LOGICAL WARM-UP 1

1
Logical Warm-up

If mathematics is regarded as a language, then logic is its grammar. In other words, logical precision
has the same importance in mathematics as grammatical correctness in a language. Both can be acquired
either through a formal study or through good practice. In this chapter we do not proceed to study
formal logic. Rather, we emphasise those aspects of it which are vital to the study of mathematics.
Indeed, the essence of what we are going to say is the same as what any layman would tend to think as
common sense, provided he is in the habit of carefully weighing his statements. Some people are
naturally good at these sorts of things just as some privileged persons have a musical ear. The moment
they come across a flaw in some reasoning, some inner sense within them keeps them on the alert.
Unlike musical expertise, however, anyone can acquire this skill through careful practice.

1. STATEMENTS AND THEIR TRUTH VALUES

It is by no means an easy task to give a complete and rigorous definition of a statement. Indeed, such a
definition would cut deep into formal logic, linguistics and philosophy. For our purpose, a statement is
a declarative sentence, conveying a definite meaning which may be either true or false but not both
simultaneously. Incomplete sentences, questions and exclamations are not statements.
Some examples of statements are:
(1) John is intelligent.
(2) If there is life on Mars then the postman delivers a letter.
(3) Either grandmother chews gum or missiles are not costly.
(4) Every man is mortal.
(5) All men are mortal.
(6) There is a man who is eight feet tall.
(7) Every even integer greater than 2 can be expressed as a sum of two prime numbers.
(8) Every man with six legs is intelligent.
Notice that (2) and (3) are perfectly valid statements although they sound ridiculous, because there
is no rational correlation between their components. Statements (4) and (5) can be logically regarded as
same or equivalent statements. Note that they are statements about members of a class (in this case the
class of all men) as opposed to the first statement which is about a particular individual called John.
Statement (6) is an example of what is known as an existence statement. It asserts the existence of
something but does not do anything more. For example in the present case, statement (6) does not say
 2 INTRODUCTION TO GENERAL TOPOLOGY

anything about the name or the whereabouts of the eight-feeter nor does it say that he is the only such
man. There may or may not be other men who are eight feet tall. The statement is silent on this point.
Statements must be either true or false but not both. The truth value of a statement is said to be
True or T or 1 if it is true and False or F or 0 if the statement is false. The beginner is warned not to
be misled by the word value here which has the connotation of numerical value. In particular, 1 and 0
are merely symbols here and not the integers which they usually represent. There is indeed some good
reason why they are used here despite the likely confusion, but it is beyond our scope to discuss it here.
For our purpose, giving the truth value of a statement is merely a fancy way of saying whether it is true
or false. What is important is that there is no in-between stage. There is no such thing as saying that a
statement is somewhat true or almost true. True means absolutely and completely true, without
qualifications. This is especially important to understand in the case of statements such as (4) and (5)
which deal with members of a class. Such statements are to be regarded as false even when there is just
one instance in which they fail. Indeed this is where a logician or a mathematician differs from a
layman for whom the exception proves the rule. In mathematics, even a single exception (a counter-
example as it is called) renders false a statement about members of a class.
Of course, the truth value of a statement may depend upon the way it is interpreted. For example in
statement (1) above, the truth value depends upon what one means by intelligent. But once intelligence
is clearly defined then the statement must be declared as either completely true or false. Actually unless
such a precise definition is implied and understood, (1) is not a valid statement. Truth values of statements,
such as (2) or (3) will be discussed later.
Statement (7) is the famous conjecture of Goldbach. Conjecture means a statement whose truth
value is not known at present. So far nobody has proved the statement nor has anyone disproved it.
(Note that since the statement is about a class of integers, in order to disprove it, that is, to prove that it
is false, all one has to do is to exhibit just one even integer greater than 2 which cannot be expressed as
a sum of two primes. But no one has found such an integer so far.) Thus, at present no one knows
whether (7) is true or false. However, this does not prevent it from being a statement. For it does have a
definite truth value, the only trouble is that we do not quite know what it is. This is indeed a fine point.
As was remarked earlier, the existence of something is to be distinguished from other knowledge about
the thing. This observation applies to the truth value of a statement.
Another interesting point is presented by statement (8), which sounds as a statement from mythology
or from fairy tales. We know no man has six legs. And, therefore, the statement (8) (and indeed any
statement about a six-legged man) is logically true. Such a statement is said to be vacuously true. This
certainly appears very puzzling to a layman. Amoments reflexion will, however clear the mystery.
For, if statement (8) is not true, it must be false, as there is no other possibility. But the only way
statement (8) can be shown to be false is by finding (that is, proving the existence of) a six-legged man
who is not intelligent. Since no man has six legs, this is impossible, the question of intelligence does
not arise at all. So the statement (8) cannot be proved to be false. Therefore, we are forced to conclude
that it is true. (This argument is not convincing to all persons. Such persons have founded their own
school of mathematics called the constructivist school. The main point where they differ from the
classical mathematical logic is that they do not accept proofs by contradiction, or the so-called reductio
ad absurdum arguments of which the present one is an example. We shall, however, agree to bend our
logical conscience if necessary in order to accept that (8) is true.)
As a practical application of vacuous truth, a student who has not appeared for any subject in an
examination can boast that he secured a first class in every subject that he appeared for. Logically, his
claim is absolutely correct. It is easy to set him right, however, if one merely notes that in such a case it
is equally true to say that he has failed miserably in every subject he appeared for. Both statements are
 LOGICAL WARM-UP 3

