0 and 1, and
these are represented electronically by voltages, one “high”, one
‘Jow’’. A short conversion table is as shown.

Base Ten Base Two

I om tire
Zten 10two
Sten deve
Aten 100two
Oten Oe
Bten LEO ire
iikvern is
Sten 1000 two

Arithmetic in base two is very simple but tedious. The addition

and multiplication tables are as follows:

+) (Oe el en Oe?

OF 0 ie 0;90 O
Litt oLO Teed

These are easy to remember, but even simple computations are

lengthy when done by hand. For example consider 6ten X 5ten in
base two notation. We must compute

110
101
110
000
110
11110
Numbers 39

where the numerals are in base two. We then verify that

LEO To a dd x 2*) = (1 x 2°) = dl x 27) =f dl x 2) 0

1648.2
30ten-

Arithmetic to base two is frequently called binary arithmetic.

By contrast, we could choose a large base, say sixty, for our
arithmetic. Then we would need sixty different numerical symbols,
and learning the addition and multiplication tables would be an
enormous job. But once they were learned, arithmetic computations
would be very simple. Our decimal system seems to be a good com-
promise between these extremes.

Negative Numbers

If no one ever ran a deficit, or if the temperature were always

comfortably warm, these counting numbers would serve for the
normal purposes of business. But deficits do appear, and red ink is
not always available, so we must extend our number system. The
convenient device is the introduction of the negative integers: —1,
—2, —3, —4, . . . , which are defined to satisfy the equations

1+ (-1) = 0,
2 (= 2) a 0,
3 + (—3) = 0,
etc.

When you went to school, it was customary to read —2 as

minus two, but now some authors call it negative two. The reason is
that “minus” connotes the idea of subtraction, whereas we are not
subtracting here at all. As I shall explain in Chapter 7, the minus
sign has several distinct meanings in mathematics, and so it is help-
ful to distinguish among these, at least in the early stages of algebra.
After these negative numbers have been introduced, it is necessary
to settle on rules for their arithmetic. As we shall see in Chapter 8,
these rules are by no means arbitrary, but are derived from other
properties of numbers which I shall discuss in that chapter.
40 MATHEMATICS FOR PARENTS

Rational Numbers

We are also used to the appearance of common fractions in our

daily lives, and it is necessary to expand our number system to
include these. There are many approaches to fractions, but the most
direct is to define 34, say, as a number such that
(34) X4 = 3.
More generally, we must define a/b where a and bare integers (posi-
tive or negative, with b different from zero). This is defined to be a
number such that
(a/b) X b =a.
Numbers of this kind are called rational, since they are expressed by
means of a ratio.
You will observe that there are many names, or numerals, for
the same rational number. For example:
1 2 3 500
a —) ete.
5 ey eG Soon
can all be used to represent the rational number which we call one-
half. When we write the decimal expansion of a rational number, we
find that after a certain number of places the digits begin reappearing
in cycles. For example,
Veo 0333 5.8;
4 = 0.25000... ,
1¢ = 0.142857142857 ... ;
Decimals of this kind are called repeating decimals. Conversely, any
decimal whose digits repeat in cycles can be expressed in the form
a/b, and therefore is a numeral representing a rational number.

Real Numbers

In ancient times the Pythagoreans believed that all numbers

were necessarily rational, and that a harmonious human relationship
could always be expressed by numbers of the form a/b where a and
b were small integers. This belief was exploded when it was shown
Numbers Al

that there is no rational number whose square is 2, that is, that ~/2
must be irrational. In order to make room for such numbers, we
again expand the number system by including nonrepeating as well
repeating decimal expansions. The collection of all such numbers is
called the real number system.
The adjective real is quite misleading here, but is so well im-
bedded in usage that it cannot be replaced. The real numbers are
high-order abstractions from nature, and are far from “real” in the
customary sense of this word. The word “‘real’’ is used to distinguish
these from another system of numbers which our ancestors called
imaginary. Imaginary numbers arose because of the problem of
introducing +/ —1 into algebra. Since there is no real number whose
square is negative, ~/ —1 was completely excluded from considera-
tion for many centuries. Then it began to creep in as an zmaginary
number. We are now so used to it that we treat it like any other
abstract number, but the traditional names have stayed with us.
In order to complete our number system, finally we introduce
complex numbers of the form a + 62, where a and 6 are real and 7 isa
symbol that stands for ~/—1. As before, the word “complex” is a
misnomer derived from our ancestors; it is an inappropriate name,
but we are now stuck with it. We add and subtract these numbers
as in the illustrations:

(2+ 31) + (447i)

= 6 + 10i,
(lee (2ea he 6r
In multiplying two complex numbers, we proceed by ordinary alge-
bra, and then replace 7? by —1 in the next to last step. For instance,

(3 + 5i) X (4
+2%) = (8 X 4) + (3 X 2a) + (Si X 4) + (Si X 21)
12 + 61 + 202 + 102?
12 + 267 + 1022
12 + 26¢ + 10(—1)
2 + 261.

Since most of you have no occasion to use complex numbers in

your daily lives, you may wonder why we bother with them or why
your children should be asked to learn them. Although there are
practical uses for complex numbers in science and engineering, the
most important reason for discussing them is purely aesthetic. In
42 MATHEMATICS FOR PARENTS

elementary algebra we find that we can solve any equation of the

form
ar -- b= 0F

where a and b are real numbers and a # 0. Indeed, the solution is

mb
f= >
a
We run into difficulty, however, if we try to solve

az? + bs) ¢ =10)

where a, b, and c are real numbers and a ¥ 0.

For in particular
zg? +1 = 0

is a special case of such an equation, and it is not satisfied by any

real number zx. It is awkward and inelegant to have to admit that
simple equations of this kind do not have solutions. So we introduce
complex numbers to provide these solutions. We can then prove
that there is always at least one complex number x which satisfies
our quadratic equation

ax? ++ br +c = 0.

You might think that we would have to introduce fancier and fancier
kinds of numbers to solve equations of higher degree such as

Con
ae lee),
or
BEA 4 eG eae
It is remarkable, indeed, that complex numbers still do the job, so
that we need not introduce further systems of numbers for such
purposes. ,
The use of the terms real, imaginary, and complex is an illustra-
tion of how mathematical terminology differs from ordinary English.
These and only these words are proper mathematical language for
the ideas involved, and common English synonyms cannot be sub-
stituted for them. One of the more amusing but frustrating con-
fusions of this type occurred several years ago in connection with a
mathematical manuscript. This was being edited by a professional
editor for publication as a book. The editor found the use of “‘real”’
Numbers 43

repetitive and awkward, and so “genuine” was substituted for ‘‘real’’

in numerous places. Similarly, “complicated” was thought to be
better English than ‘‘complex’’, and the substitution was made. As
a result the manuscript became unintelligible to both layman and
mathematician alike. So you must learn the special vocabulary of
mathematics if you wish to read mathematics. There is no use fight-
ing against it or trying to use substitutes. There are no short cuts,
but it is not all that hard anyway.

PROBLEMS

1. Write the following in base ten notation. (a) 456seven; (b) 5623seven}
(c) 20Gsecons

2. Write the following in base seven notation. (a) 2196ten; (b) 4503 ten;
(c) 7208ten-

3. Carry out the indicated operations on the understanding that all

numerals are expressed to base seven. Then convert the given numerals
to base ten, repeat the arithmetic, and check your answer.
(a) 25 (b) 513 (c) 36 (d) 432
16 642 x15 23
+39 +135 ae “ab
4. Write the following in base ten notation. (a) 10,110two; (b) 1,110,001 two;
(c) 1,100,101 two.
5. Write the following in base two notation. (a) 46ten; (b) 123ten;
(c) 256 ten.
6. Carry out the indicated operations on the understanding that all
numerals are expressed to base two. Then convert the given numerals
to base ten, repeat the arithmetic, and check your answer.
ayia (b) 111 (c) 111 (d) 1011
10 101 Sd 101
+1 100 7 eae
= +11
7. Write the addition and multiplication tables for numerals expressed in
base three notation.
8. Use the tables obtained in Prob. 7 to carry out the following operations
where the numerals are in base three notation. Check as directed in
Probs. 3 and 6.
(a) 12 flan ol (c) 121 (d) 1222
21 210 21 «112
+11 al12 ci
44 MATHEMATICS FOR PARENTS

9: a as repeating decimals (numerals are to base ten):

(a) 16;
1 (b) 24; le 26 ; (d) Piel
3 ae
10. Carry out the ion operations on complex numbers (numerals are
to base ten):
(a) (1 + 41) + (2 + 61) (d) M4 bt
(b) (2 — 2r) + (—5 + 31) (e) (8 +1) X (2 — 31)
CC) Art) F(T 130) (f) d +7) xX Gd—14)
11. Show that 2 + 37 and 2 — 37 satisfy the equation

z?—4r+ 13 =0

where all numerals are to base ten.

Problems 12-15 are not covered in the text, but if you think hard you
should be able to solve them.
12. Write the following in base ten notation: (a) 35.2seven} (b) 21.13 seven}
(ey 10. (yl 10N te
13. Carry out the following operations on the understanding that all
numerals are expressed to base seven. Check as directed in Probs. 3
and 6.
(a) 6/66 (d) 45]1353
(b) 21|630 (e) 3426
re —1645
(c) 16/643 (f) 1536
a — 642
14. Express in the form a/b (all numerals are to base ten):
(a) 0.44444 ... Hint: Letz = 0.4444... ;
then 10z = 4.4444... ,
S010 c— 04, OOO0R me
T= Fi;
(D0 7777) eric),
0,121210 een end lecags oeee
3441 | BoP 4 a2 te 1 + 22
15. (a) +
24 Hint:
v T ry ie x b
pgm vtCerra
a+ ib godt bd) + i(be — ad)
16. Show that ———
c+ id c+ d?

ANSWERS

1.1 (8)| 237t6en3; (Db)! 2026ien; (Ce) 104¢e5.

2. (2) 6255¢even; (b)) 16,062zsven;) (©) 30; 000. von:
Numbers 45

32 (8) 106sever, (b) 1623scven; (¢) 36 (d) 482

<I SB
252 1626
36 1164
G42seven 13566.even
Check:
(a) 19 (b) 255 (c) 27 (d) 219
es 324 a2 Dalia
23 15 54 1533
55ten 654 ten 20 219
324 ten ieoten
Sia) 2een, (b) 1132; (e) 10l ten.
5. (a) 101,110tw0; (b) 1,111,011iwo; (¢) 100,000,000
two.
Bee 10,-,: (b) 10,0lliwe; (c) 111 (d) 1011
x11 101
111 1011
111 1011
10101iwo 1101 13.6
Check:
(a) 3 (b) 7 (c) 7 X 3 = 2lten; (d) 11 X 5 = Sdten.
2 5
: 4
Gren 3

19ten
io i} Mal Oia d eee

ia, 1 2 Onl Omienc0

i 62 10 (ei cO eee
eae 10 11 CO eee
Be 2 ienres (D) 1220 sircas (6) 121 (d) 1222
x21 112
121 10221
1012 1222
TiO tines 1222
1000111 three
Check:
(a) 5 (b) 16 (c) 16 K 7 = 112tea; (d) 53 X 14 = 742ten.
7 21
4 14
16a: Slten
46 MATHEMATICS FOR PARENTS

. (a) 0.5000... ; (b) 0.6666... ; (c) 0.2222... ;

(d) 0.454545... ; (e) 1.428571428571 . .
10. (a) 3-102; (b) —3 + 7; (¢) 2 40; (d) —11 — 4; (¢) 9 — Fae
11. (2 + 32)? — 4(2 + 32) + 138 = (—5 + 122) — (8 + 121) + 13 = 0.
12. (a) (2624) ten; (b) (151949)ten; (Cc) (534)ten} (d) (634)ten.
13. Llseven 30 seven 34 seven
(a) 6/66 (b) 21]630 (c) 16|643
54
103
103
22 seven
(d) 45/1353 (€) 1451,even; (f) 56 4.even.
123
123
123
Check:
Sten 21 ren 25 ten
(a) 6/48 (b) 15/315 (c) 13|325
16r00 a
(d) 33/528 (e) 1245 (En Gis
ans —670 —324
coe ii.
14. (a) 49; (b) 76; (c) 443; (d) 12369.

15. ern a) a
CHAPTER © SETS, FUNCTIONS,
! > AND RELATIONS

Before introducing the main ideas to be discussed in this chapter,

tic
let us look at the problem of building a course of study in arithme
in a systema tic fashion from the early
and algebra which will extend
arith-
grades through the high school. My first observation is that
names for two parts of the same
metic and school algebra are only
subject, that artificial distinctions between them only serve to con-
these as a single subject
fuse everyone, and that we must consider
when we plan a twelve-year curriculum.
The first steps in instruction in the early grades will be con-
ed
cerned with the properties of numbers. I have already mention
in which concret e
the need for an intuitive approach at the outset
materials such as blocks, pebbles, leaves, etc., are used to help develop
a more
a proper feeling for the subject. Gradually intuition leads to
in arith-
systematic treatment of numbers, including the basic skills
d in
metic and the fine points of positional notation which I discusse
ly called
the last chapter. For a further advance into what is common
ation and simplifi cation of the ideas and facts of
algebra, an organiz
to con-
arithmetic are now essential. Before we are finished we need
47
 48 MATHEMATICS FOR PARENTS

struct a full mathematical model of arithmetic and algebra of the

type which I described in Chapter 1. Remember the two essentials:
(1) careful terminology (both undefined and defined), and (2) state-
ments of general relationships which are commonly called axioms.
The traditional textbooks in algebra have been weak on both of
these counts, and much that may seem new to you as parents has
been introduced to remedy these faults.
In introducing terminology it is necessary to walk a fine line
between confusion and pedantry. Some mathematicians have been
so revolted at the sloppy use of words in the older books that they
have adopted language so artificial and complicated that almost no
one can understand them. Some teachers have interpreted the ‘new
mathematics” as being little more than a fancy vocabulary which
must be learned if one wishes to join the club, and they have thus
missed all the main points of the current reform movement. What is
needed is a sensible intermediate position whose objectives are clar-
ity, understanding, and simplicity. In this chapter I shall discuss
how this is accomplished in one important area of our subject. In
the next chapter I shall turn to the general principles of algebra
which can be introduced once the language is clear.

Sets

The use of the word “set” is perhaps the most controversial

topic in the current reform movement. Some think that this one
word separates the conservatives from the radicals: if you do not use
sets, you are an old fogy; if you do, you are up to date. This over-
simplification of the facts ignores the essential matter of how sets
are used. Before I can explain this problem to you, I must explain
what a set is.
The idea of a set permeates all mathematics, for it is a short
word which means a group, a collection, or an aggregate of numbers,
lines, points, etc. It is thus clear and elementary. Those objects
which belong toa set are called its elements or members; for instance,
Lincoln is a member or element of the set of presidents of the United
States.
Let us begin with some examples of sets and the notation for
them.
 Sets, Functions, and Relations 49

(1) {1, 6, 15} is the set whose elements are the numbers 1, 6,
and 15. We use braces { } to represent sets in this way.
(2) {1, 2, 3,4, . . .} is the set of all natural numbers.
(3) {Bob, Jim, Bill} is a set of boys whose last name is
Allendoerfer.
(4) {z|z? = 1}. This notation is to be read “The set of num-
bers x such that x? = 1.”’ The vertical bar stands for “such that’’.
This set can also be written {—1, 1}.
(5) &, the empty set which contains no elements. An example
of such a set is the set of men born in 1700 who are still alive.
(6) {x|az? — 52 + 6 = O}, the set of solutions of the equation
aor | 6 = 0, namely, {2, 3}.
(7) [a, b], a special notation for the set of all real numbers which
lie between a and 6 inclusive, that is, those which satisfy the inequal-
ities a < x < b. This set is called a closed interval.
(8) Ja, b[, a special notation for the set of real numbers satisfy-
ing a < x < b. This set is called an open interval.
(9) {(a, y)|8a — 4y + 12 = 0}. This notation requires some
further explanation. The symbol (z, y) stands for a pair of numbers,
x and y; moreover, it is understood that the pair (x, y) is to be dis-
tinguished from the pair (y, x)—that is, the order of the two num-
bers in the pair is of importance to us. For this reason we call (z, y)
an ordered pair of numbers. The set defined above is, therefore, the
set of those ordered pairs (x, y) whose components x and y satisfy
the equation 3x — 4y + 12 = 0. Examples of elements of this set
are (—4, 0), (0, 3), (8, 9); there are in fact infinitely many such pairs,
and so we can never write down a complete list of them.
(10) {(2, y, 2)|Ba — 4y + 62 + 12 = 0}. By analogy with (9)
we call (x, y, z) an ordered triple. The elements of this set are those
ordered triples whose components satisfy the given equation.
Examples (9) and (10) raise a special problem in that it is not
possible to give a complete list of the elements of the given sets or to
use a notation such as in (2) to suggest what these elements are.
Thus we must invent another technique to help us see what the set
is; namely, we construct the graph of the set. To do this we draw two
perpendicular lines in the plane, as in Fig. 5, which we label the
X-axis and the Y-axis. We plot the point P as the graph of the
ordered pair (x, y) by measuring z units along the X-axis and y units
50 MATHEMATICS FOR PARENTS

along the Y-axis. Along the X-axis we measure to the right of O if x

is positive and to the left if x is negative. Similarly, along the Y-axis,
positive y are measured above O and negative y below O. Thus we
have a point in the plane corresponding uniquely to each ordered
pair of numbers (x, y). The graph of {(z, y)|8e — 4y + 12 = 0} is
then the set of points in the plane, each of which is the graph of an

Fig. 5. Coordinate system.

ordered pair of numbers (x, y) whose components satisfy 3x — 4y +

12 = 0. It can be proved that for an equation of this type (in which
x and y occur to the first power only) the graph is a straight line as
in Fig. 6.
We can do the same thing for the set in (10), but here we require
three dimensions. We set up axes as in Fig. 7 and find a unique point
corresponding to each ordered triple (x, y, 2). We can prove that the
Sets, Functions, and Relations 51

graph of a set such as {(z, y, z)|38x — 4y + 62 + 12 = 0} isa plane

in our space.
In both of these cases the graph helps us to visualize the given
set, but it is not the given set. There are two sets involved in our
discussion of (9). First, we have the set of ordered pairs of numbers

Yi

Fig. 6. Graph of {(z, y)|8a — 4y + 12 = 0}.

