Symbolic Logic

An Introduction

Geoffrey Bagwell, Ph.D.

Faculty of Philosophy
Spokane Community College

Spring 2023
 i

This work is licensed under the Creative Commons Attribution-NonCommercial-

ShareAlike 4.0 International License. To view a copy of this license, visit
http://creativecommons.org/licenses/by-nc-sa/4.0/.
Acknowledgements

This book began as a set of handouts for students in my symbolic logic course in
the fall term of 2015 at Spokane Community College. I want to thank them and
students in subsequent sections of this course for being subjected to an exper-
iment which at times probably caused needless confusion. I greatly appreciate
your patience and value your feedback.
The handouts and the book into which they transformed were inspired by
many of the introductory textbooks I have used in my courses over the years,
particularly Patrick Hurley’s A Concise Introduction to Logic (Cengage, 2011),
Paul Herrick’s An Introduction to Logic (Oxford, 2012), Paul Teller’s A Modern
Formal Logic Primer, Wilfrid Hodges’s Logic (Oxford, 2007), and Volker Hal-
bach’s The Logic Manual (Oxford 2010). I want especially to thank Paul Teller
for making his excellent textbook freely available and P.D. Magnus whose own
open-source textbook For all X: An Introduction to Formal Logic has arguably
done more for the teaching of symbolic logic than any other textbook on the
market. Their generosity inspired me to make this textbook freely available to
anyone wishing to learn from it.
No doubt I owe a debt of gratitude as well to the numerous other texts in logic
and philosophy of language that I have studied over my years as a philosophy
student and teacher, but these are too numerous to list and I hope that my
gratitude to all of those philosophers who have contributed to the development
of logic over the past century and a half will suffice. I am very grateful.
Finally, I want to thank my wife, Jenny, who is always hounding me to ex-
plain my reasoning, which I often mistakenly take to be obvious. This textbook
is dedicated to her.

ii
Preface

This book is an introduction to symbolic logic, sometimes called ‘formal logic’.

Logic is the branch of philosophy that studies reasoning, especially correct and
incorrect forms of reasoning. In particular, symbolic logic is the branch of logic
which employs algebra to identify precisely logically valid and logically invalid
patterns of deductive reasoning.
There are various kinds of symbolic logic, which logicians typically separate
into two groups: classical and non-classical logics. These groups are defined by
the assumptions they make. Classical logic, for example, assumes that there are
only two truth-values a sentence may have “True” and “False”. This assump-
tion is called the “principle of bivalence”. Consequently, classical logic takes
as a law the principle of excluded middle and with it the elimination of double
negation (e.g. “I was not unaware of this text”) as well as the principle of
non-contradiction and with it the possibility of deductive explosion, that from a
logical contradiction, any claim can be proven. Classical logic also assumes that
there is always at least one object in a domain of discourse. Non-classical logics
reject one or more of these assumptions, adopting others with sometimes pecu-
liar consequences. An important non-classical logic called “intuitionist logic”,
for example, rejects the principle of excluded middle with the result that dou-
ble negations cannot be eliminated. I do not discuss non-classical logics in this
book, but I do discuss each of the assumptions classical logic makes.
Classical logic has a long history. It was invented by Aristotle who wrote the
first systematic treatment of the subject entitled Prior Analytics as well as sev-
eral other treatises on related topics.1 It was developed in a different direction
by the Stoic philosophers, particularly Chryssipus, and refined by the scholastic
philosophers of the middle ages before languishing during the Renaissance and
the early modern period. Meanwhile, on the other side of the world, Hindu and
Buddhist philosophers in India, who may or may not have had access to Aris-
totle’s writing, developed their own systems of classical logic called “Nyaya”.
During the European enlightenment, the German philosopher Immanual Kant
remarked that “Since Aristotle . . . logic has not been able to advance a single
step, and is thus to all appearance a closed and completed doctrine” (Critique
of Pure Reason B, viii) reflecting the fact that no developments in logic of any
1 These texts are Posterior Analytics, Categories, On Interpretation, Topics, Sophistical

Refutations.

iii
PREFACE iv

significance had occurred in Kant’s lifetime. By the late nineteenth century,

however, mathematicians such as George Boole, John Venn, and Augustus De
Morgan took an interest in the subject and made their own mark on it. But
it is not until the German mathematician and philosopher Gottlob Frege pub-
lished his Begriffsschrift that the study of logic became modern. Along with
Bertrand Russell and Ludwig Wittgenstein, Frege brought algebra to the study
of reasoning launching an entirely new phase of logical discovery that continues
to this day. The contributions of all these men and many others are included
in this book.
Unlike other textbooks in logic, I explicitly adopt a normative approach to
classical logic. What I mean by ‘normative’ is that I treat classical logic as
consisting of a set of norms or standards for the way a person ought to reason.
This position has two implications. First, from a normative point of view it
is possible for a person to reason well or badly rather than simply correctly or
incorrectly. Second a normative approach to classical logic means that people
can be held morally responsible for their failures and successes in reasoning
according to these standards. In other words, a person can be praised for
reasoning well or blamed for reasoning poorly. Good people reason well and
bad people reason poorly.
Such a position may seem harsh and unnecessary but it has a purpose. The
point of this book is to help students improve their ability to reason through
the use of symbolic logic. The most effective way to achieve this goal is to
approach classical logic from a normative point of view. Some philosophers
may find classical logic’s assumptions to be problematic and in some circles
these assumptions may even be controversial. But the overwhelming majority
of logicians and philosophers accept them as reasonable assumptions to make.
I do so as well.
Finally, I have intentionally adopted a writing style aimed at helping students
to understand the mechanics of logical notation and the evaluation of arguments.
Logicians and philosophers will find this writing style pedantic and imprecise.
I respectfully ask them to remember that this book is not for them and not for
people who find logic intuitive or easy. This book is for students who have no
knowledge of symbolic logic. It assumes only that its readers can read English
and are familiar with basic algebraic notation, particularly the use of variables
in mathematical equations. It assumes nothing else. So, excepting any serious
mistakes, I ask my philosophical and logical colleagues to please forgive my
redundant explanations and use of non-standard terminology. I appreciate your
patience.
Contents

Acknowledgements ii

Preface iii

1 Consistency 1
1.1 Stating Beliefs . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2 Truth and Falsehood . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3 Quotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.4 Counterexamples . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.5 Logical Consistency of Sentences . . . . . . . . . . . . . . . . . . 9
1.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

2 Arguments 11
2.1 Arguments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.2 Non-Arguments . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.3 Evaluating Arguments . . . . . . . . . . . . . . . . . . . . . . . . 24
2.4 Logical Validity . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.5 Counterexamples to Arguments . . . . . . . . . . . . . . . . . . . 29
2.6 Implicit Premises . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

3 Logical Form 36
3.1 Simple and Compound Sentences . . . . . . . . . . . . . . . . . . 37
3.2 Sentence Constants and Variables . . . . . . . . . . . . . . . . . . 38
3.3 Connectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.4 Formation Rules for the Logical Connectives . . . . . . . . . . . . 45
3.5 Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.6 Bracketing Conventions . . . . . . . . . . . . . . . . . . . . . . . 49
3.7 Sentence Logical Form . . . . . . . . . . . . . . . . . . . . . . . . 51
3.8 Sentence Logical Analysis . . . . . . . . . . . . . . . . . . . . . . 52
3.9 The Logical Form of Arguments . . . . . . . . . . . . . . . . . . . 55
3.10 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

v
CONTENTS vi

4 Truth-Functionality 57
4.1 Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
4.2 Truth-Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
4.3 Truth-Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
4.4 Building Truth-Tables . . . . . . . . . . . . . . . . . . . . . . . . 64
4.5 Contingency, Tautology, and Contradiction . . . . . . . . . . . . 66
4.6 Logical Consistency . . . . . . . . . . . . . . . . . . . . . . . . . 68
4.7 Logical Equivalence and Contradiction . . . . . . . . . . . . . . . 69
4.8 Arguments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
4.9 Counterexamples . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
4.10 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

5 Conditionals 74
5.1 Conditional Sentences . . . . . . . . . . . . . . . . . . . . . . . . 74
5.2 Necessary and Sufficient Conditions . . . . . . . . . . . . . . . . 75
5.3 The Arrow and Double Arrow . . . . . . . . . . . . . . . . . . . . 78
5.4 The Arrow and Necessary Conditions . . . . . . . . . . . . . . . . 80
5.5 The Double Arrow . . . . . . . . . . . . . . . . . . . . . . . . . . 82
5.6 Formation Rules for Arrow and Double Arrow . . . . . . . . . . . 83
5.7 The Truth-Conditions of Conditionals . . . . . . . . . . . . . . . 86
5.8 Truth-Tables for the Arrows . . . . . . . . . . . . . . . . . . . . . 87
5.9 Building Truth-Tables for Arguments . . . . . . . . . . . . . . . . 89
5.10 Logical Equivalence and the Double Arrow . . . . . . . . . . . . 92
5.11 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

6 Simple Sentences 95
6.1 Singular Sentences . . . . . . . . . . . . . . . . . . . . . . . . . . 97
6.2 Universal Sentences . . . . . . . . . . . . . . . . . . . . . . . . . . 99
6.3 Formulae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
6.4 Particular Sentences . . . . . . . . . . . . . . . . . . . . . . . . . 105
6.5 The Formation Rules for Quantifiers . . . . . . . . . . . . . . . . 109
6.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

7 Models 112
7.1 Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
7.2 Truth-Conditions of Singular Sentences . . . . . . . . . . . . . . 114
7.3 Truth-Conditions of Universal Sentences . . . . . . . . . . . . . . 114
7.4 Truth-Conditions of Particular Sentences . . . . . . . . . . . . . 116
7.5 Truth-Conditions of Compound Sentences . . . . . . . . . . . . . 119
7.6 Working with Models . . . . . . . . . . . . . . . . . . . . . . . . 121
7.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127

A Common Indicator Words 128

B The Truth-Conditions of the Logical Operators 129

CONTENTS vii

C Basic Truth-Tables 130

D Building Truth-Tables 131

E Glossary 132

Index 135
Chapter 1

Consistency

We have beliefs about many different kinds of things. We have religious beliefs,
scientific beliefs, political beliefs, moral beliefs, and common-sense beliefs.
Our beliefs are the basis of our actions. When we act, we act on the basis
of some belief we have. When I open a door, I believe that there is a door to
open and I believe that turning the knob will unlatch the door, allowing me to
pull it open. When I vote for a political candidate, I believe that the person is
running for office and is eligible to hold the office.
Because beliefs form the basis of our actions, it is important that our beliefs
be logically consistent. Two or more beliefs are logically consistent if—and only
if—it is possible for them to be true at the same time. For instance, my belief
that

I live in Spokane.

and my belief that

I work at Spokane Community College.

are logically consistent; it is possible for both of them to be true at the same
time. In fact, they are!
If, on the other hand, a person believes that

It is dangerous to run a red light.

and that

It is safe to run a red light.