equally true and there is no logical contradiction, because both statements deal with something non-
existent. We shall return to this point later when we shall discuss logical implications.
To conclude this section we give some phrases or sentences which are not statements.
(1) Will it rain?
(2) O! Those heavy rains.
(3) I am telling you a lie.
The first two are not statements because they are not declarative sentences. The third one is not a
statement because it cannot be assigned any truth value. If it is true then according to its contents it is
false. On the other hand, if it is false then according to its contents it is true. The trouble with this
sentence is that it refers to itself. Many paradoxical situations arise because of self-referencing and the
reader must have come across some popular puzzles which are tacitly based upon self-reference or
some variations thereof.

EXERCISES

1.1 Which of the following expressions are statements? Why?

(1) If it rains the streets get wet.
(2) If it rains the streets remain dry.
(3) It rains.
(4) If it rains.
(5) John is intelligent and John is not intelligent.
(6) If John is intelligent then John is intelligent.
(7) If John is intelligent then John is not intelligent.
(8) For every man there is a woman who loves him.
(9) There exists a woman for whom there exists no man who loves her.
(10) There exists a woman such that no man loves her.
(11) This sentence is false.
(12) There is life outside our solar system.
1.2 On both the sides of a piece of paper it is written, The sentence on the other side is false. Are
the two sentences so written statements? Why? What if on one side The sentence on the other
side is true is written and on the other side The sentence on the other side is false?
1.3 A barber in a village makes an announcement, I shave those (and only those) persons in this
village who do not shave themselves. Is this announcement a statement? Why?

2. NEGATION, CONJUNCTION, DISJUNCTION AND TRUTH TABLES

There are three ways of manufacturing new statements from given ones. Any complicated statement
can be shown to be obtained from some very elementary or simple statements (such as it rains or
John is intelligent). Let us study one by one the three ways of generating new statements.
(i) Negation: To negate a statement is to make another statement which will be opposite to the original
one in terms of truth value. That is, we want that the negation be true precisely when the original
statement is false, and vice versa. The simplest way to achieve this would be to merely prefix the phrase
it is not the case that before the original statement. For example the negation of John is intelligent
would be It is not the case that John is intelligent. However, this is a very mechanical way and
sentences formed like this tend to be linguistically clumsy. It is much better to say that John is not
intelligent which conveys exactly the same meaning. We can even go further and abbreviate not
 4 INTRODUCTION TO GENERAL TOPOLOGY