(x, y) which satisfy the given equation. Second, we have the set of
points in the plane which lie along the graph we have plotted. These
two sets are closely related in that to each pair (a, y) there cor-
responds a unique point, and to each point there corresponds a
unique pair (x, y). Situations like this occur in many other places in
mathematics, and so it is worth while for us to give a name to such
a correspondence.
32 MATHEMATICS FOR PARENTS

ONE TO ONE CORRESPONDENCE. Two sets A = {ai, do, a3, .. -}

and B = {b, be, bs, . . .} are said to be in 1 to 1 correspondence if
and only if there exists a pairing of the a’s and b’s such that each a

Fig. 7. Three-dimensional coordinate system.

corresponds to one and only one b and each b corresponds to one and
only one a. We write such correspondences in several ways. For
example,
John Jim _ Jerry
J 1 if
Bill Bob Bert

represents a 1 to 1 correspondence between the sets John, Jim, Jerry

and Bill, Bob, Bert. Another such correspondence is

John Jim Jerry

J J J
Bob Bert Bill
Sets, Functions, and Relations 53

If we assign the integer 2n to each positive integer n, we thus

establish the 1 to 1 correspondence n © 2n or
Toeee ae
a aay
2 oe sO
between the set of positive integers and the set of even positive
integers.
This notion of a 1 to 1 correspondence is well understood by
children, even though they do not know the name for it. Indeed, the
process of counting on one’s fingers is based precisely on this idea,
for this establishes a 1 to 1 correspondence between fingers and the
objects to be counted.
Let me use examples (9) and (10) to illustrate how set notation
can clear up ambiguities which frequently bother students. In the
traditional books students are asked to “Graph the line 382 — 4y +
12 = 0” or to “Graph the plane 3z — 4y + 62 + 12 = 0”. This
causes no great problem, but difficulties arise when we consider the
equation of a plane in which the coefficient of z is zero, such as
3x2 — 4y + 12 = 0. Such planes are parallel to the Z-axis. You can
already see the confusion: Is the graph of 3x — 4y + 12 = 0a line
or a plane parallel to the Z-axis? The answer depends on the point
of view or the instructions given with the problem. Unfortunately
on tests or sets of review problems there can easily be ambiguity
regarding the meaning of ‘Plot the graph of 3x — 4y + 12 = 0.”
This can easily be clarified by using set language:
Plot the graph of {(z, y)|82 — 4y + 12 = 0} (two dimensional)
or
Plot the graph of {(z, y, z)|]Ba — 4y + 12 = 0} (three dimensional)
Students greatly appreciate having problems stated in unam-
biguous language so that they can search for the solution without
having to try to guess what it was that the teacher or textbook had
in mind. Set notation is one tool that helps to make such clarity
possible.
Perhaps another example of the use of set language would be
appropriate. In geometry students are faced with locus problems. A
simple illustration is: ‘Find the locus of a point in the plane which
54 MATHEMATICS FOR PARENTS

moves so that its distance from a fixed point is a constant.” The

answer, of course, is “‘a circle’. There are two troubles, however,
with this statement of the problem. First, the word locus is unfamil-
iar and is usually not well understood. In any case it is a special
word that needs careful explanation. Second, there is the reference
to motion of a point. In physics particles do move, but in math-
ematics a point is not a particle and its motion is at best a pseudo-
physical idea and not a mathematical idea. The situation becomes
clearer and more precise if we formulate it in set language: ‘Given
a fixed point O in the plane, find the set of points P in the plane such
that the distances PO are all equal.’ If the concept of set is under-
stood, there are no unfamiliar words and no fuzzy problem with
motion. This is the way mathematics should be written.

Doing Things with Sets

Mathematics is much more, however, than a collection of words
and definitions. After we know what the words mean, we must do
something with them. We must then turn to doing things with sets.
First we need to be clear on the matter of when two sets, de-
scribed in possibly different ways, are the same, or are zdentical.
1. ipeNTITY. Two sets are said to be identical if and only if they
contain precisely the same elements. The order in which these ele-
ments are written is immaterial. In such a situation we write A = B.
For example,
{6,115 9°23 = 4172, (67995
foley ft Ly.
In the succeeding paragraphs I shall define other operations or
ideas which are useful in working with sets. .
2. sussET. A set A is a subset of a set B if and only if every ele-
ment of A is an element of B. The notation for this is A C B, read
“4 is a subset of B.”’ When B contains elements that are not in A,
we write A C B and say that A is a proper subset of B.
For example,
(Gol Geel 27 Onl 2a i,
{2 Al68 ce ol nena eee
The set of all red-haired girls is a proper subset of the set of all girls.
Sets, Functions, and Relations 55

3. UNIVERSAL SET. Within a fixed mathematical framework such

as algebra or geometry, we are dealing with sets whose elements are
members of some large, fixed set, called the universal set, U. For
example, U may be the set of all real numbers, the set of all points
in the plane, etc. The specific sets with which we deal will then be
subsets of U.
4. COMPLEMENT. Let A be a subset of the universal set U. Then
its complement, written A’, is the set of all elements of U which are
not members of A. As an illustration, let U be the set of all
integers, and A be the set of even integers. Then A’ is the set.of odd
integers. Or, let U be the set of all living people and A be the set of
all blind people. Then A’ is the set of all people who can see.
5. UNION OF TWO SETS. The union of sets A and B, written A U B,
is the set of those elements which are members of A or B or both

GQ) A
Gy B

(a) (b) (Cc)

Fig. 8. Union of two sets.

A and B. The idea is illustrated in Fig. 8 for pairs of subsets of the

plane. The union of A and B is shaded.
For example, let A be the set of positive integers and B be
the set of integers whose square is 4. Then A U B is the set
mero 34. ..}.
As another example, let A be the set of red-haired girls and B
be the set of blue-eyed girls. Then A U Bis the set of girls who have
red hair, blue eyes, or both.
6. INTERSECTION OF TWO SETS. The intersection of sets A and B,
written A ()\ B, is the set of those elements which are members of
both A and B. See Fig. 9 for an illustration of this idea. A ( B is
shaded. For example, let A be the set of red-headed girls, and B be
56 MATHEMATICS FOR PARENTS

the set of blue-eyed girls. Then A ()/ B is the set of red-headed girls
who also have blue eyes. As another example, if A = {1, 6, 9, 12},
Be 1Ont? a1> pethen:
AB = (9, 12}.
Now let us apply these ideas to. some familiar problems in math-
ematics. You are used to questions such as ‘“‘What is the solution of
x? + 7x +12 = 0?” but before attempting an answer we must
understand the problem. In the first place we call x a variable, and
we need to say what “variable”? means.

Es . @
(a) (b) (Cc)
Fig. 9. Intersection of two sets.

7. VARIABLE. A variable is a symbol which may be replaced in a

mathematical expression by any element of a given universal set.
Some authors say that a variable is a place holder which may then be
replaced by any element of U.
A formula (or a sentence) containing a variable is called an open
sentence if after the variable is replaced by an element of U there
results a formula which can be meaningfully called true or false. For
example,
r+3=5

is an open sentence. For when x = 4 we have

4+3=5 (which isfalse),

but when x = 2 we have

2+3=5 (which is true).

8. TRUTH SET. The truth set of an open sentence is the set of ele-
ments in U whose substitution for the variable in the open sentence
Sets, Functions, and Relations 57

gives true statements. For example, the truth set of

ta ee oO

is the set {—3, —4}. The truth set of

x was president of the United States on Jan. 1, 1964

is the set {Lyndon B. Johnson}. In general we write truth sets in

the notation:
{a|z2 + 7x + 12 = O},
{tlx? > 4},
{(x, y)|2x + 4y — 6 = O}.

Thus the notion of a truth set includes the familiar idea of the solu-
tion of an equation, but it is a broader concept. It applies to inequal-
ities and to open sentences in two or more variables, such as

{c|z? + 524 + 6 > 0},

{(z, y)|8a + 4y — 5 = 0},
{(a, y)|e? + y? < 1}.
Thus by using this notation we can formulate an important
class of mathematical problems in a precise fashion, namely, we are
asked to find the truth sets of given open sentences.
As an illustration of the use of the concept of wnion let me re-
mind you of the method of solving quadratic equations by factoring.
Suppose we wish to solve

ee Aah tte) 2 ae):

This is equivalent to finding the truth set

{alz? + 7x + 12 = 0}.

From elementary algebra we know that

r+7x+12 = (4+ 3)(a4+ 4).

We also know that, if a product is equal to zero, then at least one of

the factors is zero. Hence, if x satisfies x? + 7x + 12 = 0, it must
satisfy at least one of the following: x + 3 =0, or x+4=0.
58 MATHEMATICS FOR PARENTS

Further, if x satisfies either x + 3 = Oor« + 4 = O, it also satisfies

2? + 7x + 12 = 0. A summary of these two statements is

{ele? + 7¢ +12 = 0} = {ale +3 = 0} U {ale +4 = O}.

Now {zie +3 =0} = {-3} and {ale +4 =0} = {4}.

Therefore,

{xla? + 7a + 12 = 0} = {—3} VU {-4} = {-3, —4}.

A particular use of the concept of intersection occurs in finding

the “simultaneous solution” of

22+ 3y = —1,
z—-y=2.

What we want is an ordered pair (x, y) which is an element of both

{(z, y)|2e + 38y= —1} and {(@, y)|z& — y = 2}.

Thus we are in fact looking for the truth set:

{(x, y)|2e + 8y = —1} A {(@, yz — y = 2}.

Geometrically we are seeking the point of intersection of the
two lines which are the graphs of the two given equations. The
answer is (1, —1), but we shall not enter into the method of finding
this solution in this chapter.
9. FuNcTIONS. When you were in school, a function was generally
described as some kind of a formula, such as
y = 327 — 22 + 1

In the traditional notation 3x2 — 2x + 1 was called a “function of

x’’, and the students were taught to write

tt) = Be 2a 1.

The symbol f(x) is read “f of x” and means “function of x’. This

idea was so well impressed on my contemporaries in college that our
revered professor of mathematics was nicknamed “‘f of x’, or “Effie”
for short. Although no one addressed him by this name, I gather
that he was aware of it, for in his will he left money to endow the
 Sets, Functions, and Relations 59

“f(x) Scholarship in Mathematics”. You can imagine the problem

this presented to the administrator of his estate, who was a math-
ematically illiterate business man.
In contemporary mathematics functions play a much wider role
as relationships between two sets. Let us consider the following pairs:

A B
Real number x The real number x? — 2x + 7
A letter The postage required to send this letter by
airmail
A positive real number x The logarithm of x
A telephone subscriber His phone number
A college student His age
A circle The length of its radius
A town The dot on a map which represents this town

What is it that is common to these very assorted pairs of num-

bers, people, objects, etc.? First, the element on the left is a typical
member of a well-defined set: the set of real numbers, the set of
letters, the set of circles in the plane, etc. This set is called the
domain of the function we are defining. The element on the right is
a member of another set called the range of the function. So in each
pair two sets are involved, the domain and the range. But there is
still more to the idea of a function. In each case in the table there is
a procedure which associates a particular element of the range with
a chosen element of the domain. I call this the rule.
A function, then, has three components:
(1) A set, called the domain.
(2) A set, called the range.
(3) A rule. In applying this rule we choose an element of the
domain and are led by the rule to a unique element of the range.
It is also possible to look at functions from a more sophisticated
point of view. Let us consider two sets, A and B. We define the
Cartesian Product of A and B (written A X B) to be the set of
ordered pairs (a, b) where a is an element of A and b is an element of
B. Let us direct our interest toward various subsets of A X B. Any
such subset is called a relation.
For example, consider the case where A and B are each the set
of real numbers. Then A XBis the set of ordered pairs of real num-
bers. As a subset consider those ordered pairs (a, b) where b = 2 — a.
60 MATHEMATICS FOR PARENTS

We get an infinite collection of pairs such as (1, 1), (0, 2), (3, —1l),
(114, 14), ete. The set of all such ordered pairs is the relation so
defined.
Another relation is the set of ordered pairs whose elements sat-
isfy the inequality: a? + b? < 16. Particular elements of this set are
( Ly d, 2), ae 3), (2, iD) (2,2), etc.
These two relations differ in one important respect. In the rela-
tion defined by b = 2 — a, each value of a determines a unique
value of b. Hence this is an example of a function. On the other hand,
in the relation defined by a? + b? < 16, there are infinitely many
values of b that can be associated with a = 1. Indeed any b such that
b2 < 15 will work. Thus this relation is not an example of a function.
So that we may identify those relations which actually are functions,
we shall define a function as a special case of a relation.
Definition. Consider a relation which is a subset of the Cartesian
Product A X B. This relation is a function if and only if no two |
ordered pairs (a, b) which are elements of this relation have the same
first element, a.
From this point of view a function is a set of ordered pairs,

{(a1, bi), (aa, be), (aa, bs) . « «4,

in which no two a’s are the same. There is no such restriction on

the b’s, which may be equal or different as you wish. The domain
of the function is the set {ai, a2, a3 .. .}. The range is the set
{bi, bo, bs . . .}. The rule is: to each a in the range associate the b
with which it is paired above.

PROBLEMS

1. Which of the following sets are identical:

(a) {1, 3, 7, 9}; (b) {xle? = 4}; (ec) {alz = 2}; ) 1-2, 2};
(e) {7, 3, 1, 9}; (f) {a|5a = 10}.
2. Which of the following pairs of sets can be placed in a 1 to 1 corre-
spondence:
(a) A = the set of nations in the world.
B = the set of heads of state of the nations in the world.
(b) A = {1, 3, 5,7}; B = {1, 8, 5, 9}.
(c) A= ae 3, 5, ae i te 3, 5, i 9}.
Sets, Functions, and Relations 61

(d) A = the set of married men in the United States.

B = the set of women in the United States.
(e) A = the set of your children.
B = the set. of PTA’s to which you do (or should) belong.
(f) A = the set of problems in this book.
B = the set of problems in this book which you can solve.
. Suppose that the universal set U is {1, 2, 3, 4, 5, 6, 7, 8, 9, 10}. Find
the complements of:
fino, 6}; (b)*{4,°7, 89, 1035" (c) {1 2,3, 4, 5, 6, 7,8, 9, 10}.
. Find the union (A U B) and the intersection (A (\ B) of the pairs
of sets:
(year 71, 3, 5,7, 9}; B= {1 2, 3}.
(b) A = The set of positive even integers; B = The set of positive
odd integers.
(c) A = the set of boats with sails; B = the set of boats with motors.
mee it? — 724+ 12 = O}2 B= {elr = 3h.
. Find the truth sets of the following open sentences:
(a) 27 — 4 = 0.
(b) #2? —4¢4 +4 =0.
(c) 22? — 7x + 10 = 0.
(d) x was the secretary of state on Jan. 1, 1964.
(e) x (teams) played in the World Series in 1963.
. Plot the graph of the set {(a, y)|e¢ + y = 1}.
. Write down the sets of ordered pairs that are elements of the function
for which:
_ The domain is the set of positive integers less than 4.
The range is the set of positive integers less than 16.
The rule is: to an integer x in the domain assign the integer x? in the
range.
. Write down the sets of ordered pairs that are elements of the function
for which:
The domain is the set of your children.
The range is the set of their grades last term in mathematics.
The rule is: to each child assign his grade.
. Write down three sets of ordered pairs belonging to the relation defined
by a? + y? < 25, where z and y are real numbers.
10. State whether the following equations define functions, or only rela-
tions. Assume that x and y are real.
(2) eo cel or (d) y= 3a? + 7x = 2:
ae iy? =a; (ey? = 77-4.
(ey
> 22 -+1;
62 MATHEMATICS FOR PARENTS

ANSWERS

1. (a) and (e); (b) and (d); (c) and (f).

2. (a); (b); not (ec); not (d); (e) depends on your situation; (f) hope-
fully yes, probably no.
. (a) {2, 4, 5, 7, 8, 9, 10}; (b) {1, 2,-3, 5,-6}5 (ce) the empty set.
. (a) Union: {1, 2, 3, 5, 7, 9}; intersection: {1, 3}.
(b) Union: the set of positive integers; intersection: the empty set.
(c) Union: the set of boats with either motors or sails or both; inter-
section: the set of motor sailers.
(d) Union: {38, 4}; intersection {3}.
. (a) {2}; (b) {2}; (ce) {2,5}; (d) Dean Rusk; (e) {Dodgers, Yankees}.

G1); 2, 4);.@, 9). ,

. Hopefully something like this: (John, A), (James A), (Mary, A),
(Margaret, A).
. (1, 1), (2, 3), (0, 0), and many more.
10. (a) function; (b) not a function; (c) not a function; (d) function;
(e) not a function.
a O<
CHAPTER / THE LAWS OF
= -. ALGEBRA

In Chapters 1 and 2, I spoke about the organization of knowledge

into a mathematical model, once the facts were known from exper-
ience and intuition. In the elementary school the children meet num-
bers'in many situations, and when they reach high school they should
be ready to organize what they already know about numbers. In due
course they will be able to deal with the operations on numbers in
terms of an abstract mathematical system, known as a field. They
will be able to do this more readily if their intuition and thinking are
guided by a teacher who knows the details of the ultimate goal. In
this chapter we shall describe the nature of a field, so that you can
see where your children are being led in their earlier years. When
they are juniors or seniors in high school they will then have no
trouble with this abstract system.
I wish to make it clear, however, that I do not advocate the
introduction of these laws as axioms until the last year of the high
school. In the elementary school and the early years of high school
they should be introduced as summaries of information which is
already known to the children. In this way the youngsters will be
63
64 MATHEMATICS FOR PARENTS

helped to organize their mathematical knowledge. Finally, in the

senior year, the approach should shift, the laws should become ax-
ioms, and the day-to-day algebraic operations should be proved on
the basis of these axioms. In this chapter I shall assume that you
have seen algebra (at some time or other) and that you are ready for
a summary of what you remember, or at least a review of what you
once knew.
The easiest way to approach any abstract idea is to start with
a concrete illustration of it. So we approach fields by examining the
algebraic properties of the real number system.

Properties of the Real Numbers

1. cLosurE. If you add two real numbers it is evident that the

sum is another real number. Similarly, the product of two real num-
bers is a real number. For this reason we say that the real numbers
are closed under addition and multiplication.
This property may seem so trivial that it is not worth stating.
But there are familiar systems that are not closed in this way. The
sum of two prime numbers is not in general a prime, for 3 + 5 = 8,
which is not a prime. The product of two irrational numbers is not
necessarily irrational,
for \/2 X +/8 = +/16 = 4, which is rational.
The difference of two positive numbers is not necessarily positive,
for 3 — 5 = —2. Thus this closure property of the real numbers
describes something about them which is not shared by many other
sets of numbers.
2. COMMUTATIVE LAws. Another simple property of real numbers
is that their addition and multiplication are commutative.. This
means that
atb=b+a and axXb=bXa.
Children learn these in the early grades from actual experience with
the addition and multiplication tables, and it may be shocking to
visualize algebraic systems in which these laws fail. Perhaps the
simplest such example is that of the multiplication of 2 X 2 matrices.
A 2 X 2 matrix is a square array of real numbers written in the
form
a bY
c ad
The Laws of Algebra 65

By definition the product of two such matrices is obtained from the

formula
a b (ae aw+ by ax+ bz\
c d y 2) \cewt+dy cx+dz
I am sorry that space does not permit me to motivate this definition,
but please believe me—there is a good reason for it. If we apply this
in a particular case, we get

whereas
(ey(uaecas)
GS Or)
Since

Gay
this type of multiplication is not commutative.
3. ASSOCIATIVE LAWS. The addition and multiplication tables tell
us how to add or multiply a pazr of real numbers. But there is nothing
in these tables that helps us with the sum 2 + 5 + 9 or the product
4 X 6 X 2. The sums and products of triples of numbers need to be
defined, and similar problems occur when we deal with quadruples
and larger sets of numbers.
If this matter bothered you when you were in school, you have
undoubtedly forgotten about it by now, but the situation is more
serious for your young children. What shall we do about 2 + 5 + 9?
There are at least two possibilities. First, you might add 2 to 5 and
get 7, and then add 7 to 9 and get 16. We write this as
OPER eee Jer het Osa G,
On the other hand, you might add 5 to 9 and get 14, and then add
2 to 14 and get 16. We write this as 2 + (5 + 9) = 16. From this
example we observe that
(2cteap eat Oo) 2-5 (5 9).
After some experience we realize that, for every triple of real num-
bers, a, b, c,
(a+b)+c=a+(b+ 0).
66 MATHEMATICS FOR PARENTS

This is the Associative Law for Addition. The parallel law for mul-
tiplication is
(dix Dy Ge = 08x sO c)-

We must return to the definition of a +b +candaXb Xe.