then this person’s beliefs are logically inconsistent; it is not possible for running
a red light to be safe and dangerous at the same time.
Why is it a problem to have logically inconsistent beliefs? The problem
is that inconsistent beliefs make it impossible for people to trust one another.
Trust requires people to be able to form expectations of one another. In order

1
CHAPTER 1. CONSISTENCY 2

to trust a person, we must be able to form some expectation of what the person
will and will not do. Logical inconsistency erodes confidence and trust in others.
Consider a very ordinary example. When you get in your car with the
intention of driving to work, what do you expect of other drivers? You expect
them to obey the rules of the road and you trust that they won’t deliberately
cause an accident. If you couldn’t expect other drivers to follow the rules because
you thought that they might do so inconsistently, then I imagine you would not
bother to drive anywhere.
To be more specific, suppose you approach a traffic signal. When the light
turns green, I imagine that you would proceed through the intersection as people
normally do. If so, what would you be expecting of the other drives as you
proceed through the intersection. My guess is that you expect the other drivers
to stop for the red light at the intersection, allowing you to proceed safely. But
what if you thought that the other drivers might have an inconsistent set of
beliefs about traffic laws? What if you thought that other drivers inconsistently
believed that it is wrong to violate the traffic laws but also sometimes okay to
do so? What would you expect the other drivers to do in this situation? Would
you expect them to stop for the red light and allow you to proceed? Or would
you expect them to ignore the red light and speed through the intersection?
If the other drivers really did have an inconsistent set of beliefs about traffic
laws, then you would have just as good a reason to expect them to stop at the
light as proceed through it. The question, then, is what would you do? Would
you proceed through the intersection or would you stay put? How would you
make a decision?
We might of course ask the same question in reverse. How can the other
drivers predict what you are going to do? If your beliefs about the traffic laws are
inconsistent, then how can other drivers trust you to obey the traffic laws? They
would be in no position to form any expectation of your behavior if your beliefs
were inconsistent just as you would be in no position to form an expectation of
theirs if their beliefs were inconsistent.
Of course people do sometimes run red lights. But most of us think of this
as an exception to the rule. People generally don’t run red lights because they
believe that it is dangerous to do so. When someone does run a red light, we
assume that the person did it on accident, was not paying close attention, or had
a good reason for doing so. We typically don’t think that the person believes it
is generally safe to run red lights. Yet, if we did come to believe that someone
thinks inconsistently that it is both safe and dangerous to run a red light, then
we would not trust them to stop at a red light.
So, the problem with inconsistent beliefs is that they undermine trust and
we need to be able to trust one another in order to act and in order to live our
lives.
There is another problem with inconsistent beliefs: they can justify any
course of action. Suppose a driver runs a red light and hits your car. For-
tunately, both you and the driver are uninjured. When the police arrive to
write up a report of the accident, you find out that the driver who hit you
believes—inconsistently—that it is dangerous to run red lights but safe to do
CHAPTER 1. CONSISTENCY 3

so in this case. “Why”, you ask him, “did you think it was safe to run the red
light?” He responds: “Because we didn’t get hurt.”
If the driver’s response in my examples puzzles you, it should. His response
justifies his action. It was safe to run the red light because no one was hurt.
However, this justification offers no consolation to you because you know that
someone could have been hurt. The fact that no one was doesn’t change the
fact that running a red light is dangerous. And the driver knows this. So, you
are left with nothing but his inconsistent beliefs that it is both dangerous and
safe to run a red light.
What if one of you had been hurt in the accident? The driver would likely
have admitted that running the light was a bad idea or perhaps that he would
have accused you of running the light instead of him. Either way the driver
would be trying to justify this action and his justification will succeed because
he would have erroneously believed that it was safe to proceed when in fact it
wasn’t or he would have correctly believed that it is was dangerous to proceed
through the intersection.
Suppose, on the other hand, that the driver had stopped for the red light.
You, of course, would have proceeded through the intersection without incident.
But imagine that you had an opportunity to ask the driver why he stopped for
the red light. He would have said that running a red light is dangerous and it
would be right. So, his belief that running the red light is dangerous justifies
his stopping.
In both of these situations, the driver is able to justify whichever course of
action he performs because his beliefs are logically inconsistent. If he runs the
light, then his belief that it is safe to do so justifies his action. If he stops for
the light, then his belief that it is dangerous to run the light justifies his action.
For this and the previous reason, it is important to have consistent beliefs.
Our lives and our actions are based on our beliefs and our beliefs form the basis
of other people’s expectations of us. In this chapter, I explain how you can find
out whether your beliefs are consistent. By the end of this chapter, you will be
able to determine whether two or more of your beliefs are consistent.

Chapter Objectives: By the end of this chapter, you will be able to

• Distinguish between sentences and other kinds of speech
• Determine when a sentence is true and when a sentence is false.

• Come up with a counterexample that proves that a false sentence is

false.
• Determine whether two or more sentences are logically consistent
CHAPTER 1. CONSISTENCY 4

1.1 Stating Beliefs

When we wish to express our beliefs and share them with others, we state them.
When we state our beliefs, we put them in the form of a complete sentence. For
example, I believe that it is wrong to commit murder. Therefore, I will state
my belief with the sentence:

Murder is wrong.

This example is a complete English sentence. A complete sentence is

Definition (Complete Sentence). A complete sentence is an expression that
has both a subject and a predicate.
The subject of a complete sentence is what the sentence is about and the predi-
cate is what the sentence asserts about the subject. The subject of the example
is “murder” and predicate of the example is the phrase “. . . is wrong”. If I take
the word “Murder” out of the sentence, the result “. . . is wrong” is an incomplete
sentence and cannot express my belief.
Here is another example. If I believe that human beings have souls, then I
will state my belief that

Human beings have souls.

Again, this example is a complete sentence. It has both a subject and a pred-
icate. The subject is the common noun “human being” and the predicate is
the phrase “. . . have souls”. If the subject or the predicate of the sentence were
removed, the result would be incomplete and would not state my belief that
human beings have souls.
But not everything we say is a statement of belief. For instance, if I say

Don’t murder anyone!

I am not stating my belief that murder is wrong. Rather I’m ordering or com-
manding someone not to commit murder.
Similarly, if I say

Do human beings have souls?

I am not stating a belief. Instead, I am asking a question to find out whether

human beings have souls or not. In doing so I am implying that I do not yet
know whether human beings have a soul.
Besides commands and questions, we also offer suggestions, make proposals,
exclamations, etc. None of these types of speech state beliefs. Only declara-
tive sentences state beliefs. What distinguishes declarative sentences from other
types of speech? Unlike these other types of speech, a sentence has what logi-
cians describe as “truth-value”:
Definition (Truth-Value). The property of being true or being false.
CHAPTER 1. CONSISTENCY 5

A sentence, in other words, is the type of speech that can be either true or false
but not both.
Consider the examples above. The sentences

Murder is wrong.

and

Human beings have souls.

can either be true or false. In other words, it is either true or false that murder is
wrong and it is either true or false that human beings have souls. Both of these
types of speech have a truth-value. You may not know whether a particular
sentence is true or false, but it should be evident that each of the sentences
above can be true or false.
On the other hand, the examples

Don’t murder anyone!

and

Do human beings have souls?

cannot be true or false. None of them can have a truth-value. Consider the
second example, “Do human beings have souls?”. Imagine you ask someone
this question and the person responds “False”. This response makes no sense
and there is a good reason why: the example is a question and questions are not
declarative sentences; they cannot be true or false because they cannot have a
truth-value.
Because a declarative sentence can have a truth-value, questions, proposals,
suggestions, commands, or exclamations are not declarative sentences. Only
declarative sentences can have a truth-value.

1.2 Truth and Falsehood

When is a declarative sentence true and when is a declarative sentence false?
This is an old philosophical question and one that you might think is quite
controversial. In fact, it is not. Philosophers, scientists, and logicians have long
agreed that Plato’s answer is essentially correct. According to Plato,

Those sentences that speak of the things that are as they are, are
true, while those sentences that speak of the things that are as they
are not, are false. (Cratylus, 385b, my translation)

Here, Plato articulates a theory of truth known as the “correspondence theory”:

Definition (The Correspondence Theory of Truth). A sentence is true if—and
only if—it corresponds to some fact about the world.
CHAPTER 1. CONSISTENCY 6

If a sentence does not correspond to any fact about the world, then it is false.
When I say, for example, that
It is raining in Spokane today.
this sentence is true when it is—in fact—raining in Spokane today. If, on the
other hand, it is sunny in Spokane today, then the sentence is false.
This definition of truth implies the truth-value of a sentence depends on
certain conditions. These conditions are called the “truth-conditions” of a sen-
tence and, if met, they are what makes a sentence true. For example, the
truth-conditions for the sentence
Some apples are red.
are that there are things called “apples” and that some of these things have
the property of having a red color. If these two conditions are met, then the
sentence is true. If in fact there are things called “apples” and if in fact some of
these things have the property of being red, then the sentence “Some apples are
red” is true. However, if one or more of these conditions is not met, then the
sentence will be false. If—in fact—there are no such things called “apples” or
if—in fact—no apples have the property of being red, then the sentence “some
apples are red” is false.
Because all sentences—by definition—have a truth-value, all sentences have
truth-conditions. Both true and false sentences have truth-conditions. False
sentences just happened to have truth-conditions that are not met. This is
another way of saying that there is no fact in the world that makes the sentence
true. The sentence
All human beings lay eggs.
is false because its truth-conditions are not met. The truth-conditions for this
sentence are that there are things called “human beings” and things called
“eggs” and human beings lay eggs. But, in fact, no human beings lays eggs. So
the truth-conditions for this sentence are not met, making the sentence false.
Not every philosopher accepts the correspondence theory of truth as the cor-
rect theory of truth. You might have wondered, for instance, what a fact is and
what it means for a sentence to “correspond” with a fact. These questions exer-
cise the minds of philosophers as well and have led some in search of alternative
theories. Most philosophers, on the other hand, have accepted the correspon-
dence theory without much question and almost all contemporary philosophers
accept it in one way or another. Because of its provenance and because of its
near universal appeal, it is the theory of truth that we will be working with in
this book.