intelligent to unintelligent, or to, say, dumb, provided we agree that intelligent and dumb are
antonyms, that is, words whose meanings are opposite to each other. Similarly, X is rich may be
negated as X is poor and so on.
In mathematical statements symbols are often used for brevity. The negation of such statements is
expressed by putting a slash (/) over that symbol which incorporates the principal verb in the statement.
Thus x = y (read x is equal to y) is negated as x y (read x is not equal to y). Similarly, x A is
the negation of x A, (read x belongs to A).
Things are not so simple when we come to more complicated sentences. For example suppose the
original statement is John is very intelligent. Of course one way to negate it is simply to say It is not
the case that John is very intelligent. But this is hardly satisfying. Amuch more reasonable answer is
John is not very intelligent. In practice we often regard such a statement as a polite way of saying that
John is dumb (in fact very dumb). In mathematics this is not so. The original statement is about the
degree of Johns intelligence. There are many degrees of intelligence ranging from very high to very
low. Just because John lacks a very high degree does not automatically mean that the level of his
intelligence is very low. It could be that he is just average. In this case, therefore, the correct logical
negation is merely John is not very intelligent and not John is very dumb. Asimilar remark applies
in all cases where there are more than two possibilities. For example to say that a book is not red does
not necessarily mean that it must be of some other specific colour, say, green (although they are
complementary colours). Hence, the logical negation of The book is red is simply The book is not
red and not The book is green or The book is blue.
Even greater care is necessary when we come to statements such as Every man is mortal. Alayman
is most apt to negate this statement either as Every man is immortal or No man is mortal or perhaps
even as Every woman is mortal. Alogician will however negate it as There exists a man who is not
mortal or as There is an immortal man. Recall that the statement in question is about a whole class,
namely, the class of all men. To say that it is false simply means that it is false in at least one instance.
This easily leads to the correct negation. Note that the negation of a statement beginning with every is
an existence statement. We can also state the negation as Not every man is mortal. But it is better to
avoid saying Every man is not mortal as it is likely to be confused with Every man is immortal.
When we come to the negation of existence statements, the tables are turned around. As expected,
the negation will be a statement about a class, asserting that every member of that class fails to be
something whose existence is asserted by the original statement. For example the logical negation of
There exists a rich man would be that Every man is poor or No man is rich, and not There exists
a poor man. The negation of There exists a woman such that no man loves her is For every woman
there exists a man who loves her. The reader is strongly urged to master these types of sentences
thoroughly, that is, to interpret them precisely as well as to negate them correctly as mathematics is full
of such statements.
Sometimes statements are represented by symbols p, q, r, etc. just as in algebra one uses the symbol
x, y, z, . . . , etc. to denote some variable quantities. With this notation there is a special symbol,
p(read as not), for negation. For example if p stands for the statement John is strong then p is read
as not p and denotes the statement John is not strong. Note that, by the very definition, the truth
values of p and p are always opposite of each other, no matter what p stands for. Although p is
standard, the notations ~p or p are also commonly used for the negation of p.
(ii) Conjunction: When two statements are joined together by the word and, the resulting statement
is called their conjunction. For example the conjunction of John is rich and Bob is weak is John is
rich and Bob is weak. Where possible, we can paraphrase the conjunction so that it looks better English.
For example the conjunction of John is intelligent and John is rich can be stated less mechanically as
 LOGICAL WARM-UP 5