We have two choices, and because of the associative laws they are
equivalent. So we define
at+b+c=(at+b)+¢,
a Ol c= (CO Xe:
»
Now what shall we do abouta +b +c+d?Sincea+b+¢c
and a X b X c are already defined, it is legitimate to define
atb+c+d=(at+b+c) +d,
aX bX COX = (ax bo)
In a similar fashion we proceed step by step to define the sum and
product of any finite set of real numbers.
The associative and commutative laws tell us that the sum of a
set of numbers is the same regardless of the order in which we per-
form the operation. Let us prove, for example, that
atb+tc=ct+bt+a.
Proof:
at+tb+c=(at+b) +c (Definition)
(b +a) +c (Commutative Law)
c+ (b+a) (Commutative Law)
(c+b) +a (Associative Law)
lI e+b+a. (Definition)
A similar situation holds for multiplication.
4. IDENTITY ELEMENTS. In Chapter 5 we remarked that zero has
the special property that for all real numbers a
480-10 tesa
This is the definition of zero, and all other properties of it must be
proved in terms of this. Since the addition of zero leaves a identically
as it was originally, we call zero the additive identity element.
It is now reasonable to ask whether there is a multiplicative
identity element, that is, a number such that a times this number is
a itself. Of course there is; this number is 1. We observe that for
The Laws of Algebra 67

every a
G@ <i=jixXG=a

and therefore call 1 the multiplicative identity element.

5. INVERSE ELEMENTS. To define such elements we pick an a and
ask: Are there numbers x and y such that
a-2=2-+-a= 0) ay = ya
= 1?
If a = 3, we know the answers, namely, x = —3 and y = 14. If
= --2, the answers are x = +2, y = —1l4. If a = 0, we have
x = 0, but there is trouble with y. We want 0: y = 1. As we shall
see later, 0: y = O for all y, and so there can be no y with the
required property.
From these examples we see that given any real number a there
is another real number ~a such that
Ol(Oso!

The number ~a is called the additive inverse of a. Some people write

—a instead of ~a, but this is confusing for it gives the false impres-
sion that ~a is negative. The sign of ~a depends upon that of a! For
example: —(3) = —3;—(—3) = 3, —(0) = 0.
Moreover, unless a = 0, there is another real number a’ such
that
axa=a Xa= 1.
This number a’ is called the multiplicative inverse of a. For example,
2! = 44; (34)! = 2; (-3)' = — 34.
In terms of these inverse elements we can define subtraction and
division.
Definitions. (a — b) and (a/b) are defined by the equations:
a—b=a+(-d),
a/b =a X (b’) provided b ¥ 0.
There is one more law which we shall only state here. It is the
Distributive Law, and its consequences are so numerous that they
require the whole of the next chapter.
6. DISTRIBUTIVE LAW. For every triple of real numbers, a, ), c,
aX(b+c)=(@Xb)+
(Xo).
68 MATHEMATICS FOR PARENTS

These six laws contain all the necessary information about the
algebra of real numbers—apart from their order properties to be
discussed in Chapter 9. They therefore form the basis of algebra, and
every theorem and rule of algebra is implicit in them.

Fields

Our final step is to construct a mathematical model based upon

these as axioms. We define it as follows:
A field consists of:
(a) A set, F, of undefined elements, containing at least two
distinct elements.
(b) A pair of undefined operations, + and X, called addition
and multiplication.
(c) The following set of six axioms:
1. F is closed under + and X. ;
2. The operations + and X are associative.
3. The operations + and X are commutative.
4. There exist additive and multiplicative identity elements, 0
and 1 respectively.
5. Every element of F has an additive inverse, and every ele-
ment other than zero has a multiplicative inverse.
6. Multiplication is distributive over addition.

Elementary algebra is the subject in which the properties of a

field are studied. Let us, then, consider some of the consequences of
these axioms as they apply to elementary algebra. That is, we shall
prove three theorems based on the above axioms.

Theorem 1. ~(-a) = a.

Proof:
eed) eo) — a0). (Axiom 5)
Now add a to both sides. You get:
2. (Ca) ++ Ca)) + o=.0 + a,
or
3. —(a) + [Ca) + a] =a. (Axioms 2 and 4)
4, ~(Ca) +0 =a. (Axiom 5)
Don (a) star (Axiom 4)
 The Laws of Algebra 69

Theorem 2. ~(a + b) = (-a) + (-b).

Proof:
1 -(a+b)+ (a+ 5b) =0. (Axiom 5)
Add (~a) + (~b) to each side:
2. [(a+ 6) + (a + b)] + [Ca) + (b)] = 0 + [Ca) + (6)].
ee oe 0) [(a +b) + Ca) + Cb) = Ca) + Cd).
(Axioms 2 and 4)
4. (a+b) +0 = (a) + (2). (Axioms 2-5)
0. (a+b) = (a) + (bd). (Axiom 4)

Theorem 3. Let aand b be any two elements of F. Then there exists a

unique element of F, x, such that

a+ a = bd.

This theorem has two parts: existence and unicity of z.

Proof of existence: x = b — a has the required property, for:

1 a+(b—a) =a+[b+(a)] (Definition of subtraction)
2. =a+[(Ca) + }] (Axiom 3)
oe = [a+ (—a)] +} (Axiom 2)
4 =0+06 (Axiom 5)
5 = b. (Axiom 4)
Proof of unicity: Suppose there are two such x which are unequal;
that is, suppose that:
enc + v1 = b.

Dee “+ ty, = b.

ae V1 F~ Xo.
Then from 1 and 2 it follows that
4. a+t+a=a-+ xo.
Add ~a to each side; the result is:
». —“a+ 2] ([=-a+ a+ [a+ xl,
or
6. [(-a) +a] + 21 = [Ca) +a] 4+ a2. (Axiom 2)
7% O+%,=0+4+ 22. (Axiom 5)
ae (Axiom 4)
Step 8, however, contradicts step 3. This means that the assump-
tions 1, 2, and 3 cannot all be true and hence that there is only one x
with the required property.
70 MATHEMATICS FOR PARENTS

This proof of unicity is an example of an indirect proof. I shall

discuss this type of reasoning in Chapter 13. If you are confused
about it, it would be a good idea to turn there now.
There are quite a few more theorems of this type, but these
should be enough to whet your appetite (or do you in!).

Examples of Fields

I have already described how the real numbers form afield, and
you may properly ask: Are there other number systems which are
also examples of fields? Yes, indeed. There are just lots and lots of
fields lying around in mathematics. Two of the most familiar are the
fields of rational numbers and of complex numbers. But these behave
so much like real numbers that they are not particularly worth
looking at here. I should prefer to show you some fields that are
quite different from any that would easily occur to you. These are
called finite fields or modular fields.
As an introduction to such matters let me ask you the
question:
“Tf it is now 9 o’clock, what time will it be seven hours from
now?” I hope you give the correct answer, ‘4 o’clock’’, but the
important question is ‘How did you do the problem?” One sys-
tematic way of handling such questions is to add 7 to 9 and then
subtract 12: 7 +9 —12= 4. Try this method to answer the
question:
“Tf it is now 3 o’clock, what time will it be fifteen hours from
now?” Answer ‘6 o’clock”’. A slightly harder question is:
“If it is now 5 o’clock, what time will it be thirty hours from
now?” To solve this we compute 5 + 30 — 2(12) = 11 o’clock.
In this way we can construct a new kind of addition which is
the same as ordinary addition, except that from the ordinary an-
swer we subtract that multiple of twelve which will leave a remainder
between 0 and 11 inclusive. Arithmetic of this kind is called modular
arithmetic, the modulus in this example being 12.
I chose 12 as an example because of its practical use in figuring
time, but it is too complicated for the rest of what I wish to do. Let
me then carry on from here using 5 as a modulus. To see how to add
in this system, let me write down the addition table.
 The Laws of Algebra 71

ADDITION TABLE, MODULUS 5

se (yah ae

©
mwhdor
©
re
Bmwhwo
COoFW
Ne aH
bd
Ww
Pe
Oo Nor
oF
& Of,
OnrF

The entry 2 + 4 in this table, for example, is obtained by the

computation: 2 + 4 — 5 = 1. Note that there is no point in extend-
ing the table to numbers such as 6, 10, 22; for these are the same as
1, 0, and 2 from this point of view.
Let us see what properties of addition we can observe in this
system. First, you can check to make sure that this addition satisfies
those of the axioms ofa field that refer to addition. Let us try a few
samples:

(1) The sum of any pair of numbers chosen from the set
10, 1, 2, 3, 4} is another number in this set. Hence addition is closed.
(2) To test the associative law is tedious; let us check a few
cases:
24(+4)=24+2=4 (243)4+4=04+4=4,
Mea 2) el tla?) (1 $4) +2204 2=2.
Others work the same way.
(3) The commutative law certainly holds, for

2+3=0, 3+2=0,
and so forth. Indeed, the whole table is symmetrical about the diag-
onal running from upper left to lower right.
(4) From the first line it follows that, for any a,
O+a=a.
(5) The additive inverses can be computed from the table. They
are:

O50 Pel od 2 ee BBD) hae

72 MATHEMATICS FOR PARENTS

for they must satisfy a + (-a) = 0, or

O+0=0; 1+4=0; 24+28=0; 3+2=0; 4+1=0.

These facts enable us to solve equations of the form

as

ata=b

in this number system, for Theorem 3 assures us that this is possible.

Consider the equation

8+2=1.

If we were working in the field of real numbers, it would be a correct

procedure to subtract 3 from each side and obtain x = —2. In this
system, however, there are no negative numbers, and so it is best to
return to first principles. To remove 3 from the left side, the thing to
do is to add ~3 to both sides. This gives

-3+(8+4+2) =3+4+1.

But we saw that ~3 = 2, so this is

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

or
0+2=3,
or
x= 3.

In solving a problem of this kind, you can really get to the

essentials of elementary algebra, and do not need to rely upon ill-
understood short cuts such as ‘‘move 3 to the right side and change
its sign’”’. In this way problems in modular arithmetic help you to
understand ordinary algebra, for they show you how things work in
a general field and hence in the field of real numbers.
To complicate matters, it is amusing to write down the multipli-
cation table in this system.
The Laws of Algebra 73

MULTIPLICATION TABLE, MODULUS 5

x ORs ee?oar ere

Pm
© eS= ©
SS
woNWr a No
©
Ww
PR
PwOnNWF Ke
©
ke
wow

As before we can check to see that this kind of arithmetic satis-

fies the multiplicative properties of a field. The only complicated
part is finding the multiplicative inverses. They are
Val, Y=38; FJ
=2, = 4,
Notice that 0 has no multiplicative inverse.
Now we can solve more complicated equations such as
da + 4 = 3.
First add ~4 (which equals 1) to both sides. The result is
(37 +4) +1=341,
or
3a + (4+ 1) = 4,
or
dx = 4.

In ordinary algebra you would now divide by 3 and get x = 44; but
here this is nonsense since 44 is not one of our numbers. Instead we
multiply both sides by 3’ = 2:
(253)eo 4s
or
re.
or
x= 3.

If, finally, we check the distributive law, we see that the integers
with modulus 5 satisfy all the axioms of a field. Thus there are some,
strange looking “‘fields’’.
You may indeed wonder if there are any familiar systems of
numbers which are not fields. By all means! The positive integers:
74 MATHEMATICS FOR PARENTS

1, 2,3, 4, . . . do not form afield, for there is no additive identity

and there are neither additive nor multiplicative inverses. Neither
does the set of all integers,

i ite = <=, —2, —T, 0; 1,2, SA ees els

#

for there are no multiplicative inverses.

Since, however, fields constitute those number systems in which
your old familiar algebra still works, they are the major objects of
study in school algebra.

PROBLEMS

ils Jeroen (y= w,

2 Prove aan (G0) e100
8. Prove: There exists one and only one solution of the equation

ax =b where a0.
4. Prove: There exists one and only one solution of the equation

ax+b=c where a= 0;

5. Prove: (a) = a where 64#0,c#0,d #0.

(c/d) ac
6. In the number field with modulus 5, solve:
(a) 4+ 2 = 2; (d) 3% + 2 = 4;
(b) 3+2=1; (e) 22 +3 =1;,
(c) 2+2=4; (f) 424 +4
= 2.
7. Write the addition and multiplication tables for modulus 4. Do these
satisfy the axioms of a field?
8. In the number system with modulus 4, can you solve the equation:
24 + 3 = 2?
9. Show that the number system with modulus 3is a field.
10. (Hard) For which moduli do the corresponding number systems
form a field?

ANSWERS

1. Imitate the proof of Theorem 1.

2. Imitate the proof of Theorem 2.
3. Imitate the proof of Theorem 3.
The Laws of Algebra 75

4. A solution is: 2 = ~~". To prove this is unique, suppose there are two
a
different solutions x; and x» with x, ¥ x». Then

az, + 6 = ax. 4+5,

Do = Wx,
a‘(ax,) = a'(aze),
(LO = (GH > Ovex,
Laie Le 2,
1 = Xo,

which is a contradiction.

5. ue?) me = (ab’) : (cd’)'’ = (ab’) - (c’d) = (ad)(b’c’) = (ad)(bc)’ =

(e7d) ed’
ad/bc. Give reasons for each step.
6. (a) 3; (d) 4;
(b) 3; (e) 4;
(c) 2; (f) 2.
ioe ho 1 2 3 <| oO i & 2

mipGel 2 3 0} oO O O ©
ieee 3 0 Tt Polos
mie 3 0 1 Ze Oe Oe
Sao 1 2 St |O momo mel
No. 2 does not have a multiplicative inverse.
.. No. There is no solution since 2’ does not exist.
. Write the addition and multiplication tables. They are
Ore a2 a0 ale,

ORO 1 2 OREO 0
ie 2° 0 PS U0 Sol ee
ana) O° 1 2|0 1
Now check the axioms:

O10 Seelt— hte e—2 | Pees

10. The modulus must be a prime, that is, a number with no divisors other
than itself and one.
—><
CHAPTER 8 THE DISTRIBUTIVE
a

I have saved the Distributive Law for a chapter of its own because of
its very great importance. Something like one-half of the manipula-
tions of algebra depend upon this law, and much of the trouble
students have with algebra results from their ignorance of it.
The essential idea behind this law is that it establishes a con-
nection between multiplication and addition. You will observe that
all the other laws deal separately with these two operations, so that
without the distributive law much that is familiar could not be
established.
The intuitive idea behind this law is that for positive integers
multiplication is repeated addition. For example,

3X4=34+34+3+43
=44444.
Notice that this definition of multiplication does not make sense in
other cases such as 4 X 34, for how can we multiply 34 by itself
one-half time? There are similar difficulties with (—2) * (—6), and
76
The Distributive Law 77

these make algebra incomprehensible to a student who has been

assured that multiplication is nothing but repeated addition.
To avoid all these troubles, we rely upon the distributive law,
which states that, for any triple of real numbers a, b, and c,

aX (b+c) =(Xb)4+ (a Xo).

Since multiplication is commutative, an equivalent formulation is

(6 -+c¢) Xa= (6 Xa) + (c Xa).

The law is easily extended to sums with more terms, such as

aX (b+c+d)=(aXb)+(Xc)+
(aX a).

I can give many easy applications of this law. First let me apply
it to positive integers, and use it to show that, for these, multiplica-
tion is in fact repeated addition. I assume that it has already been
established, for instance, that

4=1+14+1¢41.
Then
aha —l| 3% Ue 1 it)
eto, eka Lais
by the distributive law. In other words, 3 times 4 is the sum of four
three’s. Similarly,
Brena.
4X3=4xX[1l+1+41,
=44444,
or 4 times 3 is the sum of three four’s.
To make the situation doubly clear, I remind you that there
are two statements to be considered:
(1) The Distributive Law.
(2) Multiplication is Repeated Addition.
For positive integers these two are equivalent. For various other
types of products, however, (2) is meaningless whereas (1) continues
to make sense. It is, therefore, wise to use (1) in all circumstances
and to de-emphasize (2).
As another illustration of the use of the distributive law,
78 MATHEMATICS FOR PARENTS

consider
2X (4+7) =(22X4+ (2X7)
Sera
= 22.

The usual method of long multiplication is based on this law, for

24 X 56 = 24 xX (50 4 6)
(24 X 50) + (24 X 6)
1200 + 144
1344.

We usually write this in the form

24 24
X56 or X56
“144 “144
1200 120
1344 1344
Less well understood is the use of this law from right to left in
the process of factoring. That is,

(aXb)+(aXxXc) =aX4+0).
In numbers, we can therefore write

39 + 75 (5 X 7) + (5 X 15)
5X (7+ 15)
5 X 22
= 110.

The law is also responsible for a number of short-cut methods

in arithmetic which periodically attract a substantial following.
Thus:
523 X 9 = 523 X (10 — 1)
= 5230/5238
= 4707,
32 X 16= (30 + 2) X 16
= 480 + 32
512.
The Distributive Law 79

The idea is to do most of the steps mentally, and to write down only
the answer.
The distributive law is observed to hold for positive integers,
but there is no way of testing its validity for situations where any
of a, b, or c is zero, negative, or fractional. Since such numbers are
creatures of our own minds we are free to give them any properties
we wish. Because the distributive law is so useful for positive in-
tegers, we assume that it is to be true for all real numbers. This
assumption has several very important consequences.
First, let us prove that for any real number a

a0. 0)

You doubtless believe this already, and wonder what the point of
this is. Remember that zero is defined by its properties of addition
(a + 0 = a) and that nothing was said about multiplication. You
cannot go from addition to multiplication without consulting the
distributive law. So why is a X 0 = 0?
Well:

0+ 0= 0,
aX(0+0)=axXQ0,
(aX0)+(@xX0)=ax0O, (Distributive law)
ax 0 = 0. (Adding —(a X 0) to each side)

This simple proof is typical of the power of this law.

To give another illustration, let us compute the value of
4 X (—3), supposing that no one ever told us how to multiply signed
numbers. Notice that we are now abandoning the earlier notation
~3 and are using the more conventional —3, the former notation
having served its purpose. We start with

3+ (-3) = 0, (Definition of —3)

4X [3 + (-3)]
=4 xX0,
(4 X 3) + [4 X (—3)] = 0, (Distributive law)
12 + [4 X (—3)] = 0.

On the other hand

12 + (-—12) = 0. (Definition of —12)
80 MATHEMATICS FOR PARENTS

. 12-458) = 12 = 12),
or
4% (—3) = —12.
Thus we see that a positive number times a negative number is
negative without having to rely upon some arbitrary rule handed
down by the teacher as something to be memorized or else!
Can you prove now that (—3) X (—4) = 12? I leave the prob-
lem to you, but suggest that you start with

Ala (=A) r=. 0.