1.3 Quotation
It is not easy to talk about facts about the world without using sentences. Think
about it. When you want to tell someone about an event that has occurred in
CHAPTER 1. CONSISTENCY 7

your life, you use sentences to talk about it. If the person to whom you are
speaking doubts what you are saying, then, according to the correspondence
theory of truth, he doubts whether the sentences you have used to talk about
the event correspond to any facts in the world. How do you, then, explain to
him what the facts are without using sentence so that he can see that you are
telling him the truth? As strange as it sounds, you will have to use sentences
to talk about sentences. This is very confusing.
Logicians also find it confusing to use sentences to talk about sentences.
Consequently, logicians have adopted quotation marks as a device to separate
sentences about facts from sentences about sentences. Quotation marks allow
logicians to distinguish between using a sentence and talking about one. They
describe the difference as the distinction between using words, phrases, or sen-
tences and mentioning a word, phrase, or sentence:
Definition (Mention). To mention a word, phrase, or sentence is to talk about
it.
Logicians use quotation marks to indicate when they are talking about or “men-
tioning” a word, phrase, or sentence. They also sometimes center a word, phrase,
or sentence in the middle of a printed page to indicate that they are talking about
it. In the previous sections, I have done this when I needed to indicate that I
am speaking about certain words or sentences. When I did so, I indicated that I
was talking about those words or sentences by marking them off with quotation
marks or by centering them on the page. I was, in other words, mentioning
those sentences.
To use a word, phrase, or sentence is to do just that. When logicians use
such expressions, they don’t do anything special to indicate this use. I have
been using every word, phrase, or sentence in the previous sections except for
the sentences I mentioned and I am using words right now.
Why all the fuss? This distinction between use and mention may seem to
be pointless, but it is not, because it helps us to see what the truth-conditions
of a sentence are. Let me give you an example to illustrate why the distinction
matters. The sentence

Grass is green

is true because it is a fact that grass is green; yet the sentence

“Grass” is green.

is false. Why is one sentence true but the other false? The reason is because
there is a difference between using a word and mentioning a word. The first
sentence uses the word “Grass” to say that it is green. The second sentence
mentions the word “Grass” to say that the word “Grass” is green. Does the
second sentence make any sense? How can the word “Grass” be green?
Consider another slightly different example. The sentence

“Grass” consists five letters.

CHAPTER 1. CONSISTENCY 8

is true because it mentions the word “Grass”, that is, it is talking about the
word—not about the plant to which the word refers. On the other hand, the
sentence
Grass consists of five letters.
is false and really makes no sense because it uses the word “Grass” to refer to
the plant. Grass does not consist of any number of letters; it is a plant. The
word “Grass”, however, does consist of five letter because it is a word. Without
quotation marks which separate when I am mentioning a word from when I am
using it, there is no easy way to tell when “grass” refers to the word or the
plant.
Hopefully these examples illustrate what all the fuss is about and why I
needed to explain the difference between using a word and mentioning it. In
this rest of this textbook, I rely on this distinction between using a sentence
and mentioning a sentence to clarify when I am using a sentence and when I
am talking about one.

1.4 Counterexamples
If a sentence is false, then it is possible to show that it is false by giving an
example. Examples that describe a situation in which a sentence is false are
called “counterexamples”:
Definition (Counterexample). An example which describes a situation in which
a sentence is false.
Counterexamples are useful because they quickly prove that false sentences are
false. For example, a counterexample to the claim that
Chryssippus invented logic.
is Aristotle. The Greek Stoic philosopher and logician, Chrysippus, was born
in 279 B.C.E. However, the well-known Greek philosopher and logician, Aris-
totle, died in 322 B.C.E., a full fifty-years before Chrysippus was born. Both
men made substantial contributions to the development of logic, much of which
you will learn in this book. But Aristotle presented his contributions before
Chryssippus was born. So, Aristotle is a counterexample to the sentence that
Chryssippus invented logic. It shows that the sentence is false.
Here is another example. A counterexample to the sentence
All swans are white.
is that there are black swans in parts of Australia. Before the landing on Aus-
tralia in 1606 and observing black swans, Europeans commonly believed that
all swans are white. But this belief turned out to be false because black swans
are a counterexample to it.
If you suspect that a claim is false, finding a counterexample to it is one way
to disprove the sentence quickly.
CHAPTER 1. CONSISTENCY 9

1.5 Logical Consistency of Sentences

At the beginning of this chapter, I said that a set of beliefs is logically consistent
if and only if the beliefs can be true at the same time. In terms of sentences,
we can define logical consistency:
Definition (Logical Consistency). Two or more sentences are logically consis-
tent if and only if they can be true at the same time.
From this definition, we can define logical inconsistency as two or more sentences
that cannot be true at the same time. Accordingly, these two sentences are
logically inconsistent:

1. Aristotle is a logician.
2. Aristotle is not a logician.

These sentences cannot be true at the same time. Thus, they are logically
inconsistent according to the definition of logical consistency.
Here is a harder example:

1. There is no housing shortage in Spokane.

2. There is just a rumor of a housing shortage spread by people who have
nowhere to live.

How are both of them true at the same time? Ask yourself: do the people
who have nowhere to live actually live in Spokane? If they do not, then they
are just spreading a rumor and there may be no housing shortage in Spokane.
In other words, both sentences are true at the same time. If these people do
live in Spokane, then they are not spreading a rumor which means that the
second sentence is false. But so is the first. If these sentences are not logically
consistent, then one of them can’t be true unless the other is false. Yet, in the
case where the people who have nowhere to live do not live in Spokane, both
sentences could be true; and in the case where they do live in Spokane both
sentences are false. Thus in no cases is one sentence true and the other sentence
false.
Here is one more example:

1. Geoffrey joined the Spokane club two years ago, and has been one of its
most loyal members ever since.
2. Last year he paid for the vacations of precisely those club members who
didn’t pay for their own vacations.

According to the definition of logical consistency, these sentences are logically

inconsistent. Why? Is Geoffrey a member of the Spokane club or not? If the
first sentence is true, then Geoffrey is a a member. But, if he is, then the
second sentence is false because it says that he only paid for the vacations of
CHAPTER 1. CONSISTENCY 10

club members who didn’t pay. Did Geoffrey pay for his own or not? If he did,
then he couldn’t have paid for the vacations of only those members who didn’t
because he did and the second sentence is false. If he didn’t pay for his vacation,
then he didn’t pay for the vacations of those club members who didn’t because
he didn’t pay for his own. So, again, the second sentence is false. In short, if
the first sentence is true, then the second is false and together they are logically
inconsistent.
Now, suppose that the second sentence is true. If Geoffrey paid for the
vacations of only those members of the Spokane Club that did not pay for their
own, then he cannot be a member of the Spokane Club. But, if the first sentence
is true, then he is a member. So, if the second sentence is true, then the first is
false and, again, they are logically inconsistent.

1.6 Summary
In this chapter, I have introduced you to the fundamental logical concepts of
logical consistency. As I have defined this concept, two or more sentences are
consistent if and only if they can be true at the same time. Two or more
sentences are inconsistent, on the other hand, if—and only if—they cannot be
true at the same time. Sentences have a unique characteristic called “truth-
value”. Whether a sentence is true or not depends on whether it corresponds
to some fact about the world. Sentences that do not correspond to any fact are
false.
Chapter 2

Arguments

When we reason, we often pass through a series of thoughts on our way to

some conclusion. Each step in the chain of reasoning we do is an argument.
Arguments are not fights, disputes, or feuds between people. They are efforts
to prove a point. Here is a famous example:
Different cultures disagree about what is right and what is wrong.
So, right and wrong are relative to each culture.
Philosophers call this argument “the standard argument for cultural relativism”.
Cultural relativism is the theory that right and wrong depend on the beliefs of
each culture.
Is this argument good or is it bad? The answer to this question depends
on whether the premises of the argument justify its conclusion. What do you
think? It may not be obvious but the standard argument for cultural relativism
is a bad argument. The premise does not justify the conclusion.
You probably have doubts about this claim and that’s fine; many people
consider this argument to be a good one. So, here is brief explanation of why
the standard argument for cultural relativism is a bad argument. Suppose the
argument is good. If it is a good argument, then the premise should give you a
reason to believe the conclusion. In other words, if the premise is true, then the
conclusion of the argument should not be false. What I mean is that if different
cultures disagree about what is right and what is wrong, then it cannot be
that right and wrong are independent of each culture. But consider this: is
it possible that different cultures could disagree about what is right and what
is wrong when what is right and wrong are not relative to each individual or
culture? Sure. It is possible for one culture to be correct about what is right
and what is wrong when other individuals and cultures are incorrect. I am not
saying which culture is correct about right and wrong and which one is not.
What I am saying is only that it is possible for one culture to be correct and
for another culture to be incorrect. This possibility indicates that the premise
of the argument could be true and the conclusion could be false. Consequently,
the premise does not justify the conclusion and, so, it is a bad argument.

11
CHAPTER 2. ARGUMENTS 12

You may still not be persuaded by my explanation because you do not yet
know what I know about what makes an argument good and what makes an
argument bad. In this chapter, I introduce you to arguments and some informal
methods of evaluating them. By the end of this chapter, you too will know what
makes an argument good and what makes an argument bad, and you will see
that this argument and others like it are not good.

Chapter Objectives:

• Identify the premises and the conclusion of an argument

• Determine whether an argument is logically valid or invalid.
• Come up with a counterexample that proves that an invalid argument
is invalid

• Formulate an implicit premises needed to make an invalid argument

valid.

2.1 Arguments
In chapter 1, I introduced you to the concept of logical consistency. According
to its definition, two or more beliefs are logically consistent if and only if they
can be true at the same time. A special case of logically consistent beliefs is a
logically valid argument. Here is a precise definition of an argument:
Definition (Argument). A set of sentences, one or more of which, the premises,
gives a person a reason to believe one and only one of the others, the conclusion

A premise in an argument is a sentence that is a reason to believe a conclusion,

which is another sentence. Here is the standard argument for cultural relativism
that I introduced at the beginning of this chapter:

Different cultures disagree about what is right and what is wrong.

So, right and wrong are relative to each culture.

What is the conclusion of this argument? The conclusion of this argument is

the sentence “Right and wrong are relative to each culture”. You might be
wondering how I know that this is the conclusion. I have many reasons why I
come to this conclusion, but two of them are that that conclusion of an argument
often appears at the beginning and end of the argument, and the word “so” is
an indicator word. In this case, the conclusion appears to be at the end because
people generally begin their arguments by stating their conclusion or end by
stating their conclusion. People also typically employ indicator words to draw
other people’s attention to the parts of their argument and help them to identify
those parts. Keep in mind that, if you and I cannot tell whether a sentence is
CHAPTER 2. ARGUMENTS 13

a conclusion or a premise of an argument, then you and I will not be able to

judge whether the argument is good or bad. So, it is very important for us to
be able to identify the conclusion and the premises of an argument; and it is
in the best interest of any person presenting an argument to make it easy for
other people to find them.

Identifying Arguments
Not all kinds of speech present arguments. In general, a speech or written
passage contains an argument if it gives a reason to believe some claim. If
speech does not give any reason to believe a claim, then it does not contain an
argument. More precisely, a speech or passage contains an argument if it gives
reasons which justify some conclusion.
To determine whether a speech or a passage contains an argument, you
should look for

• Words which indicate the presence of a conclusion or premise.

• An inference between the sentences in a speech or passage.