John is both intelligent and rich. Also, instead of and we can use but when we want to stress an
implied contrast between the two statements. For example we may say Mumbai is big but Delhi is
beautiful. Mathematically, it means the same as Mumbai is big and Delhi is beautiful.
The standard notation for conjunction is , read as and. Symbolically, if p and q are statements
then their conjunction is denoted by p q and is read as p and q. The reason for introducing the
special symbol instead of simply writing the word and is that we use the word and in ordinary
language so often that its use as a symbol for conjunction is sometimes likely to cause confusion. Note
that we can form the conjunction of any two statements whatsoever. There need not be any semantic
connection between the two, although in practice such a connection is usually understood. That is why
a statement such as Missiles are costly and grandma chews gum sounds ridiculous, although
mathematically it is a perfectly valid statement obtained by the process of conjunction.
When we come to the truth value of a conjunction of two statements, it should be obvious to anyone
that it is true when both the statements are true. If either one of them or both of them are false, then the
conjunction would be false. As there is no such thing as half true in mathematics, we must accept the
conjunction as false even when one of its constituents is true and the other false.
(iii) Disjunction: The third way of forming new statements out of old ones is by taking the disjunction
of two statements. For this, merely put the word or between the statements. Thus the disjunction of
John is intelligent and There is life on Mars is John is intelligent or there is life on Mars. Sometimes
we put the word either before the first statement to make the disjunction sound nice, but it is not
necessary to do so, so far as a logician is concerned. The symbolic notation for disjunction is (read as
wedge or or). If p, q are two statements their disjunction is denoted by p q.
A word of warning regarding the use of or is very important. In practice, when one says I shall
spend my vacation either in Bombay or in Pune it is generally implied that the person will not spend it
both in Bombay and in Pune. Indeed, often the very nature of the two statements is such that they
cannot hold simultaneously, for example the statement, This book is either red or green. Occasionally,
both the possibilities are also implied together. In such a case the words or both are added to emphasize
it as in the statement A person with such a handwriting must be either a doctor or a crook or both. In
logic or in mathematics, however, it is not necessary to specify or both. The word or as it is used in
mathematics always implies either one or both the two alternatives. This is known as the inclusive
sense of the word or. When only one of the possibilities is intended and the simultaneous holding of
both of them is not automatically precluded by the context, it is explicitly stated by but not both. For
example consider the statement, For each x, we take either the positive square-root or the negative
square-root but not both. In symbolic terms, p q always means either p or q (or both). The logical
expression for either p or q but not both would be (p q) ( (p q)) or equivalently (p q)
(( p) q).
Another word of caution is necessary when we come to statements about members of a class. Note
that the statement John is either a man or a woman. can be taken to be the same as the disjunction of
the two statements John is a man and John is a woman. However, the statement Every human being
is either a man or a woman is definitely not the same thing as saying, Either every human being is a
man or every human being is a woman. The essential difference is that unlike the first statement, the
second one is about members of a whole class, in this case the class of human beings. Although for
each individual member of the class one of the two possibilities holds, it does not follow that the same
possibility should hold for all members. Quite a few cases of logical fallacies arise because of failure to
appreciate this point. A similar observation holds in the case of the conjunction of two existence
statements, as would be clear from one of the exercises at the end of this section (Exercise 2.4).
In view of the inclusive sense of the word or, it should be obvious that a disjunction p q is true
when at least one of the two statements is true and false when both p and q are false.
 6 INTRODUCTION TO GENERAL TOPOLOGY

What is said so far about truth values can be conveniently represented by a device known as truth
tables. In algebra we consider functions, such as sin x, x2 y2, x tan (yz) etc. of one or several variables.
We then prepare tables showing the values of these expressions for various values of the variables x, y,
z, etc. Similarly, we can regard statements p, p q, p (q r), etc. as functions (or formulas)
of the variables p, q, r, etc. and express their truth values corresponding to various possibilities that may
arise by assigning different truth values to p, q, r, etc. Actually, the things are much simpler here
because an algebraic variable x, y or z can assume infinitely many values, but each variable p, q, r, etc.
assumes only two truth values, 1 or 0. This means that if there are n independent variables in our
expression, then the complete truth table for it would need 2n rows. The following is a truth table for the
statement p (q r) of three variables p, q, r.
Row p q r r q r p (q r)

1 1
2 1
3 1
4 1
5 0
6 0
7 0
8 0
1 1 0 1 1
1 0 1 1 1
0 1 0 0 0
0 0 1 1 1
1 1 0 1 0
1 0 1 1 0
0 1 0 0 0
0 0 1 1 0

The columns for r and q r are auxiliary columns. The reader is urged to check that the table
is correct. As an example in the first row p, q, r are all true (i.e. have truth value 1). So r is false. But
q r is true and finally p (q r) is true.
When two statements are obtained as functions of the same variables, they are said to be logically
equivalent if they have the same truth values for all possible combinations of truth values of these
variables (which in terms of truth tables means that the columns under them are identical). As a simple
example consider ( p) where p is any statement. Logically, this is the negation of the negation of p.
Note that it is not quite the same as p on the nose although in practice we would like to identify it with p.
We can do so because it is easy to verify that ( p) and p are logically equivalent. As less trivial
examples the reader is urged to verify the equivalence between p (q r) and (p q) ( p r)
and between (p q) ( (p q) and (p q) (( p) q). Note also that the statement p p
is always true and p p is always false.A statement which is always true is called a tautology
while a statement which is always false is called a contradiction.
An interesting thing happens when we take the negation of a conjunction or a disjunction. For
example let p be the statement John is rich and q the statement John is intelligent. Then p q is the
statement John is either rich or intelligent. The negation of this, that is to say that it is false, amounts
to saying that John is neither rich nor intelligent. But that means that the negations of John is rich and
of John is intelligent are both true. Thus, we expect that (p q) should be logically equivalent to
( p ) (q ). The reader can indeed verify this to be the case by writing the truth tables for both.
The perceptive reader must have been reminded of the well known De Morgans laws in set theory at
this point. Indeed, most likely he has already noted some similarity between negation and complements,
between conjunction and intersection of sets and between disjunction and union of sets. Such a similarity
is not coincidental. It is not hard to explain its full significance if one studies what is known as a
truth set of a statement. This is the bridge needed to go from statements to sets. However, we do not
pursue this line further.
 LOGICAL WARM-UP 7