Other formulas of algebra can be obtained by using this law
several times over. For example,

(a+ b)? = (a+b)

X (a+ bd)
= [(a + b) Xa] + [a + 5b) X OI
ll (a?+ ba) +, (ab - b?)
ll a? + ba + ab + DB?
= a? + 2ab + b?.
Also .
(a+b) X (ec +d) = ac+ad+be+ bd. (Prove it!)
Thus
18 X 12= (20 — 2) X (10+ 2)
= 200 — 20 + 40 —4
= 240 — 24
= LG.

Some misconceptions regarding division are also susceptible to

therapy by means of this law:

oe = 5(r+ 5)
= (144)(2z) + (4)(5)
=x -+ (54).
Students have been known to think that the correct answer is x + 5;
they have merely cancelled the 2’s.
Using this idea we can see why it is true that an integer is
 The Distributive Law
81

divisible by 3 if the sum of its decimal digits is divisible by 3. An

example makes this clear.

oe = 732 X10?) (ox 10) =F 8]

= [2 X (19%) + 15. xX (4%4)] + 84
= [2 X 83 + 4)] + [5 X 3 + 4))] + 8%
= [2 X 33) + 5 X 8)] + (24 + 54 + 84)
= [(2 X 33) + (5 X 3)] + 20) 3 8:

The answer is thus an integer if and only if 2 + 5 + 8is divisible

by 3, as it is in this case.
In a similar fashion it is possible to prove that an integer is
divisible by 9 if and only if the sum of its digits is divisible by 9. Let
me illustrate such a proof:

#86 = Mel(4 X 10?) + (6 X 10) + 8]

a) Ge ag
= [4 X (11 + }6)] + [6 X (1 + 46)] + 86
cea atl eG xX oto. 5)
= (4 x 1) + 6 x 9) +7 OS
Since 4 + 6 + 8 = 18, which equals 9 X 2, the result is an integer
and 468 is divisible by 9.
‘Less well known is a rule for determining whether an integer is
divisible by 11. From an example let us try to guess the general
situation. Consider

3964, = 144[(3 X 102) + (9 X 10) + 6]

= (3 X 10%1) + @ X 1% 1) + 44
=X @O+ KvI+9 xX @—- Kd + 1
= (8X9) +x H) +-——**.
3—9+6

0 Te
Since ae is an integer (actually it is zero), 396 is divisible
by 11.
What, then, is the general rule? It is ‘An integer is divisible by
eleven if and only if the sum of its number of units, minus its number
82 MATHEMATICS FOR PARENTS

of tens, plus its aumber of hundreds, minus its number of thousands,

etc., is divisible by 11”.
For example, 4829 is divisible by 11, for -4 +8 —2+9= 11,
which is divisible by 11.
The only trouble with the distributive law is that some students
learn it so well that they try to apply it where it does not belong.
Some of them think that

(a+b)? =a? + BD’,

which would imply some kind of distributive law for exponents.

There is no such law, since (2 + 3)? = 25, whereas 2? + 3? = 14.
The only distributive law is the one we have given, and this refers to
addition and multiplication alone!
This concludes the properties of a field. Remember that they
were obtained by intuition from the observed properties of numbers,
and that my only role here has been to organize what you already
knew. This is what I meant in Chapter 1 by the construction of a
mathematical model. Now that you have seen one, I hope that you
will not feel that mathematical abstractions are beyond you.
The next step in algebra is to assume these laws as axioms and
to prove as many interesting theorems as possible. This will be the
program for your high school son or daughter, and if you begin with
these axioms you will see that algebra need not be a stranger to
anyone.

PROBLEMS

1. Apply the distributive law to remove the parentheses in the following

expressions: :
(a) 2(7 + 5) (d) 3(2x + y)
(b) (—3)[4 + (—2)] (e) 2e(z? — 1)
(c) (—4)(5 — 3) (f) 3x(2x? + x — 4)
2. Apply the distributive law (right to left) to “remove a common factor”
from the following expressions:
(a) 2+6 (al) Bink ao
(b) —5 + 10 (e) 22° — 47 +6
(c) 2a + 4y (f) 4a? + 8x
8. Use the distributive law to show that
(a) (—4) X 8 = (—4) + (—4) + (—4).
(b) (=1).*% 5 =(-D + Cl aa.
The Distributive Law 83

4. Use the distributive law to provide a short-cut method of multiplying

the following:
(a) 64 X 9 (Hint: Try 64(10 — 1)) (c) 176 x 99
((e)) e2 < al (d) 256 X 101
. Prove that (—3)(—4) = 12.
. Prove that a X (—b) = —ab.
. Prove that (—a) X (—6b) = ab.
- Which
naan
ant of the following numbers (base ten) are divisible by 9? (a) 7351;
(b) 4257; (c) 54,325.
. Which of the following numbers (base ten) are divisible by 11? (a) 6831;
(b) 5872; (c) 496,397.
10. Prove the following test for divisibility by 7: ‘‘Write the given number
as a numeral to base 7. Then it is divisible by seven if and only if the
last digit in this numeral is a zero.”” Apply this test to 1218ten.

ANSWERS

1. (a) 14+ 10 (d) 6x + 3y

(b) —12+6 (e) 2x3 — 2a
(c) —20 + 12 (f) 6a3 + 32? — 122
2. (a) 2(1 + 3) (d) x(x + 1)
(b) 5(—1 + 2) or (—5)(1 — 2) (e) 2(73 — 2a + 3)
(c) 2(x + 2y) (f) 4a(x + 2)
emee) X3—(—4) X(1+1+ 1) =(—4) + (-4) + (-4);
(6) (—1) X5 = (-1) X (1 +14+1+4+1+)),
Bek are) ee) a1).
. (a) 64(10 — 1) = 640 — 64 = 576.
(b) 52(10 + 1) = 520 + 52 = 572.
(c) 176(100 — 1) = 17600 — 176 = 17,424.
(d) 256(100 + 1) = 25600 + 256 = 25,856.
. As suggested in the text, start with:

4 (—4) = 0,
og) tah et) eas) <0,
ees eral oes) (4) — 0,
pei MO) K—-4)] =O,
ar (14) + i{—3) X(—4)] = 124 0,
(—3) X (—4) = 12.
b+ (—b) =0,
afb + (-—b)]=a X0 =0,
—ab + ab + [a X (—b)] = —ab +0,
a X (—b) = —ab.
84 MATHEMATICS FOR PARENTS

Ne “ a+ (—a) = 0,
la (—@)| X= 2) = 0 (0),
a(—b) + [(—a) X (—d)| = 0.
From Prob. 6, a(—b) = —ab, so
—ab + [(—a) X (—b)] = 0,
ab — ab + [(—a) X (—)b)] = ab +0,
(—a) X (—b) = ab.
8. Only (b).
9. (a) and (c).
10. If a number has a representation to base seven with 0 as the last digit,
it is of the form:

(a X 74) + (6 X 73) + (ce X 72) + (d X 7) +0,

which is certainly divisible by 7. Yes, 1218t.n is divisible by 7, for it

equals 3360,even.
a<
CHAPTER 9 INEQUALITIES
oO

j=
Qo

In spite of the Declaration of Independence and its assertion that

“all men are created equal’’, inequality plays a fundamental role in
our lives. Youngsters take pride in being older than their friends;
oldsters point out that others are older than they are. Businessmen
compare this year’s profits with last year’s, and the study of com-
parative scores is a favorite pastime during football season. Since so
much of our lives is concerned with inequalities, it is surprising that
they have played so minor a role in mathematics instruction. Mod-
ern courses in algebra, however, have raised them to the same level
of importance as equalities. Since inequalities are probably new to
you, I must discuss them from the beginning.
Throughout this chapter we shall understand that all numbers
mentioned are real numbers. From our knowledge of the real num-
bers we are aware that they subdivide into three subsets:

(1) The positive real numbers,

(2) The negative real numbers,
(3) Zero, which is neither positive nor negative.
85
86 MATHEMATICS FOR PARENTS

Moreover, each number belongs in one and only one of these subsets.
This means, for instance, that no number is simultaneously positive
and negative.
This observation permits me to define inequality as follows:
Definition. For two real numbers a and b, the statement a > b (read
“a is greater than b’’) is equivalent to the statement a — b is positive.
Similarly, a < b (read “a is less than b’’) is equivalent to the statement
a — b is negative.
We can interpret this geometrically as follows. Arrange the real
numbers on the number line in the usual way with positive numbers
to the right and negative numbers to the left. Then ‘“‘greater than”
means “‘to the right of” and ‘‘less than” means “‘to the left of”’.
Since a — bis either positive, zero, or negative, we can conclude
that: Given any two real numbers a and b, precisely one of the following
as true:
a> b, a = 6}, a<b.

This statement is called the Law of Trichotomy.

We also recall several other properties of real numbers:
(1) If a and 6 are positive, then a + b is positive.
(2) If a and b are positive, then a X 6 is positive.
(3) If a is positive and b is negative, then a X b is negative.
(4) a is positive if and only if —a is negative.
From property (4) it follows that, if a > b, then b-< a. For, ifa — b
is positive, property (4) tells us that b — a is negative. In the follow-
ing series of theorems we shall, therefore, consider only the ‘‘greater
than” statements. You can easily convert these to “less than’ state-
ments if you find this necessary. Although there are many theorems
about inequalities, I shall give only those that are most common in
applications.

Basic Theorems

Theorem 1. a ts positive if and only if a > 0.

Proof: This is exceedingly simple. For, by definition, a > 0 is equiv-
alent to the statement that a — 0 is positive, or to the statement
that a is positive.
Inequalities 87

Theorem 2. a is negative if and only if a < 0.

The proof is just like that of Theorem 1.
Theorem 3. Ifa>b,thena+c>b+te.
Proof: We want to prove that a +c > b +c. From our definition
of inequality this is equivalent to proving that (a + c) — (6 +c) is
positive. To do this, let us have a look:

(a+c) —(b+c) =a-—Db.

From our hypothesis a > b, that is, a — b is positive. Hence our
proof is completed.
Notice the strategy of this proof. We convert the inequality in
the conclusion into a statement that “something” is positive. We
then examine this ‘‘something” and use our hypothesis and simple
algebra to show that it is, indeed, positive. This method will work
for our other theorems.
Theorem4. [fa >bandc>d,thna+c>b+d.
Proof: Examine the sign of

(a+c)—(b+d) = (@—b)
+ ( — ad).
Since by hypothesis a — 6 and c — d are both positive, [(a + c) —
(b + d)] is positive and the proof is complete.

Theorem 5. Jf a> band cis positive, then ac > be.

Proof: Examine ac — be = c(a — 6). Since both factors are posi-
tive, the product is positive.
Theorem 6. Jf a> band cis negative, then ac < be.
Proof: Examine ac — be = c(a — b). Since c is negative and (a — b)
is positive, the product is negative.
Remark. Theorems 5 and 6 together tell us what happens when
we multiply both sides of an inequality by the same number. If this
multiplier is positive, we leave the inequality sign (>) as it was
originally. But if the multiplier is negative, we must replace > by
< or < by >, that is, we “reverse the inequality.” Failure to
remember this is the most common source of error in handling
inequalities.
88 MATHEMATICS FOR PARENTS

Theorem 7. If a> 0, then (1/a) > 0.

Proof: 1/a = (1/a)? X a. Since (1/a)? and a are both positive, so

is 1/a.
Theorem 8. If a? > b? anda > Oand b > 0, thena > b.
a 2 —
b2
Proof: a—b=- Tips By hypothesis (a? — b?) and (a + b) are
a
positive. Hence (a — b) is positive.
Theorem 9. Jfa>bandb>c, thena>c.
Proof: a—c = (a—b)+(b —c), and both terms on the right
are positive by hypothesis.
This theorem is called the Transitive Law for Inequalities. Notice
that it does not work for football scores!

Applications

Let us now apply these theorems to some typical kinds of prob-

lems. First let us look at the inequality

3a +4
> 1.

It is hardly appropriate to ask a student to ‘“‘solve” this, for a ‘‘solu-

tion’’ implies a fixed answer, and the answer here is of another type.
What we really want is the ‘truth set” or ‘solution set”:

{2\das 44> 1}:

Since this set contains infinitely many numbers, the best thing to do
is to graph it. First, however, we should simplify matters as best we
can. From
8a + 4> 1,
we infer ;
3x > 1—4, (Theorem 4, adding —4
to each side)
or
1 eS
or
>a 1k (Theorem 5, multiplying
both sides by 14)
Inequalities 89

The graph is given in Fig. 10. The open circle above —1 indi-
cates that. —1 is not to be included in the shaded region.

IIL
ee X
=I 0
Fig. 10. Graph of x > —1.

As another illustration, consider

At eee el
This is equivalent to
—2x4 > —4,
or to
22 < 4,
or
Ea 2)

The graph is shown in Fig. 11.

LLL ALLL LLi eo X

0 1 ie
Fig. 11. Graph of x < 2.

It is sometimes desirable to introduce the notations > and <

which mean “greater than or equal to” and “less than or equal to”’
respectively. The graph of x < 2 is given in Fig. 12 where the circle
above 2 is now filled in to indicate that 2 is included in the graph.

BLE ee
0 1 2
Fig. 12. Graph ofx < 2.

Quadratic inequalities are handled in much the same manner.

Let us consider
w—474+2>27—-
4.

This is equivalent to
v?— d54+6> 0,
or to
(x — 2)(a2 — 3) > 0.
90 MATHEMATICS FOR PARENTS

In order to plot the graph of the solution set of this inequality,

I need to explain a further important idea. Let us consider the sign
of a quadratic expression ax? + bx + c as x moves along the X-axis
from left to right. For some values of x the expression may be posi-
tive and for others negative. Suppose it is positive at x = r and
negative at x = s. Then somewhere between 7 and s the value of
ax? + bx + c must be zero. In fact, the only way az? + bx + c can
change sign is for it to pass through zero. This remark suggests that
we examine the values of zx for which it is zero, for these are the places
where something interesting may happen.
In this problem, (x — 2)(x — 3) is zero at x = 2 and x = 38.
These divide the X-axis into three intervals: (—~, 2), (2, 3),
(8, +), as shown in Fig. 13. We have just seen that the quadratic

+ = +
SX
0 1 2 3
Fig. 13. Sign pattern for (x — 2)(a% — 38).

expression x? — 54 + 6 = (a — 2)(a — 3) has a constant sign in

each interval, so we can find this sign by testing suitable values. In
(— , 2), try zr = 0. At z = 0, (a — 2)(a — 3) = 6, so our quad-
ratic is positive in this interval. In (2, 3) try x = 244. At x = 214,
(x — 2)(a — 3) = (4)(—6) = —\%, so our- quadratic is nega-
tive in this interval. Finally, in (8, +), try « = 4. At a = 4,
(« — 2)(2 — 3) = 2, so our quadratic is positive there. The left and
right hand intervals have the required sign, so the graph is as shown
in Fig. 14.

ELLIE LLL ELLA) LILLE gs x

0 1 2 3
Fig. 14. Graph of (x — 2)(x — 3) > O.

This process depended upon our ability to factor the given quad-
ratic. If we are unable to do so by elementary means, we can rely on
the quadratic formula. This says that the roots of

av? ++ ba +c¢=0
Inequalities 91

are
iad —b+ Vb? — 4ac
oa 2a

This gives two values of x which are real if b2 — 4ac > 0, but which
are not real otherwise. Thus when b? — 4ac > 0, the process shown
will solve our problem.
What if, on the other hand, b? — 4ac < 0? Then az? + br + c
is not zero for any real x. Hence it cannot change sign as x moves
from — © to + ~, that is, it is definitely positive for all x or negative
for all x. Let us try an example:
Find the solution set of 22 +2+1> 0. Since b? — 4ac =
1 — 4 = —8, we know that x? + x + 1 is either positive or nega-
tive for all x. To find out which it is, try x = 0. Atz = 0,22 +a+
1 = 1, so x? + 2+ 1is positive for all x. The solution set is the
entire X-axis.
If, however, our problem had been to find the solution set of
x* + x +1 < 0, our answer would have been ‘‘The empty set.’
A frequent use of inequalities involves the notion of absolute
value. By the absolute value of a real number we mean the distance
of the number from 0, taken positively along the number line. The
symbol |a| is used to denote the absolute value of a. For example

|3| = 3, |—4| = 4, |0| =U;

Now what do we mean by stating that |x| < 4? This says that
the distance of x from 0is less than 4, or that x is between —4 and
+4. A convenient notation for this is

—4<24<4.
Similarly,
lx + 2| <6
is equivalent to
= Octet 2a 0
or to
—8 <2 < 4.
Finally,
lx —a| <b
is equivalent to
—b<x-—a<b,
92 MATHEMATICS FOR PARENTS

or
a—b<2z2<a-+b.

Sometimes the inequality goes the other way, so that we are

interested in
|x| > 4.

This means that x > 4 or x < —4.

There are many famous inequalities which occur in higher math-
ematics. I close this chapter with just one of these. You doubtless
have dealt with the arithmetic mean (AM) or average of n positive
numbers:
Git 0) a eerie
AM =
n

Perhaps you also recognize the geometric mean (GM):

GM = VarXa,X
°°Xa
The remarkable inequality to which I refer is that

GM < AM

in all cases. I shall prove this for n = 2. In this case

GM = Vam, AM = a,
From simple algebra it follows that

(ai ar a2)? - (a1 ra 2)”

Qa\a, =
4 ° 4

Since (a; — a2)? > 0, we see that

(a1 = a»)?
G10. < A
or that

Var, < HE.

If you do not believe this, try a few examples. For a; = 2,
Inequalities 93

a2 = 8, we have
——— 2+8
+
oS Sao CLASS:

For a; = 5, a2 = 5, we have
5+5
V5xX5< ; or aon

It never fails!
Further we can ask: When does equality hold in the statement
GM < AM? In our proof above for n = 2, it is clear that we obtain
inequality if (a1 — a2)? > 0 and equality if (a; — a2)? = 0. Hence
equality holds if and only if a; = a2. The conclusion is still valid for
the general case, that is, equality holds if and only if a1, a2. . . » An
are all equal.