Common Indicator Words

In English we use certain words to indicate that sentences in an argument are
its premises or its conclusion. Here are some of the most common indicators
words:

Premise Indicators Conclusion Indicators

• “Since” • “Therefore”
• “As indicated by” • “Thus”
• “Because”
• “Consequently”
• “For”
• “We my infer”
• “In that”
• “Accordingly”
• “May be inferred from”
• “We may conclude”
• “As”
• “It must be that”
• “Given that”
• “For this reason”
• “Seeing that”
• “For the reason that” • “So”

• “In as much as” • “Entails that”

• “Owing to” • “Hence”

CHAPTER 2. ARGUMENTS 14

• “It follows that” • “As a result”

• “Implies that”

Since the conclusion of the standard argument for cultural relativism is

“Right and wrong are relative to each culture”, the other sentence “Different
cultures disagree about what is right and what is wrong” must be the premise.
Now that I have identified the parts of the argument, I can reconstruct it to
make it easier to see the parts:

Premise: Different cultures disagree about what is right and what is wrong.
Conclusion: Right and wrong are relative to each culture.

I have not changed very much here by labeling the premise “Premise” and the
conclusion “Conclusion”, I can now talk about how the premise supports the
conclusion by talking about the relationship between “Premise” and “Conclu-
sion” instead. I use this way of labelling the parts of an argument in the rest of
this book. When an argument has more than one premise, I will number each
premise separately.
Now, according to the definition in section 2.1, an argument is a set of
sentences. What this means is that arguments consist only of expressions which
can have a truth-value. This fact about arguments should make sense because an
argument tries to prove that something is true. Consider the standard argument
for cultural relativism:

Premise: Different cultures disagree about what is right and what is wrong.
Conclusion: Right and wrong are relative to each culture.

This argument attempts to proving that some sentence, the conclusion, is true
and it requires evidence to do so. Evidence must be true in order to prove that
some conclusion is true. And by using evidence to prove that some conclusion
is true, everything else in the argument must have a truth-value, it be able to
be true or false. If some part of the argument has no truth-value, then it has no
role to play in the argument; it cannot prove anything. In short, no expressions
other than sentences can ever be part of an argument.
Ordinarily, you might describe a fight, disputes, feuds, or disagreement be-
tween people as an argument. But philosophers disagree. What makes a con-
versation an argument is that the people involved in the conversation try to
justify their claims. They give people reasons to think that what they believe
or say is true. Fights, disputes, feuds, or disagreements also be arguments if
the people involved attempt to justify their claims. However, if no one involved
in a fight,dispute, feud, or disagreement tries to give any reason to think that
what they believe is true, then there is no argument.
CHAPTER 2. ARGUMENTS 15

2.2 Non-Arguments
When a person presents an argument, that person gives a reason for believing
that some conclusion is true. It is easy to understand what an argument is, but
is not always so easy to identify an argument in written passage or in a speech
because arguments are easily confused with other kinds of speech. To help you
learn to identify arguments, allow me to show you some of the kinds of speech
that are not arguments and do not contain any arguments. Here is a short list:

1. Warnings

2. Advice
3. Statements of Belief or Opinion
4. Loosely Associated Statements
5. Reports

6. Exposition
7. Illustrations
8. Explanations
9. Conditional Sentences

Warnings
Warnings are expressions that inform someone of potentially dangerous situa-
tions. For example:

Be careful not to slip on the ice!

Notice that the example above has no truth-value, that is, it can neither be true
nor false; it doesn’t make sense to ask whether it is true or false. The fact that
it has no truth-value, however, does not imply that it is meaningless. There is
a purpose behind the expression and the purpose is to warn someone to avoid
a potentially dangerous situation. Unlike an argument whose purpose is to
justify or prove a point, a warning like this one is not trying to convince anyone
that some conclusion is true. This is what distinguishes an argument from a
warning. To put it another way, the premises and conclusion of an argument
have a truth-value because they are sentences. Warnings are not arguments
because they have no truth-value.
Here is another another example:

What you do, never confide personal secrets to that gossipmonger.

CHAPTER 2. ARGUMENTS 16

Advice
When a person gives someone advice, it can sometime come across as an argu-
ment, especially if the person tries to justify giving the advice. When such a
person gives reason to accept the advice, they are presenting an argument. But
the advice given is not itself an argument. Pieces of advice are expressions that
recommend a decision or course of action. For example,
You should keep a few things in mind before buying a used car. Test
drive the car at varying speeds and conditions, examine the oil in
the crankcase, ask to see service records, and, if possible, have the
engine and power train checked by a mechanic.
The purpose of this example is to advise someone about the best way to buy a
used car. Notice that the tone of the person speaking is not to prove anything;
the person is not trying to persuade or convince another about the best way to
buy a used car. The person is simply making some recommendations.
Here is another example:
Before accepting a job after class hours, I would suggest that you
give careful consideration to you course load. Will you have sufficient
time to prepare for classes and tests, and will the job produce an
excessive drain on your energies?
Like the previous example, the point of this passage to give someone advice
about whether to accept a job or not. To help them make a decision about
whether to accept the position, the passage asks the person questions about
things to consider before making a decision. Nothing tries to justify a specific
decision or course of action. So, there is no argument.

Statements of Belief or Opinions

Unlike warnings and advice, statements of belief or opinions are sentences; they
do have a truth-value though it may not always be clear what their truth-value
is. Nevertheless, statements of belief and opinion are not arguments because
they merely express what a person believes or thinks without any supporting
reasons to justify that belief or thought. For example,
We believe that our company must develop and produce outstanding
products that will perform a great service or fulfill a need for our
customers. We believe that our business must be run at an adequate
profit and that the services and products we offer must be better
than those offered by competitors.
There is no argument in this example. Instead, there are two statements of
what people believe about their company. There reasons are given to justify
these beliefs. So, there is no argument.
Here is another example to reinforce the point that mere statements of belief
and opinion are not arguments:
CHAPTER 2. ARGUMENTS 17

When I can read the latte menu through the hole in my server’s
earlobe, something is seriously out of whack. What happened to an
earring, maybe two, in each lobe? Now any surface is game. Brow,
lip, tongue, cheek, nose. I’ve adjusted to untied shoelaces and pants
that make mooning irrelevant. But when it comes to piercings, I
just can’t budge (Debra Darvick, “Service with a Smile, and Plenty
of Metal”).

The author evidently doesn’t like piercings and unkempt clothing! She states
here opinion clearly but offers no justification for it. So, there is no argument
here; there is only the expression of an opinion.

Loosely related Sentences

Another type of speech that is often confused with arguments is what I call
“loosely related sentences”. Loosely related sentences are typically about the
same subject. Apart from this similarity, they have no other obvious relation-
ship. Here is an example:

Not to honor men of worth will keep the people from contention; not
to value goods that are hard to come by will keep them from theft;
not to display what is desirable will keep them from being unsettled
of mind’
(Laozi, Dao De Qing).

The sentences in this example are all related because they are all about the
subject of honor and value. However, there is no argument in the example
because there is no attempt to give an reason to believe any of the sentences
in it. The author, Laozi, does not try to justify or support his claim that
people won’t be tempted to stealing that are not regarded as valuable, nor does
he attempt to justify the claim that preventing conflict requires one to avoid
honoring worthy men; and he makes no effort to prove that people’s mind won’t
be disturbed if one does not show something something worth desiring. Clearly,
all three of his claims are related, but there is no argument.

Reports
Reporting information as happens in history textbooks and newspapers are also
often confused with arguments. Where arguments try to justify some claim,
reports share information about some issue or event. Here is an example of a
report from a widely-used political science textbook:

The period of 1648-1789 was one of competition among the primary

monarchs of Europe. Wars among the great powers were frequent
but limited. France made major efforts to become paramount, but
the balance of power operated to block French expansion
(Steven l. Spiegel, World Politics in a New Era).
CHAPTER 2. ARGUMENTS 18

In this example, no effort is make to justify anything. Instead, the author

presents information about competition among European monarchies during
the seventeenth and eighteen centuries.
Here is a example of a news report:

Witnesses said they heard a loud crack before a balcony gave way at
a popular night spot, dropping dozens of screaming people fourteen
feet. At least eighty people were injured at the Diamond Horseshoe
casino when they fell onto broken glass and splintered wood. Inves-
tigators are waiting for an engineer’s report on the deck’s occupancy
load.

This example shares information about a tragic accident. At no point does it

try to prove that the accident happened or even justify the explanation of why
it did. For this reason, the passage is no argument and presents no argument.
Here one last example of a report:

‘The Air Force faces a serious shortage of experienced pilots in the

years ahead, because repeated overseas tours and the allure of high
paying jobs with commercial airlines are winning out over lucrative
bonuses to stay in the service,’ says a prominent Air Force official.

Like the previous examples, this passage simply conveys information about what
an Air Force official said about shortage of qualified pilots in the Airforce. It
makes no attempt to justify what the official said or that what the official said
was true. So, it offers no argument in support of the officials testimony.

Exposition
Arguments can be easily confused with exposition. An exposition begins with a
sentence that introduces a topic some kind followed by one or more additional
sentence that provide details about that topic. The point of these additional
sentence is not to justify the sentence that introduces the topic, but to expand
and elaborate upon it. For example, this passage

‘There are three familiar states of matter: solid, liquid, and gas.
Solid objects ordinarily maintain their shape and volume regardless
of their location. A liquid occupies a definite volume, but assumes
the shape of the occupied portion of its container. A gas maintains
neither shape nor volume. It expands to fill completely whatever
containers it is in.’
(John W. Hill and Doris K. Kolb, Chemistry for Changing Times,
7th ed.)

starts by introducing solids, liquids, and gases, and, then, proceeds to describe
in more detail what each of these states of matter are like. The passage does
CHAPTER 2. ARGUMENTS 19

not try to show that there are only these states of matter or even that there are
three. It simply introduces the three and describes each one.
Here is a different example:

‘There is a stylized relation of artist to mass audience in sports, es-

pecially in baseball. Each player develops a style of his own—the
swagger as he steps to the plate, the unique windup a pitcher has,
the clear-swinging and hard-driving hits, the precision quickness and
grace of infield and outfield, the sense of surplus power behind what-
ever is done.’
(Max Lerner, America as a Civilization)

This passage introduces baseball as an example of a “stylized relation of artist

to mass audience” and goes on to explain what this means by describing the
batter, the pitcher, and the fielders as artists who make their behavior on the
field stylish for the people watching them. Though at first it may seem like the
passage argues that the player’s actions on the field are evidence of a stylized
relation of artist to mass audience, the passage does nothing more than provide
an exposition of how baseball involves a stylized relation of artists to mass
audience. If the passage strikes you as an argument, then you might ask yourself
whether you have doubts about the first sentence in the passage. If you doubt
that there is a stylized relation between artist and audience in baseball, then
the passage will feel to you like it is trying to justify itself when it is not. Does
the passage itself suggest that the first sentence in it is controversial? It does
not.

Illustration
Illustrates are sometimes thought to be arguments because, in some cases, ar-
guments use examples to justify their conclusions. Illustrations, however, are
not themselves arguments. So, it is vital to be able to distinguish illustrations
from arguments.
An illustration presents an example to show what something means or how
something is done. Here is an illustration from chemistry:

Chemical elements, as well as compounds, can be represented by

molecular formulas. Thus, oxygen is represented by “O2 ”, water by
“H2 O”, and sodium chloride by “NaCl”.