EXERCISES

2.1 Write down the logical negations of all examples of statements given in the last section (except
statements involving if ... then ...).
2.2 Verify that the statements in each of the following pairs are logically equivalent. Also, in each
case explain why the equivalence is obvious or reasonable to expect.
(i) p (q r) and (p q) (p r)
(ii) (p q) ( (p q)) and (p q) (( p) q)
(iii) (p q) and ( p) ( q).
2.3 We can define the conjunction of three statements p, q and r either as p (q r) or as
(pq)r. Are these the same? Are they logically equivalent? Justify your answers.
2.4 Suppose X has three friends, A, B and C who are, respectively, good at cricket, music and
mountaineering but not at any other fields. What is the fallacy in the following? The statement
One of Xs friends is a cricketer is true and so is the statement One of Xs friends is a musician.
So their conjunction is true, that is, the statement One of Xs friends is both a cricketer and a
musician is true. But X has no such friend.

3. LOGICAL IMPLICATIONS
As far as mathematics is concerned, the most important aspect of logic is the statements of the form If
then . These are called implications. Theorems in mathematics are commonly expressed as
statements of this form. For example the theorem that if two triangles are congruent then they are also
similar. Some theorems do not appear in this form on the surface but can easily be paraphrased into it.
For example the theorem The sum of the three angles of any triangle equals 180 degrees can be put as
If ABC is a triangle then A +B + C = 180 degrees. The importance of implication statements in
mathematics will be discussed in detail a little later. But let us first get down to the precise meaning of
such statements. They bear the form. If p then q where p and q are some statements. This statement is
also read as p implies q and also written as p q or p q. Such a statement is called a logical
implication. The statements p and q are called, respectively, the hypothesis and the conclusion of the
implication. The meaning of p q is that whenever p is true, q is also true, or in other words that the
truth of p ensures (or forces) the truth of q. But it says nothing in the contingency where p is false. In
particular, it does not say that if p is false then q is false. This is a very vital point because in practice we
tend to attach this extra meaning to implication statements. For example when a person says to his
friend If Monday is a holiday then I shall attend your wedding, in practice, we think that if Monday is
not a holiday then the person will not attend the wedding. To a mathematician, however, the present
statement is completely silent as far as the case of Monday not being a holiday is concerned. In such a
case whether the person attends the wedding or not, he is not violating his commitment. If the person
had something in mind in the event that Monday was not a holiday, he should have made it explicit by
another implication statement. For example if he had said, I shall attend your wedding if Monday is a
holiday, but not otherwise, or I shall attend your wedding if, and only if, Monday is a holiday then
this amounts to making two separate statements, If Monday is a holiday then I shall attend your wedding
and If Monday is not a holiday then I shall not attend your wedding.
The logic employed here may seem strange in the beginning, but we have already come across
something similar to it when we discussed vacuously true statements in Section 1. Recall that the
statement there was Every man with six legs is intelligent. This can be given the form of an implication
statement by If a man has six legs then he is intelligent. So, if we consider a man who does not have
 8 INTRODUCTION TO GENERAL TOPOLOGY

six legs, then whether he is intelligent or not, the statement is true in this case, because it is inapplicable
and hence cannot be shown to be false. The only way the implication statement can be false is if there
is a six-legged man who is not intelligent. Since there is no such man, the statement is true vacuously.
In general, an implication p q is considered to be true whenever p is false, that is, whenever the
hypothesis is not satisfied. When p is true, the truth value of p q will depend upon whether q is also
true or not.
With this rule in mind it is easy to construct the following truth table for p q:
Row p q pq
1 1 1 1
2 1 0 0
3 0 1 1
4 0 0 1