PROBLEMS

1. Fill in the correct inequality symbols (> or <):

(a) 5 (—3) (e) (—12) 0
()"(=10) Co (f) 13 0
(ec) (-8) 8 (2) 13) (=7)
(d) 7 (—15) (h) 48 — 10 10 — 48
2. Fill in the correct inequality symbols (>.or< ):
. (a) Since 8 > 5, then 2 X8 2 Oo:
(b) Since 2 < 7, then 3 X 2 Se
(c) 4X6 4 X 10.
(d) 4 X (—8) ____4 & (— 16).
(e) Since 6 > 2, then (—3) X 6 (=) X 2.
(f) Since —3 < 5, then (—4) X (—8) (—4) X 5.
(zg) 2 x 12 2s
in) (3) X15 ee (= 3) 7.
(i) 4 X (—8) 4 X (—5).
ye —5) X (—9) (—5) x 0.
3. Fill in the correct inequality symbols (> or <):
(a) Since (—6) < 10, then —6 + 3___ 104 3.
(b) Since (—11) < (—5), then -—11 4+ 15s —5 4-15.
(c) 17 —6 2-6.
(d) (—2) +3 Aes.
(e) Since 12 > 5 and 5 > 1, then 12 18
(f) Since 3 < 6 and 6 < 10, then 3 10.
94 MATHEMATICS FOR PARENTS

(g) Since 7 > 3, then 14 _

(h) Since 2 < 11, then 14 144.
(i) Since 16 > 9, then 4 ___ 3.
(j) Since (—4)? > (—3)?, then (—4) (—8).
. Simplify and plot the graph of the solution set:
(a) 8% > 16. (f) 22 +4 <5.
(b) —3z < 6. (g) 8a —12> 5.
(c) 4% > 2. (h) he — 2 =< 30 4 10:
(d) —5z > 15. i) =e -—- 4 = 22 — 5.
(e) x +2> 6. (j) 3a — 2 < 4% — 5.
. Write in the form a < x < bora <x < band plot the graph of the
solution set.
(a) rie 2. (f) |22 — 5| < 8.
(b) |x| < 5. (g) |8¢ + 6| < 9.
(c) ja —1| <3. (h) |—a2 + 2| < 7.
(d) |x + 2| < 4. (i) |—2r4
— 1| <3.
(e) |22| < 6.
. Write in the form: z > a or z < b, and plot the solution set.
(a) |z| > 3. (d) |22 + 1| > 5.
(b) |x| > 5. (e) |—3r + 6| > 9.
(c) |2x| > 4.
. Express as an inequality involving absolute values:
(a) The distance of 2 from 4is less than 6.
(b) The distance of x from 2 is greater than 5.
(c) The distance of x from 2 is five times its distance from 6.
(d) a is farther from 3 than it is from —4.
(e) The sum of the distances of x from 1 and from 3 is greater than 10.
. Plot the solution sets of:
(a) (a + 2)(@ — 4) > 0.
(b) a? — 7x + 12 < 0.
(c) 22 +42 —52>0.
(d) a +2+2>0.
(e) («© — 1)(a + 2)(a — 3) > 0.
. Verify that GM < AM when:
©) Gh = Bh Os = Pl
(b) ay = 2; ag = 5, a3 = 100.
(c) a; = 4, de = 4, a3 = 4.
(d) a, = 0, a2 = 5, as = 2.

10. Prove: If a > 6, and if a and bare positive, then a? > 6’.
11. Show that, among all rectangles with a fixed perimeter, it is the square
which has maximum area.
Inequalities 95

ANSWERS

ee (a) (e) <.

(b) Gi) Sy
(c) (2) t=:
(d) he

2. (a) (ie
(b) ee.
(c) (th)
(d) Gy) <<a
(e) Gia:
3. (a) (Ha
(b) (g) =.
(c) (h) >.
(d) Ga.
(e) VAVYAA
VAAY
AVYAAY Ca
408) eS 2. () << %.
2777777777777
ee
0 2 01/2

(b) 2 > —2. (g) 2 = 2.

—_ILZ4ALZ
LLL 7777777777» ee
Lefe
, -2 0 0 2

(c) x > le. (h)esaiGy

(222227777777
4 777Pw _LLZZZZZZZZAL
LYLADD PRL
01p 0 6

(d) x < —3. Ga <h

ih
ee
(ey yl

ary
96 MATHEMATICS FOR PARENTS

i, (G)) 2 << (f) —3§ <a < 1%

—<———
AZZ
+ i) -72 0 13/

(os) =<
we<< &. (g) —5 <2 <1.

___FPFAAERA EAL
ey 0 5 =5 Ome

(c) —3 <2 —1 < 3or,better, (h) —-5<2< 9.

a
eens
WINDS CE CIMA Was
__
PATA© 0
-5 9
-2 0 4
Gi) SA sx PS il
G)) Sa <q a s<

___ 5}AAG 0 2
a es

(e) —-8 <2 <3.

3 0 3

(3 @) oS Sore
<< =e (d) cx>2orz
< —3.
CZZZ7ZAD | | 777777 L777 77ZRY || CV
=} 0 3 et 0 Zz

(b) > 5 ora < —5. (@) 2 Share

& oll.

L777 py 4 | | Vs PEP A,
=o 0 5 150 5

(Cia 2 ora —2:

Inequalities 97

ls (a) |x
— 4| < 6. (d) |x — 3] > |r + 4].
(b) jx — 2| > 5. (e) |x — 1] + |x — 3] > 10.
(c) jz — 2| = dla — 6].
. (a) (d) The empty set.
L277 77AY 4 V7 77773,
20 4 (e)

(b) ay) Wengee =

a ga
0 3 4

(c)

a
-5 01

3 + 27
may \/3 X 27 < cae (8) << 115),
2
2+5 + 100
(b) V2
5 X 100 < or 10 < 35%.

(c) Vaxaxa a tite, r4

= 4,

id) 0X5 X2< OP.b er——or0 i.

10. Given: a—b is positive; a is positive; b is positive. Consider:

a* — b? = (a + b)(a — b). By hypothesis each factor on the right is
positive. Therefore, the product is positive. Hence a? > 6?.
TY. Let x and y be the length and width, respectively, of a rectangle. Then
the area A is xy and the perimeter p is 2x + 2y. From the fact that
GM < AM we conclude that

Vay <==!
or that

Mee een ee \ st(i

Thr iia 4
Therefore, the area is a maximum when equality holds in the above
statement, that is, when x = y, so that the rectangle is a square.
oe
CHAPTER IO INEQUALITIES IN
—— THE PLANE

Important applications of inequalities arise from the consideration

of inequalities for functions of two or more variables. For the sake
of simplicity we shall restrict ourselves here to linear functions of
two variables.
Graphs of Linear Inequalities
You are already familiar with functions of this type, for we have
seen in Chapter 6 that equations like 3x — 4y + 12 = 0 represent
lines in the X-Y plane. The expression f(z, y) = 3x — 4y + 12 isan
example of linear function of two variables. Let me recall how we
plot the graph of 3x — 4y + 12 = 0. First we draw the X and Y
axes as in Fig. 15. With reference to these we plot the position of
points with coordinates (zx, y). For example, the point (3, 1) is three
units along the X-axis and one unit up the Y-axis. The graph of
3x — 4y + 12 = 0 consists of those points whose coordinates (x, y)
satisfy this equation, and is a straight line. We can plot it by first
calculating two of its points. In our example, (—4, 0) and (0, 3)
satisfy the equation, so we draw the line as in Fig. 16.
98
Inequalities in the Plane 99

This gives us the points where 3x — 4y + 12 = 0, but where

are those where 3x — 4y + 12 > 0 or where 3x — 4y + 12 < 0?
Let us examine a few points below the line. At O (0, 0), 83a — 4y +
12=12>0. At (2, 1), 3 -—4y+12=6-4+12=1450,
and, at (4, 4), 32 — 4y + 12 = 12-16 +12 =8>0. Indeed, it
is true that, at any point below the line, 3x — 4y + 12 > O. Similar

Fig. 15. Coordinate axes.

calculations show that for points above the line 3x — 4y + 12 < 0.

You must not jump at generalizations here, for the situation may be
reversed for other functions of this type. For example, —2x + 3y —
6 > 0 for points above the line and —2z + 3y — 6 < 0 for points
below the line (see Fig. 17).
We can, however, make the following true statement: If we
draw the line which is the graph of ax + by + c = 0, this line
100 MATHEMATICS FOR PARENTS

divides the plane into two half-planes. In one of these ax + by +

c > 0, and in the other ox + by +c < 0. We can tell which half-
plane is which by testing the inequality for a single point not on the
line.
PROBLEM. Sketch the half-plane in which 2x + 5y — 10 > 0. (See
Fig. 18.) First we find that (0, 2) and (5, 0) are on the line, and plot

Graph of 3x
— 4y + 12 = 0

Figure 16

them. Join them with the line. Now test any point not on the line;
we choose O because the arithmetic is easiest for this point. At O,
2x + 5y —10 = 0+ 0-— 10 < 0. Hence the half-plane below the
line is the wrong one for our problem, and so the upper half-plane
gives us the solution. Just to be sure we check that at (3, 3), 2e+
5y — 10 = 64+ 15 — 10 = 11 > 0. We now shade the upper half-
plane.
 Inequalities in the Plane 101

We can complicate the problem by seeking the set of pairs

(x, y) which are solutions of both the inequalities:
2x + 5y — 10 > 0 and oi ale
We already have discussed the first of these, so let us turn to the
second. The line x — y = 1 passes through the points (1, 0) and

Graph
of : —-2x + 3y-6=0

Figure 17

(0, —1). At O, x -— y=0-—0=0 <1, so the correct half-plane

is that which includes the origin. We now plot both lines on the same
axes and shade the region which belongs to both selected half-planes.
(See Fig. 19.) We can continue this process for as many lines as we
please.
This technique of solving systems of inequalities is the first step
toward solving many problems of business and industry, and is re-
lated to a subject known as Linear Programing.
102 MATHEMATICS FOR PARENTS

The Basketball Problem

Let us illustrate by considering an imaginary situation in a

basketball game. The home team is behind by 3 points and there are
1014 minutes left to play. The home team can put on either a slow-

LZ e (3,3)

Graph of: 2x + 5y—10>0

Figure 18

moving pattern offense or a fast, hard-running offense. On the aver-

age the following figures are assumed to be accurate:
Offense Time/Play Shooting Accuracy
For THE HOME TEAM
Pattern 30 seconds 2 out of 3
Running 15 seconds 1 out of 3
For THE OPPONENTS, WHO ALWAYS RUN
Running 15 seconds 1 out of 2
Inequalities in the Plane 103

The home team naturally wishes to win, but also thinks it wise
to mix up its offenses as much as possible. What possibilities are
open to them?
Let p be the number of pattern plays the home team uses and
r the number of running plays, so that the home team has a total of
p + r plays. Assume that after each shot the ball changes hands, so
Y
2x + 5y-—-10>0
x-y<l

Figure 19

that the opponents will also have p + r plays. We must assume, also,
that there are no penalties, and no miscues which cause the ball to
change hands except after a shot.
Then we look at the time required. We see that
30p + 15r + 15(p + r) < (1014)(60) = 630 (seconds)
Dividing by 15 we simplify this to
ip ir) <42
ap ina
104 MATHEMATICS FOR PARENTS

or
3p + 2r < 42.
With regard to the score, if the home team is to win we must have
Dexa piesa Pat eae
This simplifies to
p—T> ¥.
Our system of inequalities is therefore:
3p + 2r < 42, Deza0:
p—r>9, r= 0:
Their graph is as shown in Fig. 20. The two lines intersect at
(3, 12), and the set which satisfies all the inequalities is shaded in

3p + 2r < 42
(Sh 4)) p=1r>9
p20
Te)

Figure 20

the figure. The home team can then play any combination (r, p) of
offenses which lies in the shaded region. The possibilities are:
r 0 1 py

Dp 9 to 14 11 to 13 12
 Inequalities in the Plane 105

In other words, the home team can afford to use at most two
running plays. Perhaps they would do better to use the pattern
offense for the whole time, but the team will have to decide this on
the basis of other information than that stated in this problem.
For a true problem in Linear Programing we should put the
question somewhat differently. Suppose we had said: Given 1014
minutes to play and the times and accuracies stated above, how

Pp

3p + 2r < 42

Figure 21

should the running and pattern plays be distributed so that the home
team would have the largest net gain in score?
The mathematical formulation then becomes: For what values
of p and r is the function (the net gain in score) p — r a maximum
when p and r are subject to the conditions
3p + 2r < 42, He) mage eeOt
The graph of these simultaneous inequalities is shown in Fig. 21. A
theorem says that this maximum will be achieved at one of the
vertices of this triangle. Let us check them.
AL (21, 0);
p—?r=0 —2h= —21,
106 MATHEMATICS FOR PARENTS

At (0, 0),
p—r=0.
At (0, 14),
p—r= 14.

The conclusion is that, for the largest net score, all plays should be
pattern plays. This conclusion was probably obvious to you without
all these calculations, but that is inevitable if I give you such a
simple problem. The point is that methods of this kind can solve
maximization problems of much greater difficulty for which the solu-
tions are far from obvious. Let me illustrate the method by a second
example.

The Diet Problem

This illustration is a simplified version of one of the original
problems that led to the invention of Linear Programing. Let us
suppose that we have two types of synthetic foods, which I call A
and B. You may think of K rations and C rations if you like, but the
numbers given below are purely artificial and apply to no foods on
the market. Let us suppose that the two foods contain the following
nutritional components:

Calories Protein Fat

Food | per Ounce | per Ounce | per Ounce

A 100 50 0

B 200 10 ve 30

The units for protein and fat are arbitrary and need not be specified.
Let us also suppose that the minimum daily requirements for an
active man are:
Calories: 2500.
Protein: 350.
Fat: 150.

There are several different questions that can now be asked. As a

starter consider our first problem:
 Inequalities in the Plane 107

PROBLEM 1. Which food or combination of foods should be em-

ployed in order: (a) to fulfill the minimum daily nutritional require-
ments, and (b) to minimize the total weight?
Such a problem would be of importance to a mountain climber,
or to a military expedition. In order to attack the problem I must
introduce some notation.
Let a represent the number of ounces of food A that are required,
b represent the number of ounces of food B that are required.
Then the minimum daily requirements will be met if:
Calories: 100a + 200b > 2500.
Protein: 50a + 10b> 350.
Fat: 30b > 150.
These three inequalities can be simplified to the following simul-
taneous system:
a+ 2b > 25,
5a + b > 35,
bo:
To these we should also add the practical requirements that a 210
and 6 > 0, for one cannot eat a negative amount of food. The graph
of these inequalities is given in Fig. 22. In this figure the shaded
region represents the possible pairs (a, b) which meet the minimum
daily requirements.
. Now I must turn to the quantity which is to be minimized,
namely, the total weight, W = a + b. The basic theorem on Linear
Programing says that W will be a minimum at one of the vertices
labeled P, Q, R in the figure. In order to determine which vertex to
use, let us do some arithmetic.
At P,
a = 0, b = 35, W =a+b
= 35.
At Q,
a= 5, b = 10, W=a+b
= 15.
At R,
a 15, epblsconen =a
+b = 20,
Therefore, W is a minimum at Q, and the diet should consist of 5
ounces of food A and 10 ounces of food B.
For variety we can change the problem to consider feeding the
consumer for the least cost, as in Problem 2.
108 MATHEMATICS FOR PARENTS

PROBLEM 2. Let us assume all the above information plus the fact
that food A costs 10 cents per ounce and food B costs 20 cents per
ounce, Then what combination of A and B should be used to provide
an adequate diet at least cost?

IE
(0,35)

(0, 12/2)

pa
maeC (7,0) (25,0)
a

a+ 2b225
Graph of Sa + eae
a2z0

Figure 22

The solution is just as given above until we come to select one

of the points P, Q, or R. Now we consider the following:
Cost = 10a + 20b.
Ate,
a = 0, b = 35, cost = 700 cents.
At Q,
a = 5, b = 10, cost 250 cents.
 Inequalities in the Plane
109

At R,
a= 15, be=s5: cost = 250 cents.
Thus Q and R are equally good solutions, and in fact any point
on
the segment QR is also a solution.

PROBLEMS

1. Plot the graphs of the solution sets of the inequalities:

(a) <+3y+6>0.
(b) 2a —y+4>0.
() @+ty+1)?>0.
2. Plot the graph of the solution set of the simultaneous
inequalities:
2t — yal
> 0)
3x + 4y —7>0.
3. In the diet problem suppose that the consumers dislike food
A six times
as intensely as they dislike food B. Then what combination
of A and B
should be used to minimize the amount of griping?
4. In the diet problem suppose that 1 ounce of food A occupies
2 cubic
inches and that 1 ounce of food B occupies 3 cubic inches. Then
what
combination of A and B should be used to minimize the total
volume?
Such a decision might be important to an astronaut.
5. The situation in a simplified version of a football game
is as follows.
There are just two plays, a running play and a pass play. We
assume
these facts to be true:
Distance Time
Play Gained, yards | Required, seconds

Running 3 30

Pass 12 9
Also suppose that there are 36 yards to go for a touchdown and that
138 seconds remain in the game. Ignore the requirement of having
to
make 10 yards in four downs and other considerations of score
and
strategy. Problem: What combinations of running and pass plays
will
secure a touchdown in the allotted time?
6. In the football problem (Prob. 5 above) suppose that on
the average
there is one injury in each five running plays and one injury
in each
ten pass plays. Then what combination of plays should the quarter-
back call to secure the touchdown in the allotted time with the minimu
m
risk of injuries?
 MATHEMATICS FOR PARENTS
110

into
7. The Minneapolis and Seattle Lumber Company can convert logs
a given week the mill can turn out 500 units
either lumber or pulp. In
are
of production, of which 100 units of lumber and 200 units of pulp
and
required by regular customers. The profit on a unit of lumber is $15
on a unit of pulp is $18. Problem: How many units of lumber and pulp
should the mill produce (totaling 500) in order to maximize the total
profit?

ANSWERS

1. (a) (b)
he <

(c) The entire plane.

( ih 0)
Inequalities in the Plane 11

3. Let D be the measure of dislike.

At P,a = 0; b = 35, D = 6a + b = 35.
At Q, a = 5,b = 10, D = 30 + 10 = 40.
At R,a = 15,6 =5, D = 904+ 5 = 95.
Answer: Use P where a = 0, b = 35.
4. Let V = volume.
At P, a = 0, b = 35, V = 2a + 3b = 105 cubic inches.
At Q, a = 5, b = 10, V = 40 cubic inches.
At R, a = 15, b = 5, V = 45 cubic inches.
Answer: Use Q where a = 5, b = 10.
5. The inequalities are: Distance: 3r + 12p > 36. Time: 30r + 9p < 138.
Also p > 0, r > 0. These simplify to

r+ 4p > 12,
10r + 3p < 46,
p = 0,
i Uy
The graph is shown below. The shaded triangle is the solution set. The
P

C (0, 15Y3)

A (0, 3)

(4.6, 0)

pairs of integers corresponding to pairs in this triangle are:

r 0 1 2 3 4

p 3 to 15 3 to 12 3 to 8 3 to 5 2
From among these possibilities the quarterback can choose his plays
according to his judgment of appropriate strategy.
112 MATHEMATICS FOR PARENTS

6. Let J = injuries. Then J = +=,0

ols

At A,r =0, p = 3; I = 340.

At 6, r = 0, p = 15%; I = 4% = 1.538.
I is minimized at A, so the quarterback calls three pass plays and no
running plays.
7. Let L = units of lumber, P = units of pulp.
The conditions are:
L+P = 500, TOO ee 200:

Profit = 15L + 18P. Draw the graph. From the two inequalities the
mill must operate in the shaded portion of the plane. Moreover, it must
operate along the diagonal line. Hence its operations must lie on the
segment AB. "

At A, L = 100, P = 400; profit 15L.+ 18P = 8700.