The purpose of this passage is to illustrate how chemists use molecular formulas
to represent elements and molecules. If the passage contained or presented an
argument, then it would give a reason to believe that molecular formulae do
represent atoms and molecules. However, no reasons are given in the passage
to justify this claim. So, there is no argument.
Here is another illustration:
CHAPTER 2. ARGUMENTS 20

A deciduous tree is any tree that loses its leaves during the win-
ter. For example, maples are deciduous. And so are elms, poplars,
hawthorns, and alders.

This passage is an illustration because it gives several examples of different trees

to show what a deciduous tree is. Like the previous illustration, there is no effort
here to prove or justify anything.
At the beginning of this section, I mentioned that illustrations are often
confused with arguments because arguments occasionally use illustrations to
prove a point. In this case, philosopher sometimes characterize such argument
as “arguments from example”. Consider the following:

Although most forms of cancer, if untreated, can cause death, not

all cancers are life threatening. For example, basal cell carcinoma,
the most common of all skin cancers, can produce disfigurement, but
it almost never results in death.

Despite the use of basal cell carcinoma as an illustration of a specific kind of

cancer, this passage also uses the illustration to justify its conclusion that not
all cancers are life threatening. Since cancer is commonly thought to be fatal,
there may be some doubt in you mind that not all cancers are life threatening.
Consequently, the passage gives you a reason to think that not all forms of
cancer are fatal. This reason given in this case is an illustration. Basal cell
carcinoma is an example of a non-life threatening form of cancer. So, not only
does this example illustrate one type of a cancer, it also justifies the conclusion
that not all types of cancer are life threatening.

Explanation
One of the most common forms of speech that people confuse for arguments are
explanations.
There are many good reasons why explanations and arguments are easily
confused. One reason is that both make use of the indicator word “because”.
People use “because” to explain something and they use “because” to justify
something. Another is good reason is that arguments and explanations have
parts which are similar. Where arguments have one conclusion and at least one
premise, explanations have an “explanandum” and at least one “explanans”.
The explanandum of an explanation is the sentence that it is trying to explain.
The explanans is the sentence in the explanation that does the explaining.
In general, an explanation tells you why some event happened. In other
words, an explanation identifies the cause of some observed effect. If I were to
tell you, for example, that the cause of a collision was the driver who ran the
red light, then I am giving you an explanation. I have identified the cause (the
driver running the red light) of the observed effect (the collision).
Let me be a bit more precise. Explanations have two parts: the explanandum
and the explanans. The explanandum is the sentence in the explanation that
CHAPTER 2. ARGUMENTS 21

describes the observed effect. The explanans is the sentence in the explanation
that describes the cause of the observed effect. So, in the example

The collision happened because one of the drivers ran a red light.

the “collision happen” is the explanadum and the sentence “one of the drivers
ran a red light” is the explanans. Notice that I used the indicator word “because”
to mark the difference between the explanadum and the explanans.
My example might led you to ask why it is not an argument. After all,
could I have said that the conclusion of the argument is “the collision happened”
and the premise is “one of the drivers ran a red light”. This appears to be a
reasonable interpretation of the example given my use of the word “because”.
Whether the example is an explanation or an argument in this case depends on
what the indicator word “because” is doing. In an explanation, the explanans
uses ‘because’ to show why something is the case. In an argument, the premises
of an argument use ‘because’ to justify the conclusion. The difference between
them is subtle.
Unfortunately, there is no full-proof way of distinguishing between explana-
tions and arguments. Philosophers and logician have invested a lot of time and
energy over the years trying to figure out how to draw a brightline between them
so that it is easy to tell them apart. The best and simplest way to distinguish
between is to ask whether the sentence in front of “because” is controversial
or not. If it is controversial, then “because” indicates an attempt to justify
the sentence. If the sentence is not controversial, then “because” signals an
explanation. Let’s go back to my previous example for a moment:

The collision happened because one of the drivers ran a red light.

The sentence before “because” is “The collision happened”. Do you think that
this sentence is controversial, that some people might disagree about whether it
is true or false? I doubt it, especially if they were eyewitnesses to the collision.
Since there is no controversy about whether the collision happened, there is
no need to justify the sentence. The only thing left to do is explain why the
collision happened. This means that the example is an explanation rather than
an argument.
For contrast, let’s take a look at another example:

The blue car caused the collision because I saw the blue car collide
with the red car after running the light.

Is this example an argument or an explanation? You might be able to tell right

away from the tone of the example is it is an argument; and you are right. But
there are details that it give it away as well. Is the sentence before “because”
controversial? There could be disagreement about which car caused a collision,
especially between the drivers. If so, then the example would be an argument
and the sentence after “because” would be a reason to believe the sentence in
front of it. Moreover, if this example was an explanation, then sentence before
CHAPTER 2. ARGUMENTS 22

“because” would be the explanadum and the sentence after it would be the
explanans. However, the sentence after “because” does not try to identify the
cause of the effect in the sentence before “because”. This is another reason why
the example is not an explanation.

Conditional Sentences
Finally, people often think that conditional sentences are arguments. A condi-
tional sentence is any sentence with a conditional clause. In English, a condi-
tional clause typically begins with the word “if”. For example, the sentence

If concussions in professional football cause brain injuries in player,

then the widespread approval given to this sport should be recon-
sidered.

is a conditional sentence because it has a conditional clause which is

If concussions in professional football cause brain injuries in player

...

Conditional sentences are often confused with arguments because, like argu-
ments, conditional sentence have have two parts: the sentence that comes im-
mediately after the word “if” is called the “antecedent”; and the sentence that
does not follow “if” is called “consequent”. In the above example, the sentence

the widespread approval given to this sport should be reconsidered.

is the consequent of the conditional and the sentence

concussions in professional football cause brain injuries in player

is the antecedent because it immediately follows “if”. Conditional sentence are

somewhat complicated and I will have a lot more to say about them in chapter
5. For now, I will focus on how conditional sentences are similar but different
from arguments.
Conditional sentences are similar to arguments in that there is an inference
from the antecedent of the conditional to its consequent of in the same way that
there is an inference from the premises of an argument to its conclusion. Con-
ditional sentences, however, are not arguments because they have no premises.
Compare my previous example of a conditional sentence

If concussions in professional football cause brain injuries in player,

then the widespread approval given to this sport should be recon-
sidered.

to a similar argument:

Premise: Concussions in professional football cause brain injuries in players.

CHAPTER 2. ARGUMENTS 23

Conclusion: The widespread approval of professional football should be recon-

sidered.

When you compare a conditional sentence and an argument, you should notice
that they both involve an inference. In case of the conditional sentence, I can
infer that consequent is true from the antecedent provided that the antecedent
is true. In other words, I can infer that the widespread approve of professional
football should be reconsidered if it is true that concussions in professional foot-
ball cause brain injuries in players. Similarly, I can infer that the conclusion of
the argument is true from the premise provided I assume that the premise is
true. In other words, I can infer that the widespread approval of professional
football should be reconsidered if it is true that concussion in professional foot-
ball cause brain injuries in players.
On the other hand, you might also notice a significant difference between
the conditional sentence and the argument when you compare the examples.
In the case of the argument, the premise is stated as a fact. But, in the case
of the conditional sentence, the antecedent is hypothetical. The word “if” in
front of the antecedent indicates that the antecedent is not yet known to be
true. It could be true but it also be false. The point of “if” is to spell out what
would also have to be true in case it turns out that the antecedent is true. Do
concussions in professional football cause brain injuries in players? You and I
may not know for sure whether the answer to this question is “yes” or “no”; but
what we can both be sure of is that widespread approval of professional football
might need to be reconsidered in case we learn that the answer is “yes”.
What all of this means is that the difference between an argument and a
conditional sentence is that arguments have premises which are asserted to be
true while conditional sentences have antecedents which are hypothetically true
but not yet known to be true.
The example above not only illustrates why conditional sentences and ar-
guments are confused but also demonstrates how conditional sentence can be
rewritten as arguments. By separating the antecedent of a conditional from
its consequent and, then, changing the antecedent into a premise, it is possi-
ble to transform any conditional sentence into an argument. For example, the
conditional sentence:

If people are free, then they can be held responsible for their actions.

can be rewritten as the argument:

Premise: People are free.

Conclusion: People can be held responsible for their actions.

In this example, the antecedent of the conditional becomes the premise of the
argument and the consequent of the conditional because the conclusion. Notice
as well how the antecedent changed when it became the premise. Instead of
being hypothetical true in the conditional sentence, the premise is asserted to
be true.
CHAPTER 2. ARGUMENTS 24

Ultimately, it is important to remember that not all kinds of speech and not
all kinds of writing present arguments. Different types of speech and writing
serve different purposes. A speech or passage contains an argument only if
it tries to prove something. More precisely, speech and writing contains an
argument if and only if it gives reasons which justify some conclusion. If a
speech or passage does not try to give any reason to justify a conclusion, then
it does not contain an argument.

2.3 Evaluating Arguments

Here is my reconstruction of the standard argument for cultural relativism that
I introduced at the end of the first section of this chapter:

Premise: Different cultures disagree about what is right and what is wrong.
Conclusion: Right and wrong are relative to each culture.

Is this a good argument? The answer depends on whether or not the premises
are true and whether the premises give a person a good reason to believe that
the conclusion is true. Philosophers describe such arguments as “sound”. Sound
arguments have premises that are true and justify the conclusion.
Definition (Soundness). An argument is sound when
• all of its premises are true, and
• its premises justify the conclusion.
If the premises are good reasons to believe that the conclusion is true, philoso-
phers sometimes say that premises “logically entail” a conclusion or that a
conclusion “logically follows” from the premises.
Unsound arguments have premises that are either false or appear to justify
the conclusion, but really do not. An argument is unsound, therefore, when
either

• one or more of its premises is false,

• it is logically consistent for its conclusion to be false when all of its premises
are true, or
• both

To figure out whether an argument is good or bad, it is best to ask yourself two
questions:

1. Are the premises of the argument really true?

2. Assuming the premises are true, do they provide a good reason or a bad
reason to think that the conclusion is true?
CHAPTER 2. ARGUMENTS 25

It is important to ask the first question because the premises of an argument

are the reasons given to justify the conclusion. If the premises aren’t really true,
then they are not good reasons to accept the author’s conclusion. For example,
according to the definition of soundness, this argument is unsound:

Premise: All philosophers are brilliant.

Premise: Socrates is a philosopher.

Conclusion: Socrates is brilliant.

Are the premises here true? You’ve already learned that Socrates was a philoso-
pher. So, you know that the second premise is true. What about the first? Not
all philosophers are brilliant. Many simply accept what their more brilliant
contemporaries said. Thus, while the second premise may be true, the first is
not; not all philosophers are brilliant. So, the argument is unsound because one
of its premises is false.
Now consider the standard argument for cultural relativism:

Premise: Different cultures disagree about what is right and what is wrong.
Conclusion: What is right and what is wrong is relative to each culture.