If we construct the truth table for ( p)q we see that it is identical to the one above. Therefore,
p q is logically equivalent to ( p)q. Indeed, this is often taken as the very definition of p q.
With this approach, however, it becomes necessary to explain why the statement either q holds or else
p fails conforms to our intuitive understanding of logical implication.
Using truth tables, the reader can verify that p q is logically equivalent to ( q) ( p).
Verbally, this means that p implies q is equivalent to saying if q fails then p fails. This certainly
sounds very reasonable. Indeed, this is the ,very basis of the so-called proof by contradiction or a
reductio ad absurdum argument. We use this type of argument in everyday practice too. Take for example
the statement, If it rains the streets get wet. If the streets are not wet, we conclude on the basis of this
statement that it did not rain. The statement ( q) ( p) is said to be the contrapositive of the
statement p q. Both are logically equivalent to each other.
The contrapositive of an implication statement should not be confused with its converse. The converse
of p q is defined to be the statement q p. In other words, the hypothesis of the original implication
statement is the conclusion of the converse and vice versa. For example the converse of the statement.
If two triangles are congruent then they are similar would be If two triangles are similar then they are
congruent. There is no correlation between the truth values of an implication and its converse. Numerous
examples can be given where both are true, both arc false or where one is true and the other false.
Sometimes, in the statement p q, the hypothesis, that is, p, is itself the negation of some statement,
as for example, the statement If it does not rain the crops will die. In such a case it is customary to
replace the phrase if not by the single word unless. With this change, the present statement would
become Unless it rains, the crops will die. We warn once again that this statement says nothing
whatsoever about the survival of the crops in the event it does rain. Here again, a logician differs from
a layman who would interpret this present statement to mean that if it rains crops will be saved. The
safest way to correctly interpret statements involving unless is to substitute for it if not.
In view of the immense importance of implication statements in mathematics, let us consider some
other ways of paraphrasing them. Suppose p and q are any statements. Then p q can be read in any
of the following ways:
(i) p implies q.
(ii) q follows from p.
(iii) q is a (logical) consequence of p.
(iv) If p is true then q is true.
(v) If q is false then p is false.
 LOGICAL WARM-UP 9

(vi) p is false unless q holds.

(vii) p is a sufficient condition for q.
(viii) q is a necessary condition for p.
(ix) p is true only if q is true.
Item (i) is just the definition, while (ii), (iii) and (iv) are its paraphrases. As we have seen before, (v)
is the contrapositive of (i) and (vi) a rephrasing of (v). The last three are the only versions which call for
a comment. Of these (vii) is fairly straightforward. For example to say If it rains the streets get wet
clearly amounts to saying that Raining is a sufficient condition for the streets to get wet, or that In
order that the streets get wet, it suffices if it rains. Thus, the use of the word sufficient here conforms
to its ordinary meaning.
It is a little confusing to use version (vii) in the case of some statements. For example in the example
just given, the statement would read Wetting of streets is a necessary condition for it to rain. This
sounds absurd. The trouble is with the word condition. In practice, it has the connotation of a
prerequisite, that is, something which is to exist prior to the happening of some event. In the present
case the question of streets getting wet arises only after the rain and that is why it is hard to swallow that
wetting of streets is a necessary condition for it to rain. Perhaps, another example would clarify the
situation. Consider the statement, If two triangles are congruent then they are similar. This means that
in order that two triangles be congruent they must at least be similar to each other. Congruency can
never occur if similarity does not hold. In other words, similarity of the triangles is a necessary condition
for them to be congruent. Whether it is sufficient or not is not the concern of the statement, it is the
business of the converse statement. Necessity and sufficiency should never be confused with each
other. In a sense, they are converse to each other.
About the last version, p is true only if q is true it is once again necessary to distinguish a layman
from a logician. When a layman says I shall come only if I am free, he generally means that he will
come if he is free but not otherwise. Alogician, however, makes no such commitment when he makes
the same statement. All he is saying is that his being free is a necessary condition for his coming, that is,
his coming will be impossible if he is not free. He is saying nothing at all as to what he will do if he is
free. Here, too, it is vital to distinguish between if and only if.
There is one exception to the preceding remarks. When something is defined in terms of a condition,
it is customary to cite this condition as sufficient, even though it is in fact sufficient as well as necessary.
Thus, when we say A triangle is called equilateral if all its sides are equal, it also means that a
triangle all whose sides are not equal will not be called equilateral. In other words, here if means if
and only if. This usage is unfortunate but very standard. Fortunately, it appears exclusively in definitions
and nowhere else.
In mathematics it often happens that we combine together an implication statement along with its
converse. For example take the well-known theorem, The sum of opposite angles in a cyclic quadrilateral
is 180 degrees and conversely. If we let p be the statement ABCD is a cyclic quadrilateral and q the
statement A + C = 180 degrees then the statement of the theorem is the conjunction of p q and
q p, that is, the statement (p q)(q p) or (p q)(q p). These types of statements come
up so frequently that it is convenient to have a shorter notation for them. The most natural choice is to
use arrows in both directions, that is, to use pq or p q. Here again it is convenient to list down a
number of versions of this statement.
(i) p and q imply each other.
(ii) p and q are equivalent to each other.
 10 INTRODUCTION TO GENERAL TOPOLOGY