At B, L = 300, P = 200; profit 15L + 18P = 8100.
Hence the mill should operate at A and produce 100 units of lumber and
400 units of pulp.
 ><
CHAPTER I] GEOMETRY
oO

JS
Oo

The traditional program in geometry included Euclidean plane geom-

etry in the tenth grade and solid geometry in the twelfth grade.
There was some attempt to introduce simple geometrical ideas in
the elementary school, and formulas for areas and volume were
often given in the eighth grade. In general, however, geometry was
put off till the tenth grade, and very few pupils reached solid geom-
etry at all.
For centuries geometry has been considered to be one of the
most important subjects in the education of an intelligent man, and
the course goes back with remarkably little change to the days of
Euclid in 300 s.c. It is no exaggeration to say that every educated
man in the world who lived after the time of Euclid studied geometry
much as it is taught today. What is it about geometry that has
caused it to last so long and to be so popular with men of all ages
and of all countries? Of course, it has immense practical value, for
one must know something about geometry to deal with the shapes
of physical objects, and the craftsman, the mechanic, and the artisan
must be highly skilled in geometry to ply their trades. I believe,

113
114 MATHEMATICS FOR PARENTS

however, that the real reason is deeper. Euclid’s great accomplish-

ment was to construct the first formal mathematical model of nature.
He presented the first systematic treatment of axioms and initiated
the procedure of deducing conclusions from premises. Geometry is
thus the prototype of a major trend in civilized thought, and every
civilized man needs to know it if he is to understand other aspects
of his surroundings. Geometry has been, therefore, a liberal art in
the true sense, and not merely a tool for science and technology.
There are those who have tried to reduce it to such a tool, but they
have missed the sweep of history as it applies to geometry.
When the revolution in mathematics approached the subject of
geometry, there was much less agreement than in other areas of
school mathematics. There are those who argue for drastic curtail-
ment of instruction in geometry on the grounds (1) that its place as
a prime example of the axiomatic method has been usurped by
algebra in its modern axiomatic form, and (2) that the time spent
on geometry could be more profitably devoted to more algebra and
even calculus. Others favor an overhauling of Euclid’s geometry to
take account of modern scholarship, and still others argue for the
replacement of much Euclidean geometry by analytic geometry.
The issue is far from settled, and so almost. anything is likely to be
happening in your school. Let me describe for you some of the ideas
that are in the air.

What’s Wrong with Euclid?

The first topic concerns the modern revision of Euclid’s Ele-
ments. Euclid claimed to have provided aset of axioms from which
all of the theorems of geometry could be proved by logical means.
His work was so skillful that no one found a flaw until about one
hundred years ago, some 2200 years after Euclid. There are very few
men whose work has stood up so well! But a serious flaw was found,
and it is now proposed that this be remedied in our school teaching
of geometry.
Euclid warned us to reason from the axioms and not to argue
from a figure, but in one important place he failed to follow his own
warning. Apparently he never noticed that it is necessary to be
careful about the order of points on a line. Let us illustrate the prob-
lem by attempting to define a line segment. If we are given two points,
Geometry 115

A and B, on aline (see Fig. 23), we wish to speak of the segment AB.
The natural definition of this segment is that it consists of the
points A and B and all points on the line between them. The trouble
is that Euclid’s axioms give us no grounds for deciding what between
should mean. Of course, we know what we mean from the figure, but
this is not enough—it must be in the axioms. By straightening out

Se
A B
Figure 23

this notion of betweenness, mathematicians of the last century cleared

up many obscurities in Euclid. It is strange that writers of school
textbooks have ignored this development for some seventy-five
years!
There is a similar problem in Euclid regarding points in a plane.
We surely know what we mean bya triangle, but what does it mean

Figure 24

to say that a point P is inside a triangle? In Fig. 24 this is clear, but

how else does one describe this fact? Of the several alternative
approaches, perhaps the simplest is to add the following axiom:
Axiom. Every line divides the plane into three subsets which
have no point in common:
(1) The line itself.
(2) The set I (intuitively the points on one side of the line).
116 MATHEMATICS FOR PARENTS

(3) The set II (the points on the other side of the line) such that
(See Fig. 25a and b):
(a) A segment PQ lies in I (or II) if and only if the points P and
Q are in region I (or II).

(a) (b)
Fig. 25. Division of a plane byaline.

(b) A segment PQ intersects the given line if and only if its

end points P and Q are in opposite regions, I and II, or II and I.

Fig. 26. Angle defined by two rays.

This may seem trivial to you, and it certainly is obvious. But

rigorous geometry cannot be done without this axiom or one equiv-
alent to it. Let me show you how this is applied in textbooks now
written for seventh-grade students.
Geometry 117

An angle is defined to be a pair of rays (or half-lines) each start-

ing at a point O and extending indefinitely in one direction. (See
Fig. 26.) We pick points A and B respectively on the rays, and speak
of the angle AOB. Now we must define the interior of this angle.
(See Fig. 27.) The line of which OA is a part divides the plane into
two half-planes. In one of these we find the point B. Let us call this
the B-side of OA. Similarly, we can speak of the A-side of OB.
Finally, we define the interior of angle AOB to be the set of points
which are in both the B-side of OA and the A-side of OB. We have

> aq

Fig. 27. Interior of angle AOB.

previously called this the intersection of those two half-planes. The

shaded region pictures this set. From this we can proceed to define
the interior of the triangle ABC. This is the intersection of the
interiors of the three vertex angles of the triangle.
But, you ask, why do we need to define this, for was it not
obvious before? Yes it was, but we had no way of proving anything
about this interior. If you re-examine your old geometry book, you
will find many proofs that tacitly suppose that certain points are
inside a given triangle but which never prove that this is the case.
Using the above method we can now construct such a proof. I agree
that this makes many proofs long and tedious and do not recommend
118 MATHEMATICS FOR PARENTS

such proofs in every case. We should, however, understand what is

involved and be able to fill in the details if necessary.

Analytic Geometry

The methods of Euclidean geometry are far from automatic.

They often require that certain lines be constructed (for no obvious
reason) and that great ingenuity be used. By transferring our prob-
lems to analytic geometry we can work with numbers and algebra
and hence get solutions with less mental effort. Since analytic geom-
etry is used widely in calculus and in the applications of mathe-
matics to physics and engineering, it is recommended that it be
introduced at an early stage in the high school curriculum. Clearly
this introduction would leave less time for the standard deductive
geometry, and so there is a conflict of interests. How do we weigh
the cultural and imaginative aspects of deductive geometry against
the power and applicability of analytic geometry? This is an old
battle and it is not likely to be settled easily.

Intuitive Geometry

In Chapter 1 I described how important it is to begin any math-

ematical subject with intuition before proceeding to its formal, sys-
tematic treatment. This precept applies with special force to
geometry. It is madness to begin geometry in the tenth grade in an
axiomatic, deductive fashion unless a foundation of intuition is built
first. If we have to build this foundation in the tenth grade, there is
hardly time left to do a proper job with the deductive material.
Moreover, the tenth grade is too late for students to begin their
geometrical studies; they need the information earlier.
In view of this situation the modern textbooks introduce geom-
etry in the seventh grade or earlier. The emphasis here is on the
ideas rather than on the proofs. Applications are made to daily
affairs such as explaining why a three-legged stool sits firmly on the
floor, whereas a four-legged chair may wobble. This early teaching
of geometry is an excellent scheme, which has the approval of every-
one involved.
Geometry 119

Solid Geometry
It is a paradox that in the space age the teaching of the geom-
etry of space has gone to pieces. But so it is, and the fault is due to
both neglect and misunderstanding. Everyone agreed that the old
twelfth-grade course in solid geometry had failed to meet its objec-
tives, but there was no agreement on the remedy. Students have a
serious need for an understanding of space relationships, for knowl-
edge of the facts about lines and planes in space, and for information
about the geometry of the sphere (for is our globe not a sphere?).
The deductive method of approaching these ideas was too compli-
cated and tiresome, and seemed to bog down in details while missing
the main ideas. For this reason it was recommended that the course
be abandoned, and that the material on space geometry be taught
intuitively in the junior high school and in the tenth grade. This has
been partially accomplished, but much more must be done. Too
often space geometry is taught nowhere in the school, and our chil-
dren are the worse for our neglect.

A Recommendation
My own preference for the geometry curriculum is the fol-
lowing:
Junior High. Intuitive geometry in two and three dimensions.
Emphasize ideas; no proof.
Tenth Grade. Modernized deductive geometry of the plane.
Ideas of solid geometry introduced where appropriate. A few weeks
only of analytic geometry.
Twelfth Grade. One semester of plane and solid analytic geom-
etry. A few weeks on the geometry of the sphere.
If this program is followed in addition to a proper curriculum
in algebra and trigonometry, the student will be well prepared for
his calculus course when he goes to college. It is very foolish to teach
calculus to anyone who has not first completed such a curriculum.

A Theorem on Intersections
Let us now get down to some details. First I shall show you an
application of the axiom on the division of the plane which I stated
earlier in the chapter.
120 MATHEMATICS FOR PARENTS

Theorem 1. Gwen: A triangle ABC and a line PQ (see Fig. 28)

which intersects segment BC but which does not pass through any of the
vertices A, B, C.
Conclusion: Line PQ intersects either segment AB or segment AC.
Proof:
1. Let us suppose that line PQ does not intersect segment AB. Then
we must prove that it does intersect segment AC.
2. Line PQ intersects segment BC (hypothesis).
3. B and C are on opposite sides of line PQ (axiom above).

B Q

P
Figure 28

4. Line PQ does not intersect segment AB (assumed).

5. A and B lie on the same side of line PQ (axiom above).
6. A and C are on opposite sides of line PQ (steps 3 and 5).
7. Line PQ intersects segment AC (axiom above).

Inside and Outside

The matter of defining the inside of a triangle was simple

enough, but it is harder with other figures. The generalization of the
definition of the inside of a triangle, for instance, does not work for a
polygon such as that in Fig. 29. Moreover, how can you determine
in a simple way whether point P in Fig. 30 is inside or outside the
polygon of the figure? A good method is to draw a ray PQ not through
 Geometry 121

any of the vertices and to count the number of sides which it crosses.
Regardless of how you draw PQ this number will be even if P is
outside and odd ifP is inside. Try a few positions for PQ, and check
this for yourself. Of course, this result requires proof, but since the
proof is complicated I shall leave it out. The idea, however, is most
important. Did you ever run across it when you were in school?

Figure 29

Euler’s Theorem

Finally I shall turn to a result in solid geometry which has great

intuitive appeal, but which disappeared from school geometries sev-
eral generations ago.
First consider a triangular pyramid such as that in F ig. 31. This
is called a tetrahedron, a word derived from the Greek tetra, which
means four. Let us count the number of vertices, edges, and faces.
We get: V (vertices) = 4, E (edges) = 6, F (faces) = 4. Now com-
pute: N =V—H+F =4—6+4 = 2. For late use we record
r
this:

Tetrahedron: WN = 2.
122 MATHEMATICS FOR PARENTS

Figure 30

Next do the same for a cube (see Fig. 32): We find that V = 8,
E = 12, F = 6,sothatN= V —EH+ F= 2.
Cube: WN = 2.
Finally let us do it for an octahedron (Fig. 33). We find that
V = 6, FE= 12, F = 8,sothatN= V —E+F = 2.
Octahedron: WN = 2.
Perhaps this is getting monotonous; we seem to get N = 2 no
matter what solid figure we consider. As I shall show you, this is not
necessarily the case, so we need to state some property of a solid
Geometry 123

Tetrahedron

Figure 31

Cube
Figure 32
124 MATHEMATICS FOR PARENTS

figure which will imply that N = 2. What, then, do the tetrahedron,

cube, and octahedron have in common? Let us suppose that they
are made of wire and covered with a thin rubber surface like that of a
balloon. Now blow air into the balloon—eventually the rubber surface
will become that of a sphere. That is, the surface of each of these solids
can be blown up into the surface of a sphere. The result before us is
due to the Swiss mathematician Euler (1707-1782) and is as follows:
Theorem 2. Jf the surface of a solid figure can be blown up into the
surface of a sphere, then N = V —E+ F = 2.

Octahedron

Figure 33

Again, I must omit the proof, but let us try some experiments.
First take any spherical object (say an orange or a billiard ball) and
draw a network of curves on it. Mark all the intersections and call
these ‘vertices’. Between each pair of vertices there should be one
and only one curved segment—call this an “‘edge’’. Call the open
spaces ‘‘faces’”. Now compute V — EF + F; if you have followed the
instructions this number should be N = 2. Thus you have verified
Euler’s Theorem.
As the next experiment I suggest that you do the same thing on
the surface of a doughnut. I know that it is hard to draw on a
 Geometry 125

doughnut, so I suggest that you try it on a rubber ring such as is

used in deck tennis or quoits. Each “face” must be polygonal—ring-
shaped regions do not fit the requirements. If you do this carefully
you will find that N = 0. Why not N = 2? The answer is that there
is no way of blowing up the surface of a doughnut into the surface
of a sphere. Further experiments of this kind are suggested in the
problems.

Regular Solids
Let me return to Euler’s Theorem as it applies to solids whose
surface can be blown up into the surface of a sphere. Remember that
in this case
ae ieee
A solid is called regular if and only if each face has the same number
of edges, and if each vertex is met by the same number of edges.
The question now is: What regular solids exist?
Let n represent the number of edges to a face and m represent
the number of edges meeting at a vertex. Then the three solids
mentioned above are regular with the following values of n and m:
m n

Solid (edges per vertex) (edges per face)

Tetrahedron 3 3
Cube 3 4
Octahedron 4 3

Since there are F faces, the number of edges can be counted by

considering the product nF. This counts each edge twice, so we have
the relation
ve = 2h or Fe= 2H in:
In a similar way we can find that
mV = 2E, or V = 2E/m.
Now substitute these expressions for F and Vin V — E+ F = 2.
The result is
28
—— E+ 2Hn =2,
m
126 MATHEMATICS FOR PARENTS

which can be rewritten in the form

The problem now is to find all possible values of n, m, and E (inte-

gers!) which satisfy this equation. To do so we construct the table
shown.
m n E V PF Solid

3 3 6 4 4| Tetrahedron
3 4 8 6| Cube
3 5 SOM mel 12 | Dodecahedron
3 6 — | — | — |Impossible since 1g + 146 = 14
3 of — | — | — |Impossible since 44 + 4% < 14
4 3 12 6 8| Octahedron
4 4 — | — | — |Impossible since 14 + 14 = 14
4 5 — | — | — |Impossible since 14 + 14 < 14
5 3 30 12 | 20 | Icosahedron
5 4 — | — | — |Impossible since 14 + 14 < 14

Other choices for m and n are also impossible.

Thus there are exactly five regular solids: tetrahedron, cube,
octahedron, dodecahedron, and icosahedron (Fig. 34).
This result was known to the Greeks at least as early as the
fourth century B.c., and was used extensively by Plato in the devel-

ee

Icosahedron Dodecahedron
Figure 34
Geometry 127

opment of his philosophy. Hence, the five regular solids are often
called the ‘“‘Platonic Solids’”’. It is a beautiful theorem, but of little
practical importance. Students of mathematics, however, should see
such theorems in order to be convinced that mathematics has aes-
thetic as well as practical values.

PROBLEMS

1. Find the flaw in the following proof that: All triangles are isosceles.

B B

(a) Given triangle ABC, construct the angle bisector BD at angle B and
the perpendicular bisector DH of side AC. Unless AB = BC, these
lines are distinct and meet in a point D. So far as we know at present,
D may lie inside or outside the triangle, so we draw the two corre-
sponding figures.
(b) Draw AD and DC.
(c) From D draw MD LAB and NDL BC.
(d) AADE= ACDE (right triangles with equal bases and altitudes).
(e) AD = DC (corresponding parts of congruent triangles).
(f) ABMD = ABND (right triangles with equal angles at B and equal
sides BD).
(g) MD = ND and BM = BN (corresponding parts of congruent
triangles).
(h) AADM = ACDN (right triangles with two pairs of equal sides).
(i) AM = CN (corresponding parts of congruent triangles).
(j) In the first figure above, In the second figure above,

BM = BN, BM = BN,
Mies ONG AM =CN.
Adding: AB = BC. Subtracting: AB = BC.
128 MATHEMATICS FOR PARENTS

2. Compute VN = V — E + F for the subdivision of the doughnut sug-

gested by the figure.

3. Compute VN= V — E +Ffor the doughnut with two holes, as sketched

below.

4. On the basis of your previous experiences guess the value of N for a

doughnut with q holes. This is an exercise in intuition—now someone
must prove your guess to be right or wrong. Probably this is beyond you,
but it is fun to try!
5. Compute N for the doughnut using a subdivision different from that in
Prob. 2. You should again get N = 0.
6. Compute N for the doughnut with two holes using a subdivision different
from that in Prob. 3. You should again get N = —2.

ANSWERS

1. D must lie outside the triangle, for it lies on the circumscribed circle.
Even so, the figures above are incorrect. Hither MJ lies on segment AB
or N lies on segment BC, but not both. So on one side we should sub-
tract and on the other side we should add.
Geometry 129

This fallacy illustrates the point that we should never argue from a
figure. In order to handle this problem we need the concept of the order
of points on a line, phrased in a way such that we can prove results with-
out the use of a figure. The modern treatment of geometry permits this;
Euclid’s does not.
2. V =16; EF = 32:F =16:N=V—E+F =0.
Seve 240 = 48: Ff = 32)N =V=B+F= —2,
4. N =2 — 2¢.
SSS]

CHAPTER 12 TRIGONOMETRY
Oo

i
Oo

The word trigonometry means the science of measuring a triangle.

This science was invented to enable us to calculate the magnitudes
of unknown sides and angles of a triangle provided that we were
given the magnitudes of enough sides and angles for the triangle to be
determined. In this way it is possible for us to measure inaccessible
distances such as heights of mountains and widths of rivers by
measuring other distances and angles and then using the formulas
of trigonometry. The method was also applied to the geometry of
triangles on a sphere, and this led to important applications to
astronomy and celestial navigation. Since computations of this kind
involve lengthy arithmetic, it has been customary to carry them out
by means of logarithms, and so instruction in the use of logarithms
has commonly found its place in courses on trigonometry.
In developing the relations among the trigonometric functions
which were needed for these purposes, mathematicians amassed a
considerable theory which is not directly related to the measurement
of the triangle. This subject is known as analytic trigonometry, and
it has many applications in calculus and in other branches of higher
130
Trigonometry 131

mathematics. Thus, trigonometry has been viewed as a subject with

both practical and theoretical aspects, both of which were important
in the education of a prospective scientist or engineer.
The traditional course in trigonometry defined the trigonomet-
ric functions, used them to solve right and oblique triangles by
means of logarithms, and then spent the little time that was left on
trigonometric identities, graphs, and equations. Sometimes the in-
verse trigonometric functions were included.
Such a course, no matter how well conceived in the past, is
extremely unsuitable as part of a modern school curriculum in math-
ematics. The original purposes of trigonometry had to do with sur-
veying and celestial navigation, but how many of our children are
going along one of these paths? Surveying is a small and essential
profession, but its methods are a far cry from those suggested in
textbooks on trigonometry. In particular, I have been told by profes-
sional surveyers that computation by logarithms has long since been
obsolete in the profession—they use desk calculators to work up
their data. Similarly, celestial navigation—once the greatest friend
of sea captains—has been replaced by modern electronic aids. I
agree that I would feel more comfortable on a sea voyage if the
skipper knew how to shoot the sun and compute his position, and I
hope this art is not lost. Nevertheless, the basic navigation of ships
and aircraft is done by other means, and celestial navigation is for
the occasional specialist.
’ If the old applications are as obsolete as this, you may ask why
the course has persisted in its present form. The only reason that I
can give is that of sheer inertia and conservatism. There is no course
in the mathematics curriculum that could stand a greater change.