The premise here is “Different cultures disagree about what is right and what is
wrong”. Is this premise really true? Most people probably think that it is. After
all, different cultures usually do have different standards of right and wrong. I
say “usually” because it is possible that someone could disagree, pointing out
that different cultures also have the same standards in some cases. Murder, for
instance, is considered wrong in every culture. So, it is possible for someone to
deny that the premise is really true. And, if this premise really is false, then
it is not a good reason to believe the conclusion of the argument. Thus, the
argument is no good and should persuade no one of its conclusion.
However, let’s assume that the premise is really true. Even so the premise
may not be a good reason to believe the conclusion. This is why we need to ask
the second question: Assuming the premises are true, do they provide a good
reason or a bad reason to think that the conclusion is true?

2.4 Logical Validity

If the premises of an argument are good reasons for believing that the conclusion
is true, then argument is logically valid.

Definition (Logical Validity). An argument is logically valid if and only if

the falsehood of its conclusion is logically inconsistent with truth of all of its
premises.
This argument, for instance, is logically valid:
CHAPTER 2. ARGUMENTS 26

Premise: All philosophers are brilliant.

Premise: Socrates is a philosopher.
Conclusion: Socrates is brilliant.

How do I know that this argument is logically valid? Look again at the definition
of logical validity. It says that an argument is logically valid if and only if the
falsehood of its conclusion is logically inconsistent with the truth of all of its
premises. Suppose that the premises of this argument are true: Socrates is
a philosopher and all philosophers are brilliant. Now suppose the conclusion,
that “Socrates is brilliant” is false and ask yourself: can I consistently believe
that the first and second premises are true and that the conclusion is false.
The answer is “no”. If I assume that both premises are true, together they
imply that the conclusion is true. Why? Think about it. If it is true that all
philosophers are brilliant and that Socrates is a philosopher, then together these
imply that Socrates is brilliant. If I suppose that Socrates is not brilliant, then
either Socrates must not be a philosopher or not all philosophers are brilliant
because Socrates is the exception. Do you see? This argument is valid because
believing that the conclusion is false is logically inconsistent with believing that
the premises are true.
The definition of logical validity states that an argument is logically valid
if and only if the falsehood of its conclusion is logically inconsistent with the
truth of all of its premises. What happens when we encounter an argument
where the falsehood of the conclusion is logically consistent with the truth of
all of the premises? In this case, the argument is logically invalid if and only if
the falsehood of its conclusion is logically consistent with the truth of all of its
premises. For example, this argument is logically invalid:

Premise: The majority of philosophers are brilliant.

Premise: Socrates is a philosopher.

Conclusion: Socrates is brilliant.

Suppose that the first and second premises are true. Now suppose that the
conclusion is false and ask yourself: can I consistently believe these suppositions?
The answer here is ‘yes’. It is logically consistent to believe that the majority of
philosophers are brilliant and that Socrates is a philosopher without believing
that Socrates is brilliant; he may be the exception, in which case the conclusion
can be false while the premises are true.
From this definition of logical validity, we can extract a rough method for
figuring out whether an argument is logically valid:

1. Assume that all the premises of an argument are true.

2. Assume the conclusion is false.
CHAPTER 2. ARGUMENTS 27

3. If these assumptions are logically inconsistent, then the argument is logi-

cally valid.
4. If these assumptions are logically consistent, then the argument is logically
invalid.

The reason that this method is rough is because it suggests that whether an
argument is logically valid or invalid depends whether all of the premises are
true. However, the definition of logical validity states that logical validity de-
pends only on the relationship between the premises and the conclusion of an
argument. It is important to see that the validity of an argument does
not depend in any way on the truth of its premises or its conclusion.
In fact, it is possible for an argument with false premises and a false conclusion
to be logically valid. For example,

Premise: The president of the United States must be younger than 35 years
of age.

Premise: Bagwell is president of the United States.

Conclusion: Bagwell must be younger than 35 years of age.

According to the Constitution of the United States, a person is eligible to be

president only if that person is at least 35 years old. I am certainly not president
of the United States. I am also older than 35. In other words, the premises and
the conclusion are false. However, this argument is logically valid. Using the
rough method of figuring out whether an argument is logically valid, assume that
the first premise is true, that anyone eligible to be president must be younger
than 35 and that the second premise is true, that I am indeed president. Assume
also that the conclusion is false, that I am in fact older than 35 years and, then,
ask yourself whether it is logically consistent to assume that the conclusion is
false while assuming that both of the premises are true. The answer is “no”.
It is logically inconsistent to assume that I am older than 35 and president of
the United States while assuming that the president must be younger than 35.
Because these assumptions are logically inconsistent, the argument from the
premises to the conclusion is logically valid despite the fact that both of the
premises and the conclusion are in fact false. This example illustrates how the
logical validity of an argument does not depend on the truth of the premises or
the conclusion of an argument.
It is also possible for an argument with true premises and a true conclusion
to be logically invalid. For example,

Premise: The president of the United States must be at least 35 years of age.
Premise: Joe Biden is at least 35 years of age.
Conclusion: Joe Biden is president of the United States.
CHAPTER 2. ARGUMENTS 28

Again, the Constitution of the United States requires a person to be at least

35 years old to be eligible to be president. Joe Biden is currently serving as
president of the United States and, when he was elected in 2020, he was 77
years old. In other words, the premises and the conclusion are all in fact true.
You might already see that this argument is logically invalid. Suppose that
Joe Biden lost the election. In this case, the conclusion would be false, but
the premises would still be true because they really are true. Is it logically
consistent to suppose that Joe Biden lost the election and assume that Mr.
Biden is at least 35 years old and that the constitution requires anyone eligible
to be president to be at least 35 years old? Yes, of course, they are logically
consistent. But, then, the argument is logically invalid despite the fact that the
conclusion and all of the premises are true. So, again, you can see how the
logical validity of an argument does not depend on the truth of the
premises or the conclusion of an argument.
The lesson here is that truth and logical validity are independent concepts;
one does not depend on the other. Logicians use the concept of truth only as a
way of testing whether an argument is logically valid or invalid. This, however,
does not mean that the fact of an argument’s validity somehow depends on the
truth of its premises or its conclusion.
Finally, here again is the standard argument for cultural relativism:

Premise: Different cultures disagree about what is right and what is wrong.
Conclusion: What is right and what is wrong is relative to each culture.

Is this argument sound? Since the premises of a sound argument justify

its conclusion, the answer to this question depends on whether the premise is
actually true and on whether the premise justifies the conclusion. What do you
think? Consequently, the whole point of logic is to find out when the premises
of an argument give us a reason to believe its conclusion and when they do not.
The standard argument for cultural relativism is an unsound argument because
the premise does not justify the conclusion. You probably have doubts about
this claim and that’s fine; many people consider this argument to be a good
one. So, here is brief explanation of why the standard argument for cultural
relativism is unsound. Suppose the argument is sound. If it is a sound argument,
then the premise should give you a reason to believe the conclusion. In other
words, if the premise is true, then the conclusion of the argument cannot be
false. What I mean is that if different cultures disagree about what is right and
what is wrong, then it cannot be that right and wrong are independent of each
culture. But consider this: is it possible that different cultures could disagree
about what is right and what is wrong when what is right and what is wrong
are not relative to each culture? Sure, it is possible for one culture to be correct
about what is right and what is wrong when other cultures are incorrect. I am
not saying which culture is correct about right and wrong and which one is not.
What I am saying is that it is possible for one culture to be correct and for
another culture to be incorrect. This possibility indicates that the premise of
CHAPTER 2. ARGUMENTS 29

the argument could be true and the conclusion could be false. Consequently,
the argument is logically invalid and, therefore, unsound.

2.5 Counterexamples to Arguments

It is not always easy to tell whether an argument is logically valid or in-
valid. Even knowing the definition of logical validity precisely and following
the method for determining whether an argument is valid will not guarantee
that one will correctly identify an argument as valid or as invalid. Mathematics
can solve this problem by providing us with precise and rigorous methods of
testing the logical validity of arguments called “proof theories”. I cannot intro-
duce you to proof theories yet, but I will introduce you to one of these methods
in chapter 4. So, for now, I will show you an informal and imprecise method of
determining validity. An informal way of testing an argument for logical valid-
ity is to look for a counterexample to the argument from the premises to the
conclusion.
Definition (Counterexample to Arguments). An example which describes a
circumstances in which the premises of an argument are true and the conclusion
of that argument is false at the same time.
According to this definition, if an argument is invalid, then it will have a coun-
terexample.
For any logically invalid argument, a counterexample will show that it is
possible for the conclusion of the argument to be false and for all of its premises
to be true. Here, for example, is an invalid argument:

Premise: The majority of philosophers are brilliant.

Premise: Socrates is a philosopher.
Conclusion: Socrates is brilliant.

What is a counterexample to this invalid argument? One counterexample to

the argument might be:

Just because Socrates was a philosopher and most philosophers are

brilliant doesn’t entail that Socrates was also brilliant; he might be
the exception.

(In fact, Socrates was actually extremely brilliant but not simply because he
was a philosopher.)
Here is another example of an invalid argument:

Premise: If North Korea is developing nuclear weapons, then North Korea is

a threat to world peace.
Premise: North Korea is a threat to world peace
CHAPTER 2. ARGUMENTS 30

Conclusion: North Korea is developing nuclear weapons.

This argument is an instance of invalid pattern of argument called “affirming

the consequent”. What is a counterexample to this invalid argument? One
counterexample to this argument might be:

North Korea could be a threat to world peace without developing

nuclear weapons by funding terrorist organizations around the world.

Finally, here again is the standard argument for cultural relativism:

Premise: Different cultures disagree about what is right and what is wrong.

Conclusion: What is right and what is wrong are relative to each culture.

Since the argument for cultural relativism is invalid, we should be able to come
up with a counterexample to it. So, let’s assume the premise is true and the
conclusion is false. Can we describe a situation where the premise is true and
the conclusion is false? Here is one possibility:

One individual or culture may correctly believe that a moral judg-

ment is true while a different individual or culture may incorrectly
believe that it is false.

This counterexample describes a situation where the premise is true and the
conclusion is false at the same time, which proves that the argument is invalid.
In sum, a counterexample can prove that an invalid argument is logically
invalid but only if the argument is in fact logically invalid. If an argument is
logically valid, then it will be impossible to come up with a counterexample to
it. This is one reason the counterexample method is rough. If an argument is
logically valid, then you will simply not be able to come up with a counterex-
ample. But how do you know when to stop trying to come up with one? There
is no way to know. If you try and fail to come up with a counterexample, it
may be that the argument is logically valid. On the other hand, it may also be
that your imagination has failed you. Failures of imagination do not mean that
an argument is invalid.