(iii) p holds if and only if q holds.

(iv) q is a characterisation of p. (This version is generally used only when p, q express some
properties of the same object.)
(v) q holds if p does and conversely.
(vi) q holds if p does, but not otherwise.
(vii) if p is true then q is true and if p is false so is q.
(viii) q is a necessary as well as a sufficient condition for p.
Of course many other formulations are possible in view of the symmetry of p and q. Such statements
are called if and only if statements. The expression if and only if appears so often in mathematics
that it is customary to abbreviate it to iff. Thus, the geometric theorem quoted above can be stated as
A quadrilateral is cyclic iff the sum of its opposite angles is 180 degrees.
A theorem of this sort is really equivalent to two separate theorems which are converses of each
other. If we write the statement symbolically as p q (or as p q) then the implication p q is called
the direct implication or the only if part of the theorem while the other way implication, q p is
called converse implication or the if part of the theorem. In general, separate proofs are needed for
both the parts. Occasionally, it so happens that the steps used in the proof of the direct implication are
all reversible. In such a case, the converse is said to follow by reversing the proof of the direct implication.
It is by no means the case that both the implications are of the same degree of difficulty. There are many
theorems in which one of the implications is simple almost to the point of being trivial, but the other
way implication is fairly involved. As an example take the well-known remainder theorem which states,
Let f (x) be a polynomial in the variable x. Then a real number b is a root of f (i.e. f(b) = 0) iff f (x b)
is a factor of f (x). In this case, the if part is trivial, but the only if part is fairly hard to prove.
The concept of implication leads naturally to that of comparison of relative strengths of statements.
In practice, we say that a certain statement or piece of information is stronger than another if the
knowledge of the former subsumes knowledge of the other. For example we say it is stronger to say that
a certain person lives in Kerala than to say that he lives in India. This is so because anyone can infer the
latter from the former by sheer commonsense, provided of course, that he knows that Kerala is a part of
India.
Mathematically, we say that a statement p is stronger than a statement q (or that q is weaker than p)
if the implication statement p q is true. Afew comments are in order. First of all, stronger does not
necessarily mean strictly stronger. Note for example that every statement is stronger than itself. The
apparent paradox here is purely linguistical. If we want to avoid it we should replace the word stronger
by the phrase stronger than or possibly as strong as. However, the use of the word stronger in this
context is fairly standard. If p q is true but its converse is false, then we say that p is strictly stronger
than q (or that q is strictly weaker than p). For example it is strictly stronger to say that a given
quadrilateral is a rhombus than to say it is a parallelogram. Second, given two statements p and q it may
happen that neither is stronger than the other. Indeed, the two statements may not be related at all. In
such a case we say that their strengths are not comparable to each other. For example the statement
ABCD is a rectangle and the statement `ABCD is a rhombus. The word sharper is used sometimes
for stronger. This usage is common when the two statements deal with estimates or approximation of
22
something. For example, 3.1416 is a sharper statement than p .
7
What happens if out of two statements, each is stronger than the other? As we have already noted, in
such a case we say that the two statements have the same (or equal) strength or that they are (mutually)
equivalent. For example the statement, ABCD is a cyclic quadrilateral and the statement, ABCD is a
quadrilateral in which A + C = 180 degrees are equivalent to each other. In the last section we