Definitions of Trigonometric Functions

How, then, is it being modernized in the better schools? The

most striking change is that modern trigonometry (in spite of its
name) has very little to do with triangles, or even with angles for
that matter. The first change that I shall discuss is the definition of
the trigonometric functions themselves, so that they are functions
not of angles but of real numbers like most other functions. From
your school days you will recall that, in the right triangle ABC, we
define the sine of angle A to be the opposite over the hypoteneuse or
132 MATHEMATICS FOR PARENTS

a/c. Similarly, the cosine of A is b/c. (See Fig. 35.) We write these
for short:
sin A = G/c,
cos A = b/c.

In these formulas A is supposed to be an angle measured in degrees

or possibly in radians.

c a

A Cc
b
Figure 35

In other mathematical functions the variable does not stand for

an angle but a real number. For instance in

f(a) =x? — 42 + 7,
or
g(x) = logio x,

we understand that we are to substitute numbers for x. In this way

we compute f(2) = 4—8+7 =8 and g(2) = 0.3031. Tradition-
Trigonometry 133

ally, however, it has been nonsense to write sin 2 unless we state

whether 2 is measured in degrees or radians.
This is an intolerable state of affairs, for in calculus and in
modern applications of trigonometry to electrical engineering we
regularly meet sin 2 with no such specification of units for 2. Let me
give you the modern definition of sin x and cos x. First we draw a

Figure 36

Figure 37
134 MATHEMATICS FOR PARENTS

circle of radius one, and to the same scale a segment PQ of length z

(Fig. 36). Let us think of PQ as a flexible piece of inextensible wire
and place the end P at the point A on the circle (Fig. 37). Now wrap
PQ around the circle as far as it will go. The result may look like

z positive
Figure 38

that in Fig. 38. The point Q will then lie on the circle and has coor-
dinates x and y. We then define

sin2 = y,
COs 2 = &,

where y and x are the coordinates of the point Q.

The above construction assumes that z is positive. If z is nega-
Trigonometry 135

tive we wrap the segment of length |z| around the circle in a clock-
wise direction and thus find the position of point Q (Fig. 39).
From such a figure we can immediately compute the values of
these functions for certain convenient values of z.

Vv

z negative
Figure 39

Since the length of a quarter of a circle is 7/2, we see that

ro : er
sin 5 = 1, cos 7 0. (See Fig. 40.) Since the length of a semicircle

is 7, we see that (Fig. 41) sina = 0, cost = —1. The values of

sin z and cosz for most other values of z cannot be computed so
easily, but good decimal approximations can be obtained from a
large scale graph. In practice one refers to a table.
136 MATHEMATICS FOR PARENTS

With these definitions the trigonometric functions lose their

special character of being associated with angles, and join the rest
of mathematics as useful and important functions of numbers. Now
that they are defined as functions of a real variable, which hereafter

Q(0, 1)
3
|

Figure 40

we call x, the modern course proceeds to find their properties. We

prove identities such as:
sin? z + cos? x = 1,
sin 2x = 2 sin z GOS Z.

We solve equations such as

4sn?z2—4snz+1 =0,
and we define the inverse functions such as

are sin 2, arc COS x.

Trigonometry 1:37

Applications

This is the theoretical part of the subject which was badly

neglected in so many traditional courses. In addition to this we need
to include a full range of contemporary applications to physics and
engineering. One of the most appealing of these is concerned with

Mf

Figure 41

wave motion. The graph of y = sinz is the basis of this and has the
form shown in Fig. 42. This can be modified hy inserting coefficients
to read
y = Asin Bz.
The factor A is called the amplitude and controls the height of the
maximum and the depth of the minimum of the curve. The factor
B is related to the frequency of the curve, and tells us how many
138 MATHEMATICS FOR PARENTS

complete vibrations occur within a fixed horizontal distance. In

ordinary house current, the amplitude corresponds to the voltage
(usually 110 volts) and the frequency to the number of cycles per
second (usually 60). By varying the amplitude continuously, we can

Figure 42

Carrier wave - Transmitted wave

Amplitude Modulation
Figure 43

send a signal on such a wave transmitted by a radio station. This is

the basic idea of AM, or amplitude modulation radio. Similarly, by
varying the frequency, we can obtain F'M or frequency modulation.
(Sée Figs. 43 and 44.) These applications are only a few of the ways
in which trigonometry is now used in science, and it is very important
that we introduce applications such as these into the curriculum in
place of the obsolete materials of yesteryear.
Trigonometry 139

Logarithms

Since logarithms are so commonly associated with trigonometry

in school courses, I should comment on their place in contemporary
mathematics. Originally logarithms were invented to provide a
means for carrying out long series of multiplications and divisions.
By means of the basic formulas

log czy = log x + log y, log 2s log x — log y,

Y
multiplication was converted into addition, and division into sub-
traction. This was a major advance, for addition and subtraction

MUU MA
VV VVVVVVVY WY VV VWI
Carrier wave Transmitted wave

Frequency Modulation

Figure 44

are numerically simpler than multiplication and division. This need

for logarithms, however, has entirely disappeared, for modern desk
calculators and high-speed computing machines now make arith-
metic a trivial and mechanical operation. There is no point whatever
in teaching logarithms to present-day students for this purpose.
On the other hand, there are many theoretical and practical
aspects of logarithms that make them an essential part of a math-
ematical education. In the first place there are many physical
processes whose law of behavior can be expressed only in terms of
logarithms. For example, consider radioactive decay. If one starts
with A grams of a radioactive substance, the number of grams, 2,
remaining after ¢ seconds is given by the formula
a =as Ae-‘tlos.
—(tl 2 Mrs

where ¢ is an irrational number approximately equal to 2.71828 and

tr is the half-life of the substance, that is, the time required for A
 MATHEMATICS FOR PARENTS
140

n is given
grams to decay into A/2 grams. The graph of this functio
in Fig. 45.
tation
Another use for logarithms occurs in the graphical presen
stock market and
of percentage change. Suppose that you play the
You may well plot
wish to chart the behavior of your favorite stock.
in Fig. 46.
the price against time and obtain a graph like that shown
x

Radioactive Decay
Figure 45

78
77
76

Price74

73
72
71
70

Time

Figure 46
Trigonometry 141

This is satisfactory if you are interested in absolute gains or losses,

but many investors place greater importance on percentage gains or
losses.
Suppose that on successive days a stock sells at 76 and 81 for an
absolute gain of 5. The percentage gain, however, is °+4 = 0.07 = 7
log 100

log 90

log 80

log 70

log 60

(logarithmic
Price
scale)
log 50

log 40
Time

Logarithmic Rulers

eee ae eee ee) eee |

0 10 20 30 40 50607080 100
Percentage gain

(Seer | ees eee eee |e ee |

0 10 £20 30 40 50 60
Percentage loss
Figure 47

per cent. This involves arithmetic, and you may prefer to do it

graphically. The trick is to plot log 76 and log 81 against time
instead of plotting 76 and 81. Then the vertical distance
81
log 81 — log 76 = log 767 log 1.07.
 MATHEMATICS FOR PARENTS
142

ruler, you can

By using a logarithmic vertical scale and a logarithmic
y from the graph. Moreov er, a gain
read percentage changes directl
to the same vertica l distanc e no
of 10 per cent (say) corresponds
Thus percen tage change s at differe nt
matter what the base price.
it would be a nuisan ce to have to
times are easily compared. Since
graphs of this type, special paper
look up logarithms in a table for
l scale is
called “semilog” is available. On this paper the vertica

Figure 48

already logarithmic, and the prices can be plotted directly without

looking up their logarithms. (See Fig. 47.)
Asa final application suppose that you wish to plot the graph of
=e Ol

This is as shown in Fig. 48. A simpler procedure (and a more accurate

one) is to write the given equation as
log y = log 2 + 3log z.
Trigonometry 143

Now plot log y against log x. The result is a straight line. The easiest
way to do this is to use “log-log” graph paper on which both the
vertical and horizontal scales are logarithmic. (See Fig. 49.)
It is for applications like these that logarithms need to be taught;
let us leave computing to the machines.

log 100

log 50 (log 3, log 54)

log 30

log 20
(log 2, log 16)

log 10

log 5

log 3

log 2 (log
1,log 2)

log 1
log 1 log2 log3 log4
Figure 49

PROBLEMS

1. Using the definitions of the trigonometric functions compute:

(a) sin 2r (f) cos 51
(b) cos 2r (g) sin (—7)

(c) sin= (h) cos (—7)

Ey con (i)sin(- -
Sea . 2
(e) sin 5 (j) cos(- *)
144 MATHEMATICS FOR PARENTS

. Explain what is meant by the numbers on your radio dial.

Why is sin? z + cos? z = 1?
. Why is there no solution to the equation sin = 2?
Pf
oO
bY
w . Plot y = 27 on a conventional set of axes. Then plot logio y against x.
What is the nature of the curves?
6. Suppose that banks finally go the limit and compound savings interest
not annually, not quarterly, not weekly, not hourly, but continuously.
Then an amount of P dollars ar r per cent interest is worth
A= Pert/100

dollars after t years. How much interest at 4 per cent per annum would
be paid on a deposit of $100 during the first year? How much would the
first year’s interest have been if it had been compounded quarterly?
7. For small values of z, a good approximation to sin z is

; ae
Sa = o> ==
6

Compute sin (0.1) to four decimal places, and compare your result with
its value in a published table. You will need to consult a table which
gives the sine of an angle of 0.1 radians, for published tables have not
caught up with the times.
8. From the definition show that sin (2r + 2) =sinz. Hence using
the result of Prob. 7 compute sin 6.38.
9. Give an example to show that in general

sin (zx + y) ¥sinz+siny.

Hence a possible extension of the distributive law is false.

10. From the definition show that sin (—xz) = —sin z.

ANSWERS

ab, (@)) Wh (f) —1.

(b) 1. (g) 0.
(ec) —1. (h) —1. x
(d) 0. (i) —1.
(e) 0. (j) 9.
2. The number 550 refers to 550 kilocycles. This means that the carrier
wave of this station has 550,000 periods per second. A period of a sine
wave is that illustrated in Fig. 42.
Trigonometry 145

3. By definition sin z = y, cos z = z, where z and y are the coordinates

of the end point of an aro of length z starting at point A in Fig. 38. For
any point on a circle of radius one, x? + y? = 1. Hence

sin? z + cos? z = 1.

4. Sin z lies between —1 and +1 inclusive.

logy=x log,)2

6. Compounded continuously, interest = $4.08. For 9-4 = 1.0408. Com-

pounded quarterly, interest = $4.06.

eee) — 0.1 — 2-0!

= 0.1 — 0.0002
= 0.0998.
8. An arc of length 2x + z wraps around the circle completely with z left
over. Hence the end point of such an arc is the same as the end point
of an arc of length x. Ans. 0.0998.

9. Letzx =F" =yety = x. Then sinz = 1,siny =1,sin(z + y) =

0.0 ~1+1.
10. The end point of the are corresponding to (—z) is directly below that
corresponding to z. Since the figure is symmetrical relative to the
X-axis, sin (—z) = —sin z.
ae
CHAPTER 13 LOGIC
oO

ste es
oO

The process of proof runs throughout all branches of mathematics

and is an essential ingredient of a modern high school course in
mathematics. In order to understand a proof a student needs to have
an elementary introduction to logic. In the past it has been assumed
that an intelligent youngster could learn his logic by seeing how it
was used by the teacher and in the textbook, that is, by a process of
osmosis. For the very bright student this was, indeed, satisfactory,
but for others it is now clear that formal instruction is necessary.
This chapter has, therefore, been written as a brief summary of the
essentials of logic and the nature of proof.

Sentences and Open Sentences

A sentence is a simple statement which is so clearly expressed

that it is meaningful to call it true or false. Examples of sentences are:
Boston is in Massachusetts.
Diamonds are agirl’s best friend.
146
 Logic 147

Although mathematics uses sentences to some extent, the most com-

mon type of mathematical statement is an open sentence. This is a
statement which contains a variable and which cannot be said to be
true or false unless we specify the value of the variable. Examples of
open sentences are:

x+3=7. True if x = 4, false for other values of z.

|
x+édzx+4=0. Trueifx = —1or —4, false for other values of z.

Some open sentences are true for every value of the variable; these
are called identities. Examples are

Ri teat COE CALCD eee9

Vx? = |z\.

Similarly, some open sentences are true for no value of the variable.
Examples are

z= —1, where x is to be real;

0X x = 5.

Compound Open Sentences

In constructing mathematical arguments, we often combine two

or more open sentences containing the same variable into a new,
compound, open sentence by the use of connective words such as and
and or. We must then give rules for deciding about the truth of the
compound open sentence for a particular value of the variable. Of
course, this depends on our previous knowledge of the truth of the
component parts for this same value of x. A simple example is

ge tA and (a — 2)(x4 — 3) = 0.

So that we can analyze this situation, let me introduce some nota-

tion to be used later.

Let pz be the open sentence: x? = 4;

gz be the open sentence: (x — 2)(x — 3) = 0;
pz /\ gz be the open sentence: p, and q-.

The symbol A stands for ‘“‘and,” and p, A q; is called the con-

Junction of pz, gz. Now for what values of x is p, A qr true? The
148 MATHEMATICS FOR PARENTS

answer is based upon a definition which itself is intended to follow

common sense.

Definition. pz A qx 1s true for a particular value of x uf both pz and qz

are true for this value of x. Otherwise px qz 18 false for this value of x.

To illustrate this definition, let me apply it to the example above.

Values of x for Which the
Open Sentence Is
Open Sentence

True False

Dx —2,+2 | Allz except —2, +2

(x 2,3 All x except 2, 3
De \ Qz 2 All x except 2

A standard way of formalizing this definition is a “truth table”,

such as that below.

CONJUNCTION

Pz Qx De IN Gz

HeaHHA Sd

In this table T stands for ‘“‘true” and F for:“false’’. To read the

table, choose a value of « and determine whether, for this x, ps and qz
are individually T or F. Enter the corresponding line of the table
and read T or F under pz A dz:
We can put this another way by considering the corresponding
truth sets (see Chapter 6). The truth set of pz A qr is then the inter-
section of the truth sets of p, and of q, respectively.
In a similar fashion we may connect two open sentences with
the word “or”. In mathematics ‘“‘or’’ means “either or both” and is
used in the sense of the legal phrase “and/or”. The symbol V stands
for “or” and p, V qzis called the disjunction of pz, dz. The correspond-
ing truth table is:
Logic 149

DISJUNCTION

Pr Ux Dz V Qe

a ly a
Hl F a
F fi a
F F F

In terms of truth sets we see that the truth set of p, V q, is the

union of the truth sets of p, and of gq, respectively.

Equivalent Open Sentences

Two open sentences may have quite different forms of expres-

sion, but still have identical truth sets. For example,
x?+ 5x = —6,
and
(x + 2)(@ + 3) =I 0.
have the same truth sets. This leads us to the definition of equivalent
open sentences.

Definition. Two open sentences are equivalent if and only if they have
identical truth sets.

The corresponding truth table is:

EQUIVALENCE
Pe dx Pr x

Miwaeg
ke Sey

The symbol «= is read “is equivalent to”. Much of ordinary

algebra consists of learning how to transform a somewhat compli-
cated open sentence into a simpler one which is equivalent to it.
 MATHEMATICS FOR PARENTS
150

The Negation of an Open Sentence

Associated with each open sentence is another open sentence

called its negation. The negation of pz is written ~pz. The basic idea
behind this is that the truth set of ~p, consists of those values of x
for which p; is false. That is, the truth set of ~pz is the complement
of the truth set of p,. Often we can form negations by inserting a
“not” in an appropriate place. For example:

Pz —~DPxz

g=4 x? is not equal to 4; i.e., x? # 4

Triangle x is isosceles Triangle zx is not isosceles

In other cases the formation of negations is more complicated.

The corresponding truth table is:

NEGATION

Dx pe

ay F
F cD

Implication
The great bulk of mathematical theorems are of the form: Ii pay
then q:. For example:
If x is an even number, then x + 1 is an odd number.
There is, moreover, a subtlety here that is all too frequently ignored.
In our example above, we really mean:
For all integers «: If x is an even number, then x + | isan
odd number.

In other words, our theorems are general statements which are to be

true for all values of our variable. We write such an implication in
the symbolic form
Vz (pz — QQ),
which is read: ‘‘For all x: If p., then q.’.
Logic 151

In order to make progress, we must say what is meant by the

truth of p,— qz for a particular value of x. To gain some feeling for
the situation, let us consider the example:

For all x: If 2? = 4, then (« — 2)(x — 3) = 0.

Let p, be a? = 4, gz be (x — 2)(x% — 3) = 0.

Then we have

Values of x Pz dx Dz — Qz

2 ue oi
—2 cl F To be
3 F eli defined
Any other x F F

The question now is what to write in the right hand column.

The answer may seem arbitrary, but it works out quite well. It is
given by the truth table:

IMPLICATION

Dz a Pz dx

elacia
eolae| ea

In our example above we see that p,— q; is true for all values
of x except « = —2, and that it is false for c = —2. Hence it is not
true for all x, and the statement

For all x: If x? = 4, then (a — 2)(a — 3) = 0

is false. Hence this is not a theorem which we are entitled to use in

later work.
On the contrary, consider the example

For all x: If z = 2, then 2? = 4.

152 MATHEMATICS FOR PARENTS

Let px be x = 2, and let g. be x? = 4. We have:

Values of x De

2 a
—2 EF
Any other x F

In this table there is no entry where p, is T and q, is F, and

(therefore) pz — qz is F. Thus our implication 7s true for all values
of x. It is, indeed, a useful theorem.

Figure 50

There is one very important point to notice, namely, we really

need not bother with those values of x for: which p, is false, for
regardless of the truth or falsehood of g, for these values, p: — qz 18
true. We can then concentrate on those values of x for which p; is
true. For every one of these we must show that q, is true. If there is
a single value of x for which p, is true and q, is false, then our main
statement: V, (p: — qz) is false.
This situation can be explained in another way by Fig. 50. Let
the points of the page in your book represent all possible values of
z, that is, the universal set. Let those values of « for which pz is
true lie inside the circle labeled P, and those for which q, is true lie
inside the circle labeled Q. Of the many relative positions in which
Logic 153

circles P and Q may lie, there are just two which correspond to the
statement that V, (p:— q,) is true, namely, the situations when P
lies inside Q or is identical with Q (Fig. 51).

P Pand Q

(a) (b)
Figure 51

Figure 52

In our example

Forally:ii a= 2, then?’ = 4

the figure looks like Fig. 52. The large circle includes points where
x? = 4, namely, x = —2 and xz = 2. The small circle contains x = 2
154 MATHEMATICS FOR PARENTS

only. The shaded region corresponds to x = —2. Since P lies inside

Q, the theorem is true.
In our previous example:

For all x: If 2? = 4, then (2 — 2)(a — 3) = 0

Fig. 53 applies. Since P contains points not in Q, the theorem is

false.
In summary, to prove a statement of the form V, (pz — qz), we
consider all x for which p, is true. We then check to see whether q, is

Figure 53

true for all these values of x. If it is, the statement is true. If, how-
ever, there is a single x for which p, is true and qz is false, the state-
ment is false.

Implications Related to a Given Implication

Associated with the implication V.(p:— qz) there are three
other implications with which it is often confused:

Converse: Via (Gz — Dz).

Inverse: V,(~p,—> —@:)
Contrapositive: Vz (~q:— ~ pz).

The question is whether any or all of these are equivalent to the

given implication. We can settle this in two ways.
Logic LO)

First, let us use an extended form of our truth tables.