2.6 Implicit Premises

Most people do not intend to give invalid arguments. When reconstructing
another person’s argument there is a risk of committing a logical mistake that
philosophers call “the straw man fallacy”:
Definition (Straw Man Fallacy). A mistake in reasoning that occurs when a
person misrepresents another person’s point of view or opinion in such a way
that it will be easily refuted or rejected.
CHAPTER 2. ARGUMENTS 31

In other words, since most people do not intend to make invalid arguments,
reconstructing another person’s argument in such a way that it will turn out to
be invalid is a straw man.
The straw man fallacy is a mainstay of political campaigns. Here are two
recent examples of the fallacy. During the 2016 election, Donald Trump insisted
that his opponent Hilary Clinton wanted to take away everyone’s guns because
she has endorsed gun-control policies. This characterization of Clinton’s position
commits the straw man fallacy because it misrepresents Clinton’s point of view
on gun control in such as a way as it makes it easy to reject.
On the other hand, Hilary Clinton insisted that Donald Trump wants to start
a nuclear war because he had endorsed expanding the United States’ nuclear
arsenal. This characterization of Trump’s position also commits the straw man
fallacy because it misrepresents Trump’s point of view in order to make it easy
to reject his position.
Philosophers try to avoid the straw man fallacy because they are lovers of
wisdom and it is unwise to misrepresent another person’s argument deliberately.
The way to avoid the straw man fallacy is to adopt the principle of interpreta-
tive charity or the principle of charity for short. In the context of evaluating
arguments,
Definition (The Principle of Charity). A person ought to interpret another
person’s argument as valid rather than invalid if it is possible to do so.
In other words, the principle of charity requires us to try out best to interpret
other people’s arguments as valid if it is at all possible to do so. But, what if
someone presents an argument that really is invalid? In this case, the argument
is invalid and there is nothing to be done. However, even in a case like this
there is good reason to stick with the principle of interpretive charity:

The principle of charity can reveal the assumptions a person uncon-

sciously believes that makes their arguments valid.

If a person accepts an invalid argument, that person must either be irrational

or be unconsciously assuming a premise to be true which makes the argument
valid. In most cases, it is unwise to assume that such a person is irrational.
Thus, it is better to assume that the person has neglected to state explicitly
some premise which his or her argument needs in order to be valid. Such a
premise is called an “implicit premise”:
Definition (Implicit Premise). A unstated premise required for an argument to
be valid.
I urge you to abide by the principle of charity which requires you to interpret
this person’s arguments as valid and, thus, to interpret an argument as relying
upon an as yet unstated and perhaps unknown premise. Doing so enables you
to help another person see for the first time the assumptions he or she is making
which allows him or her to examine his or her own assumptions.
It is worth mentioning that the term “assumption” is ambiguous:
CHAPTER 2. ARGUMENTS 32

Definition (Ambiguity). An expression is ambiguous if and only if it has more

than one discrete meaning.
In some cases people use the term “assumption” to describe a belief or opinion
which someone accepts without any good reason to do so. Thus, according to
some people, religious beliefs are assumptions because such people think that
there is no good reason to accept them. In other cases, people use the term
“assumption” to describe a belief or opinion which someone accepts without
realizing. The difference is that in one sense of the term “assumption” a person
has no good reason to make an assumption but does anyway, where in the other
sense of the term, a person has no reason at all—good or bad—to make the
assumption but has done so anyway without knowing it.
In philosophy, the term “assumption” is used in the second sense, that is, an
assumption is a belief or opinion which someone accepts without realizing he or
she accepts it. This sense of the term does not imply that assumptions are good
or bad. In the context of an argument, an assumption is a general belief which
a person relies on to make their inferences valid. A person may or may not have
reason to believe the assumption; all that matters for our purposes is that such
a person doesn’t at first realize that he or she is making an assumption and
don’t realize that he or she needs to do so in order for his or her arguments to
be valid.
The straw man fallacy and the principle of charity present us with an oppor-
tunity to examine our own and other people’s assumptions. In the case of other
people, our efforts to avoid the straw man fallacy and to follow the principle
of charity encourage us to identify the assumptions other people make and to
understand whether their assumptions are justified. In our own case, we do not
have to worry so much about the straw man fallacy since people do not tend
to misrepresent their own opinions or their own points of view; but we do ulti-
mately commit the straw man fallacy in characterizing our own beliefs when we
neglect or overlook the assumptions we make which make our arguments valid.
So, by avoiding the straw man fallacy and following the principle of charity, we
open ourselves up to the possibility of discovering the assumptions we make for
the first time so that we may be able to examine them for ourselves, in order to
understand whether we have any reason to hold the assumptions we do.
In the context of an argument, an implicit premise is a belief which a person
relies upon to make his or her argument valid. How do we identify implicit
premises? There is no precise method of correctly identifying implicit premises.
One way to identify them in an invalid argument is to

1. Identify the predicate of the sentence in the premise.

2. Generalize the predicate to include every case to which it applies.
3. Make this generalized predicate of the premise the subject of a new sen-
tence.
4. Identify the predicate of the sentence in the conclusion.
CHAPTER 2. ARGUMENTS 33

5. Make the predicate of the conclusion the predicate of the new sentence.
6. Add the completed new sentence to the premises of the argument.

In chapter 1, I mentioned that the predicate of a sentence is the part of the

sentence which excludes the subject. Here’s a simple example. Suppose someone
presents you with the following argument:

Premise: Kurt Gödel is a logician.

Conclusion: Gödel is paranoid.

This argument is invalid. Here’s a counterexample:

Gödel is a logician, but he might not have been paranoid if his health
had been different.

In order words, it is possible that Gödel is a logician but not paranoid. So,
you can reject the argument as invalid and move on with your life. However, it
is unlikely that the person arguing this way would be so careless. Dismissing
the argument as invalid, therefore, commits the straw man fallacy. Instead of
dismissing the argument as invalid, I will adopt the principle of charity and try
to identify the implicit premise that this argument requires to be valid. The
rough method outlined above instructs me, first, to identify the predicates of
the premise and the conclusion, which are

. . . is a logician

and

. . . is paranoid

The “. . . ” are called an “ellipsis” and mark where I have left out the subjects
of these sentences, in this case “Gödel”.
Next, the method directs meto generalize the premise’s predicate to every
case and turn it into the subject of a new sentence which includes the predicate
of the conclusion as its own predicate:

All logicians are paranoid.

This is the implicit premise. Now, I add it to the original argument:

Assumption: All logicians are paranoid.

Premise: Kurt Gödel is a logician.

Conclusion: Gödel is paranoid.
CHAPTER 2. ARGUMENTS 34

Is this argument valid? It is. The truth of the implicit and the explicit premises
of the argument are logically inconsistent with the falsehood of its conclusion.
How do I know that this implicit premise is one the person who gave this
argument has in mind? I don’t. It may be the one that the person has in mind,
but the person may have some other implicit premise in mind that also makes
the argument valid. The only way to know for sure is to ask the person. But,
it doesn’t really matter whether I have correctly identified the implicit premise
the person has in mind or not. What matters is that I’ve identified an implicit
premise that makes the argument valid. If it turns out that this is not the
implicit premise the person giving the argument relies on, then that person may
supply the correct implicit premise and I can, then, check to see whether the
argument needs this implicit premise to be valid. Otherwise, the argument is
invalid and the game is over.
Now consider, once more, the standard argument for cultural relativism:

Premise: Different cultures disagree about what is right and what is wrong.
Conclusion: What is right and what is wrong are relative to each culture.

You already know that this inference is invalid and I have already proven that
it is invalid by finding a counterexample to it. Suppose that the person making
the argument meant for it to be valid? What implicit premise does the validity
of this argument rest on? The argument assumes that

There is no way to settle disagreements about what is right and what is wrong.

Adding this assumption to the argument gives us:

Premise: Different cultures disagree about what is right and what is wrong.
Implicit Premise: There is no way to settle disagreements about what is right
and what is wrong.
Conclusion: What is right and what is wrong are relative to each culture.

This inference is valid. The truth of both the explicit premise and the implicit
premise is logically inconsistent with the falsehood of the conclusion. Good! The
argument is now obviously valid, but you must still ask whether the argument is
sound. To determine whether this argument is sound, you need to know whether
its premises—including its implicit premise—are in fact true. Notice that you
would not be in a position to evaluate the soundness of this argument fairly
if you had not taken the time to identify the implicit premise responsible for
making the inference valid. If you had neglected to identify the implicit premise,
then you would have prematurely rejected the argument as invalid. Now that
the implicit premise is explicit, you are in a position to evaluate the soundness
of the argument. You already know that the premise is true; cultures do as a
matter of fact disagree about what is right and what is wrong. Is there no way
to settle disagreements about right and wrong? I’ll leave you to figure out the
answer to this important philosophical question on your own.
CHAPTER 2. ARGUMENTS 35

2.7 Summary
In this chapter, I have introduced you to arguments and explained how to eval-
uate one. An argument is a set of sentences, one of which is the conclusion and
the rest are the premises. The premises of an argument are reasons to believe
the conclusion. A good argument is sound and a bad argument is unsound. A
sound argument is logically valid and has premises which are really true. The
premises of a logically valid argument guarantee that the conclusion is true.
To put it more precisely, the inference from the premises of the argument to
its conclusion are logically valid or logically invalid depending on whether the
truth of the premises is logically consistent with the falsehood of its conclusion.
If the premises are logically consistent with the falsehood of the conclusion,
then the argument is logically invalid. If, however, the truth of the premises is
inconsistent with the falsehood of the conclusion, then the argument is logically
valid. One way to prove that an argument is invalid is by finding a counterex-
ample to it. A counterexample is an example which describes a situation in
which all of the premises of the argument are true and its conclusion is false.
Chapter 3

Logical Form

So far you have learned that a person’s beliefs are logically consistent when all
of them can all be true at the same time. At this point, you might be wondering
why it is possible for some beliefs to be consistent and not possible for others.
Whether it is possible for a set of beliefs to be consistent or not depends on
the logical structure of the sentences that express them. The logical structure
of a sentence is called its “logical form”. A logical form looks like this

(A ∧ (¬B))

. You may not understand what this logical form means right now, but logical
forms are very important. The logical form of a sentence determines whether
it is possible for a set of sentences to be true at the same time. If the logical
form of a set of sentences prevents the sentences from being true at the same
time, then the sentences are inconsistent. However, if the logical form of the
sentences does not prevent the sentences from being true at the same time, then
the sentences are consistent.
Thus, in order to understand why some sentences are consistent and why
some are not, you need to be able to identify the logical structure of each
sentence. To do this, logicians have developed a logical notation which I call
“sentence logical notation” that describes the logical form of a sentence by
stripping the sentence of its content and leaving only its form.
In this chapter, I will introduce sentence logical notation and explain how to
use this notation to describe the logical form of compound English sentences.
By the end of this chapter, you will know how to identify and describe the
logical form of a compound sentence.