Contra-
Inverse, positive,
NV 7 Mf ~dz 7 ~Dzx

HHAmaH

To construct this table we give p, and gq, all possible combinations

of values of T and F in the first two columns. We then write T or F
on each line of succeeding columns as we are directed to do by our
earlier truth tables for negation and implication.
From this table we see that in the case where pz — qz is true for
all values of x (that is, there are no x’s corresponding to line 2) there
may be x’s for which its converse is false (line 3). Similarly, there
may be 2’s for which the converse is true and the given implication
is false (line 2). Hence the given implication and its converse may
not be equivalent. For example,
Ifx = 2, then x? = 4
is not equivalent to
If x? = 4, then x = 2.
On the other hand, an implication and its converse may be
equivalent. This occurs when there are no z’s corresponding to either
line 2 or to line 3. For example,
If 3z = 6, thenx = 2
is equivalent to
If+ = 2, then 3x = 6.
In a similar way we can show that:
An implication and its inverse may not be equivalent.
The converse and inverse of an implication are always equivalent.
An implication and its contrapositive are always equivalent.
A second way of seeing this is by means of circular diagrams
like those above. Let us consider an implication and its converse
(Fig. 54). The given implication states that P is inside Q or is
identical with Q, and the converse that Q is inside P or is identical
156 MATHEMATICS FOR PARENTS

with P. These two statements will be contradictory unless P and Q

are identical; hence they are not equivalent. I leave consideration of
the inverse and the contrapositive to the problems.
In summary, it is false reasoning to conclude that the converse
or the inverse of a true implication is true. It may or may not be

We Deed) Vx(9,—>

(a) (b)
Figure 54

true, and more investigation is required. It is correct, however, to

conclude that the contrapositive of a true implication is true.

Direct Proof
The most common type of mathematical proof is a direct proof
involving a chain of implications which have previously been shown
to be true. Suppose that we are given that the following are true:

Va (pz aa 2), Vi (qxcae: ia)s Ve ( a: 8;).

O | coy (0)
Figure 55
Logic Wa 74

We wish to prove that V, (p;— sr). The argument is immediate; for

we are given the diagrams in Fig. 55. In order to save space we shall
draw such diagrams with circle P inside circle Q, but we interpret
this to mean that P is either inside Q or identical with Q. From these
we derive Fig. 56. Thus P is contained in S and V, (p; — sz) is true.
There are endless variations of this pattern, and great skill is
required in arranging the work so that the proof is valid. One com-

Figure 56

mon trick, for instance, is the replacement of an implication by its

contrapositive if this will simplify matters. For example:
Given: Wie (pz 2, ~r); Vie (Mr ar r)-

Prove: V2 (pz— ~?c).

Proof: Since V; (7: — qz) is equivalent to Vi (~q: > ~r,z), we have
the chain: V, (pz > ~Qz), Vz (~G: ~?x). Hence we conclude that
V. (pz: — ~rz) is true.
As a specific example, consider the following proof of
For all integers x: If x is odd, then 2? is odd.
This amounts to the following:
Given: For all x: If x is odd, then x = 2a + 1, where ais an integer.
For all z: If x = 2a + 1, where a is an integer, then « is odd.
The rules of algebra.
158 MATHEMATICS FOR PARENTS

Prove: For all x: If x is odd, then 2? is odd.

Proof:
1. For all z: If x is odd, then 2 = 2a + 1. (Given)
2. Forall x: If c = 2a+ 1, then 2? = 4a? +4a+1. (Algebra)
3. 402+ 4a+1 = 2(2a2 + 2a) +1 =2b+1, where 6b is an
integer. (Algebra)
4. For all x: If « = 2a +1, then 2? = 26+ 1, where b is an
integer. (Substitution)
5. For all a: If 22 = 2b + 1, where b is an integer, then x? is odd.
(Given)
6. For all a: If x is odd, then 2? is odd. (From 1, 4, and 5)

Indirect Proof

This method of proof is a bugbear. to students and teachers

alike, but it is really very straightforward. The idea is that
V; (pr — dz) must be either true or false. If we can show that it is
not false, then it must be true. We proceed by assuming that it is
false, combining this assumption with other known facts, and (with
luck) arriving at a contradiction. Since contradictions are impossible
in correct thinking, we must have made a mistake somewhere. Our
only dubious statement was the assumption that V, (p:— qz) is
false. Hence this must be in error, and V, (p:— qx) must be true.
The difficulty in practice usually arises when a student tries to
state that V.(p:— qd) is false. But from what we saw above the
correct procedure is crystal clear; the only way that Vz (pz — qr) can
be false is that, f
For some x, pz ts true and q; is false.

That is, some point inside the P circle must be outside the Q circle.
Let us try an example:
Prove: For all integers x: If x? is even, then x is even.
Proof: Weassume that the given implication is false, namely, that:
For some x: 2? is even and x is odd (not even).
From our previous proof, above, however, we know that it is impos-
sible for x to be odd and x? to be even. Hence we have arrived at a
Logic 159

contradiction; our assumption is incorrect, and the statement to be

proved is true.

Disproof

Students sometimes have difficulty in disproving statements

which are asserted to be true for all x. For example, how do you
disprove
For all a: 22 + 16 = (a — 4)(a + 4).

There is a natural procedure: find one value of x for which the open
sentence is false. Such an z is called a counterexample. In the above
example xz = 0 will do the job.

Conclusion

This has been a very brief and informal discussion of logic and
methods of proof and disproof, but I hope that you have some idea
of these topics. The construction of a proof often involves ingenuity
of a high order, and this takes time to develop even if the brain
power is adequate. It is nowhere nearly so difficult to recognize a
correct proof when you see one, and I hope that this discussion will
help you find your way through some of the modern textbooks your
children are using.

PROBLEMS

1. Using truth tables show that (p.) V (~pz) is true for all values Ofeas
Hint; Fill in the right-hand column of the table:
Dx ~Dz (pz) V (~Pz)

© F
F T
2. Using truth tables show that ~[(p:) A (~pz)] is true for all values of x.
oo.Using truth tables show that p, — qz is equivalent to ~[(pz) A (~¢4z)]-
4. In order to treat the inverse and contrapositive of Vz (pz — 2) in our
diagrammatic fashion, we draw a square to represent the universal set
consisting of all values of x. The circle P inside this square contains
those values of x for which p; is true. The remainder of the square con-
160 MATHEMATICS FOR PARENTS

tains those x for which p, is false (figure below) that is, for which
>
(~pz) is true. Using this idea show that V. (pz — qr) and Vz [(~qz)
(~p,)] are equivalent.

As in Prob. 4, by diagrams show that V. (qz— pz) and V; [(~pz) >

(~qz)] are equivalent. ,
6 Construct a proof for the following, or show that no such proof exists.
Given: For all x: If 24 = 16, then x? = 4.
For all v7: If 2 = 2 then x? = 4.
Prove: For all x: If x4 = 16, then x = 2.
a Construct a proof for the following, or show that no such proof exists.
Given: For all x: If 24 = 16, then x? = 4.
For all x: If z is other than —2, 2, then x? ¥ 4.
Prove: For all x: If 2! = 16, then 2 = —2 or 2.
8. Give a direct proof of:
For all integers x: If x is even, then x? is even.
9 By means of an indirect proof, establish the truth of: For all integers z:
If x? is odd, then z is odd.
10. Prove or disprove:
For all real numbers a, 6, c, d: Ifa > band¢ < d,thena +e >6 +d.
11 Is the following a correct indirect proof?
Prove: For alla, 6: If a and 6 are integers, then (a/b)? # 2.
Proof:
(1) Assume that for one pair of integers, a and 6, (a/b)? = 2.
(2) Let k be the greatest common divisor of a and b so that a = kp,
b = kq, where p and q are integers with no factors in common. Then
(a/b)? = (kp/kq)? = p?/q? = 2.
(3) p? = 29.
(4) p? is even, so p = 2r.
Logic 161

ea 267)
(6) e272) — g?.
(7) q? is even.
(8) gis even, so q = 2s.
(9) p and q have 2 as a common factor.
(10) Step 9 contradicts step 2.
(11) Therefore, the assumption in step 1 is incorrect and the theorem
is proved true.
12. Is the following a correct indirect proof?
Prove: For all 2: If z > 3, then 2? > 9.
Proof:
(1) Assume: For all x: If e< 3, then 2? < 9.
(2) Let x = —5. Then —5 < 3, but (—5)? = 25 and is not <9. Thus
z = —5 is counterexample to step 1, and step 1 is false.
(3) Since the assumption in step 1 is false, the theorem is true.
13. Some theorems in mathematics cannot be written conveniently in the
form Vz (pz — gz), but indirect proof can still be applied to them. Con-
sider the theorem:
There exist an infinite number of primes.
Recall that a prime is a positive integer which has no integral divisors
except itself and 1. To prove this theorem we proceed as follows:
(1) Assume that the theorem is false, that is, assume that: There are
only n prime numbers, where n is a (finite) positive integer.
(2) Then we can make alist of these primes, say, Pi, Pa, + + + 5 Pn
(3) Consider P = pi X p2 X ++ * X Pn +1. We wish to produce a
contradiction by showing that P is a prime not in the above list of
all primes.
(4) First, P is not equal to any of pi, . . . , Dn, for it is clearly larger
than any of them.
(5) Second, P is a prime (prove it).
(6) Hence there is a contradiction with our assumption in step 1 and
the theorem is true.

ANSWERS

qe Dz | ~DPz | (pz) V (~D:)

oy F ali
F el iL

2. Pz | ~pz | (pz) A (~pz) | ~[(p2) A (~D:)]

i. F F ar
F L F fi
 MATHEMATICS FOR PARENTS

~qa
(pz) \ (~q@) ~[(pz) A (~9az)] (Dz — qz)
————$_$———_——— SCT
Logic 163

Hence the region where ~; is true is included in the region where ~qz
is true, or Vz [(~pz) — (~qz)] is true. Now reverse the process.
. There can be no proof, for z = —2 is a counterexample to the conclu-
sion.
. Proof:
(1) For all x: If ct = 16, then 2? = 4. (Given)
(2) For all x: If xz is other than z = —2, 2, then 2? # 4. (Given)
(3) For all 2: If 2? = 4, thenz = —2or2. (Contrapositive of step 2)
(4) For all x: If x! = 16, then x = —2 or 2. (From steps 1 and 3)
. Proof:
(1) For all x: If z is even, then « = 2a where a is an integer. (Given)
(2) For all x: If x = 2a, then x? = 4a’. (Algebra)
(3) 4a? = 2(2a?)= 2b, where 6is an integer. (Algebra)
(4) For all x: If = 2a, then x? = 28, where 0 is an integer.
(Substitution)
(5) For all x: If x? = 2b, where bis an integer, then x? iseven. (Given)
(6) For all x: If z is even, then 2? is even. (From steps 1, 4, and 5)
NEroor:
(1) Assume that, for some z, x? is odd and zis even.
(2) This contradicts the result of Prob. 8.
(3) Hence the assumption is incorrect and the theorem is true.
10. False. Let a = 2, b = 1, c = 3, d= 10. The na
> b and c <d. But
a+ec=5andd+4+d = 11. Hence it is false thata +e > 6 +d.
on This is correct. It shows that there is no rational number a/b whose
square is 2, or that / 2 is irrational.
12. This is incorrect. The assumption in step 1 is the wrong one if you wish
to assert that the theorem is false. Actually the implication in step 1 is
the inverse of the given implication, which has nothing to do with the
case! The correct assumption to make is: For some z:z > 3 and 2? < 9.
This does not lead to any reasonably evident contradiction, and so the
method of indirect proof is not suitable for this theorem.
We can give a simple direct proof as follows: Consider x? — 9.
We know that x? —9 = (rx+ 3)(c4 — 3). Now x > 3; sor+3>0
and (x — 3) > 0. Hence (x + 3)(x — 2 9.
3) =>0, or 2?
13. To prove that P is a prime, we use indirect proof again. Assume that P
is not a prime. Then P must be divisible by at least one of the existing
pened, py, =: . , Pa. Sin ce
P = p, XP2X~"* +1, this is a
X pn
contradiction. Hence P is a prime.
CONCLUSION

If you have followed me this far, you should have an understand-

ing of the general trends of the Revolution in Mathematics. I have
left many things aside such as the introduction of probability,
matrix algebra, and calculus into the high school. I have failed to
mention by name the host of college and school mathematicians who
have contributed their time and energy. I have necessarily slid over
many delicate points of mathematics. i
I hope, however, that I have given you the flavor of the move-
ment. If you wish to go farther, you should sit down and work
through some of the modern textbooks which seem so hard to the
teachers, but which are so easy for your children.
INDEX

mm (1 Between, 114
Absolute value, 91 Binary system, 39
Abstraction, 3
Acceleration, 15 Calculus, 14
Actuary, 31 Careers, 31
Addition, repeated, 77 Cartesian product, 59
Advanced Placement Examination, 16 Closure, 64
Advice to school districts, 19 Commercial aids, 22
Algebra, laws of, 63 Commission on Mathematics, 17
“All triangles are isosceles’, 127 Commutative law, 64
Analytic geometry, 14, 118 Complement of a set, 55
Angle, 117 Complex numbers, 41
interior of, 117 Compound interest, 144
Applied mathematics, 5 Computation, 13
Associative law, 65 Conjunction, 147
Audio-visual aids, 22 Contrapositive, 154
Axioms, 3 Converse, 154
in geometry, 4 Coordinate system, three dimensional,
in physics, 4 52
two dimensional, 50
Base seven, 35 Cosine of, acute angle, 132
ten, 34 real number, 134
two, 38 Counterexample, 159
165
166 Index

Courses, 29 Identity elements, additive, 66

for non-college-capable, 30 multiplicative, 67
high school, 29 Imaginary numbers, 41
Cube, 122, 123 Implication, 150
Cuisenaire rods, 23 Indirect proof, 158
Curriculum, high school, 30 Industrial mathematics, 32
Inequalities, 85
Decimal notation, 34 in the plane, 98
Decision making, 28 Inside, 115
Deduction, 5 Institutes, 21
Dienes, Z. P., 23 Interior of, angle, 117
Direct proof, 156 general polygon, 120
Disjunction, 148 triangle, 117
Disproof, 159 Intersection of two sets, 55
Distributive Law, 67, 76 Intuition, 6
Divisibility by 3, 9, and 11, 81 Intuitive geometry, 118
Division, 67 Inverse, logical, 154
Dodecahedron, 126 Inverse elements, additive, 67
multiplicative, 67
Einstein, 5
Equivalence, 149 Language, 15
Euclid, 113 Less than, 86
Euler, 121, 124 or equal to, 89
Euler’s Theorem, 121 Linear inequalities, graph of, 98
simultaneous, 101
Field, 63, 68 Linear programing, 27, 101
finite, 70 basketball problem, 102
modular, 70 diet problem, 106
Foreign language, 30 football problem, 109
Function, definition of, 58, 60 lumber problem, 110
domain of, 59 Log-log graph paper, 143
range of, 59 Logarithms, 139
rule of, 59 Logic, 146
Game Theory, 27
Generalizations, 9 Management Science, 28
Geometry, 113 Mathematical model, 2
analytic, 118 Mean, arithmetic, 92
intuitive, 118 geometric, 92
non-EHuclidean, 7 Modulation, amplitude (AM), 138
recommended curriculum in, 119 frequency (FM), 139
solid, 119 Modulus, 70
Graph of, linear equation, 51 Multibase arithmetic blocks, 23
set, 49 Multiplication of signed numbers, 79
Greater than, 86 Mutations, 9
or equal to, 89
National Science Foundation, 17, 21
Half-life, 139 Nature of mathematics, 1
How to start, 22 Negation, 150
Negative integers, 39
Icosahedron, 126 Newton, 4
Identities, logical, 147 Non-Euclidean geometry, 7
INDEX 167

Number, 33 Sets, 48
base seven, 35 Cartesian product of, 59
ten, 34 complement of, 55
two, 38 identical, 54
complex, 41 intersection of, 55
imaginary, 41 truth, 56
natural, 33 union of, 55
negative, 39 universal, 55
rational, 40 Sine of acute angle, 131
real, 40 Sine of real number, 134
Numeral, 33 Sine wave, 137
amplitude, 137
Octahedron, 122, 124 frequency, 137
One to one correspondence, 52 Skinner, B. F., 24
Open sentence, 56, 147 SMSG, 17
compound, 147 Solid geometry, 13, 119
equivalent, 149 Solids, regular, 125
Operations Research, 27 Square root of 2 is not rational, 160
Opposition, 18 Structure of mathematics, 2, 7
Order on a line, 85, 114 Subset, 54
Subtraction, 67
Percentage change, 140 Sum of three numbers, 66
Place value, 34 Support, types needed, 20
Plato, 126
Primes, infinitude of, 161 Teacher training, 21
Probability, 14 Teaching machines, 24
Programed learning, 24 Tetrahedron, 121, 123
Projectors, overhead, 25 “Three’’, 3
Proof, direct, 156 Topics, old and new, 12
indirect, 158 Transitive Law, 88
Triangle, interior of, 117
Quadratic formula, 91 Trichotomy, Law of, 86
Quadratic inequalities, 89 Trigonometry, 130
Truth, 5
Rational numbers, 40 set, 56
Real numbers, 40 table, 148
Reference materials, 20, 31
Regular solids, 125
UICSM, 17
Relation, 60
Undefined terms, 3
Understanding, teaching with, 11
Scientific Revolution, 26
Union of two sets, 55
Semilog graph paper, 142
Universal set, 55
Sentences, 146
open, 56, 147
Separation of the plane by aline, 115 Variable, 56
 ~'ee
a“

» j
. - 7
.
.
’
na ibbdpry
7 won ive 2 iow

A 5) oeM- a, ;

DAL Bit otbetitiy Ht hive x” Awe

ASE jaloge ebilok et ’

toy £40 at & lo toot sseitna Liisi oeweet UF om
iasmeritoct ly eiuioua multiple,
Da. toed . bi FS other
tate TH uteitsavhink Ljsgcahingns, Lt
9 odoin seuls Jo 0A lath ana eR enon
Manan maqqys trea econ! pf ee
T Awe wre 3 4

‘
 DATE DUE
2)

SEC 6
Y

B®)

9
3
fiz

” oOo wo
Oo

0
ray
 Allendoerfer, Carl Barnett
QA
a:
~A558 Mathematics for parents.

SCIENCE
QH
From the preface...
"This book has been written to answer
the many questions about the ‘new
mathematics' ... my colleagues ask me
about this 'base 7' business their children
are being exposed to; when I meet someone
at a summer resort he is likely to turn
out to be a member of a school board who
wants the real truth about SMSG; and when
I go out in the evening my friends blame
me for all the troubles their children are
having in school with mathematics that
even father never saw before. ... This
book is designed to get me off the hook,
for hereafter I can say: I know what
you are about to ask. I have it
all written up in my new book."

CARL B. ALLENDOE is an outstanding

RFER
mathematician who has been deeply involved in
developing programs to teach the “new mathe-
matics.” The same charm and wit which we-*>

IH
evident in his recent television series on the ‘
math” permeates this book. Professor of Mi
matics at the University of Washington a
|
former president of the Mathematical As ~ELIBRARY. 253
||
8212
ation of America, he is the author of widely
textbooks.

THE MACMILLAN COMPANY

60 Fifth Avenue, New York 10011
*£81L-S
PU
37