36
CHAPTER 3. LOGICAL FORM 37

Chapter Objectives: By the end of this chapter, you will be able to

• Write out the logical form of three types of compound sentences:
negations, conjunctions, disjunctions
• Identify the main operator in the logical form of a compound sentence

• Leave parentheses out of a logical form when it is safe to do so

3.1 Simple and Compound Sentences

In chapter 1, you learned that a sentence is any spoken or written expression
that has a truth-value; that is, a sentences can be either true or false—but
not both. Only complete sentence have a truth-value. A complete sentence in
English must have a subject and a predicate.
There are two kinds of sentences: simple and compound:
Definition (Simple Sentence). A complete sentence that does not consist of any
other complete sentences as its parts.
For example, the sentence

Chryssipus is a logician.

is simple because it is a complete English sentence that cannot be taken apart

into other even simpler complete sentences and remain itself a complete English
sentence. In other words, if I remove a word from the example, the result will
not be a complete sentence.
Here are three more examples of simple sentences:

1. “Aristotle wrote Prior Analytics.”

2. “Logicians are philosophers.”
3. “Logic is an academic discipline.”

None of these examples can be broken down into simpler sentences without
making them incomplete. If I remove a word from any of these examples, the
result would be an incomplete sentence. This indicates that they are all simple
sentences.
A compound sentence, on the other hand, is
Definition (Compound Sentence). A complete sentence that consists of sim-
pler complete sentences and at least one connecting word that joins the simpler
sentences together.
For example, the sentence

The will is free and it is determined.

CHAPTER 3. LOGICAL FORM 38

is a compound sentence because it can be divided into two complete simpler

sentences: “The will is free” and “the will is determined”. The word “and”
connects these simpler sentences, making them one compound sentence.
Here is another example. The sentence

Knowledge requires experience or it requires reasoning.

is also a compound sentence because it can likewise be divided into two complete
simpler sentences: “Knowledge requires experience” and “Knowledge requires
reasoning”. In this example, the word “or” joins the two simpler sentences into
one compound sentence.
Here is one last example:

The mind is not a body

This example may not look like a compound sentence at first, but removing the
word “not” leaves the simpler complete sentence “The mind is a body”. So the
example does turn out to be a compound sentence. The word “not” is logically
very important and I will have more to say about it in section 3.3.
Connecting words such as “and”, “or”, and “not” in the examples above are
called “connectives”:
Definition (Connective). A word that combines sentences into compound sen-
tences according to certain grammatical rules.
There are two types of connectives: linguistic and logical. Linguistic connectives
are the connectives found in languages such as English and they combine words
and phrases into more complex expressions, some of which have logical structure
and some of which do not. The linguistic connectives that do not create logical
structure won’t be discussed in this textbook for do not have any logical form.
Linguistic connectives such as “and’, “or”, and “not”, however, do combine sim-
pler complete sentences into complete compound sentences according to certain
logical rules and, therefore, do have logical structure. Hence, logicians label
linguistic connectives of this kind “logical connectives”. The grammatical rules
for logical connectives are an important part of logical notation and I discuss
them in section 3.4.

3.2 Sentence Constants and Variables

Logicians employ special symbols to represent simple sentence and connectives.
They use constants to represent simple sentences and connectives to represent
certain connectives in English. A constant is a symbol that has a fixed value.
In mathematics, for instance, the symbol

represents the value 3.14. The value of π is fixed, meaning that its value cannot
change unless we redefine the symbol itself.
CHAPTER 3. LOGICAL FORM 39

It is easiest to understand constants by contrasting them with variables. Like

constants, variables are symbols. But, unlike constants, the value of a variable
is not fixed; the value of a variable, in other words, can vary, which is why they
are called “variables”. The familiar variable
x
could have a value of “3”, “10.5”, or “8888.888”. Its value can change without
refining the symbol itself.
In mathematics the value of a constant or variable is usually a number. In
logic, constants and variables takes other kinds of values. Constants whose fixed
values are simple sentences are called “sentence constants”:
Definition (Sentence Constant). Any capital letter of the alphabet such as “P ”,
“Q”, and “R” other than “X”, “Y ”, or “Z” used to represent simple sentences
of a language.
Like mathematicians, logicians reserve the letters “X”, “Y ”, or “Z” as vari-
ables that take sentence constants, calling them “sentence variables”. In logical
notation, I can represent the simple sentence
Chryssipus is a logician.
using a sentence constant such as

C
We can also represent the simple sentence
Aristotle wrote Prior Analytics.
with the sentence constant

P
And I can represent the simple sentence
Logicians are philosophers.
using
L
Finally, I can represent the sentence
Logic is an discipline of philosophy.
using
D
In all of these examples, the letters of the alphabet I have used as sentence
constants are arbitrary. I could have used any other letter to represent these
sentences. For instance, I could have used the letter
A
to represent the sentence
CHAPTER 3. LOGICAL FORM 40

Chryssipus is a logician.
There is no specific letter that must be used to represent a specific simple
sentence. Any letter will do.
When logicians wish to spell out or say what sentence constants represent,
they give an “interpretation”. In logic, an interpretation is just
Definition (Interpretation). A list of simple sentences and the sentence con-
stants that represent them.
For example, this is an interpretation of the examples given above:
C: Chryssipus is a logician.
P : Aristotle wrote Prior Analytics.
L: Logicians are philosophers.
D: Logic is a discipline of philosophy.

3.3 Connectives
In any language such as English, logical connectives join simple sentences to-
gether to make compound sentences. They can also be used to combine com-
pound sentences into larger compounds. There are three basic logical connec-
tives. They are,
1. Negation
2. Conjunction
3. Disjunction
In English, phrases such as “it is not the case that . . . ” and “not” are used
to indicate a negation. Words and phrases such as “and”, “also”, “moreover”,
“but“, “however”, “yet”, “still”, “although” and “nevertheless” indicate con-
junctions. Words like “or”, “unless” indicate disjunctions. These words and
phrases are not the only words and phrases English speakers use, but they are
the most common.
Logicians use special symbols called “logical connectives” to represents con-
nectives in languages such as English. The basic logical connectives are the tail
“¬” for negations, the carrot “∧” for conjunctions, and the wedge “∨” for dis-
junctions. Here is a table which, I hope, will summarize the logical connectives
and what they represent:
Operator Name Logical Function Represents
¬ Tail Negation “not”
∧ Carrot Conjunction “and”
∨ Wedge Disjunction “or”
What I will now do is explain each logical connective in turn to help you see
how the logical connectives are used to represent English connectives.
CHAPTER 3. LOGICAL FORM 41

Logical Negation
The symbol for logical negation is “¬”, called “tail”, and it can be substituted
for English negations such as “it is not true”, “it is not the case”, and “not”.
Tail always occurs in front of the the logical form it negates. For example, the
logical form
(¬G)
is a grammatically correct use of tail.
Tail can negate the logical forms of both simple and compound sentences:

(¬(A ∧ B))

The tail cannot be used to join two logical forms together. For example,

D¬H

is not a grammatically correct use of tail. Tail can immediately follow another
connective, but not a sentence constant. For example

G¬

is a grammatically incorrect use of tail.

Tail represents English negations such as “it is not true”, “it is not the case”,
and “not”. For example, the English sentence

Bertrand Russell is not a logician.

is a compound sentences that consists of the simpler sentence “Bertrand Russell

is a logician” combined with the English negation “not”. I can represent the
simpler sentence using the sentence constant “{R}” and replace “not” with tail,
which gives you the logical form:

(¬R)

This example can be rewritten as

It is not true that Bertrand Russell is a logician.

or

It is false that Bertrand Russell is a logician.

both of which consist of the simpler sentence “Bertrand Russell is a logician”

and an English negation. In the first case, “it is not the case” is the negation
and, in the second, the negation is the phrase “it is not true”. I can replace
both of these phrases with tail which gives you the logical form

(¬R)
CHAPTER 3. LOGICAL FORM 42

Logical Conjunction
The symbol for logical conjunction is “∧”, called “carrot”, and replaces En-
glish conjunctions such as “and”, “also”, “moreover”, “but”, “however”, “yet”,
“still”, “although”, NS “nevertheless”. The carrot can represent relative pro-
nouns such as “who”, “whom”, and “which” are basically conjunctions.
The carrot can join the logical forms of no more than two simple or compound
sentences together. For example, neither the logical forms

∧AB

nor
∧ABC
correctly use the carrot. Only logical forms such as

(A ∧ B)

use the carrot correctly.

The simple or compound sentences joined by carrot are called its “con-
juncts”. For example, the conjuncts of the logical form

(A ∧ B)

are the sentence constants “A” and “B”.

The carrot represents English conjunctions such as “and”, “also”, “more-
over”, “but”, “however”, “yet”, “still”, “although” and “nevertheless”. For
example, the compound sentence

Russell is a philosopher and Frege is a logician.

consists of two simpler sentences, “Russell is a philosopher” and “Frege is a

logician” and the English conjunction “and” which joins the simpler sentences
together. I can use sentence constants such as “R” and “F ” to represent the
simpler sentences and I can replace the English conjunction “and” with the
carrot which gives you the logical form of the compound sentence

(R ∧ F )

Similarly, the compound sentences

Russell is a philosopher, but Frege is a logician.

and

Russell is a philosopher; however, Frege is a logician.

CHAPTER 3. LOGICAL FORM 43

consist of the same simpler sentences, but the first contains the English con-
junction “but” while the second contains “however”. These conjunctions have
different connotations in English, but their differences are logically irrelevant.
So, I can replace them with the carrot just like I replaced “and” with the carrot.
Doing so, gives you the exact same logical form

(R ∧ F )

The fact that the last three examples have the same logical form means that they
are logically equivalent sentences; whatever differences in meaning they may
have in English are logically irrelevant. I will say more about logical equivalence
in chapter 4.
Here is one last example. The compound sentence

Russell and Frege are logicians

consists of two simpler sentences. It is easier to see the simpler sentences if I

rewrite the example as

Russell is a logician and Frege is a logician.

Notice that I didn’t change the meaning of the sentence; I just filled out the
meaning which was implied.
The simplest sentences in the example are “Russell is a logician” and “Frege
is a logician” joined together by the English conjunctive “and”. I can replace the
conjunction with the carrot and represent the simpler sentences with sentence
constants “R”, for “Russell is a logician”, and “F ”, for “Frege is a logician”.
The result is the logical form
(R ∧ F )
I’ve said that the carrot represents conjunctions in English. But the carrot
sometimes fails to match the meaning of English conjunctions. This happens
most often when the meaning of an English conjunction depends on the order
of the conjuncts. For example, the order of the conjuncts in the sentence

Plato got up and went to the Academy.

is important. The sentence suggests that Plato first got up and second went to
the Academy. If I used “P ” to represent the conjunct “Plato got up” and “A”
to represent “Plato went to the Academy”, then I could replace “and” with the
carrot and get
(P ∧ A)
which can be read in English as

Plato got up and Plato went to the Academy.

But notice that the order of the conjuncts no longer matters. The same sentence
could be rewritten
CHAPTER 3. LOGICAL FORM 44

Plato went to the Academy and Plato got up.

where the order clearly doesn’t matter. So, I cannot use the carrot to represent
“and” in this example, because “and” is doing more than just joining together
two simpler sentences which is all that the carrot can do.
The carrot also fails to match the meaning of English conjunctions when
they imply cause and effect. For example, the conjunction in the sentence

Socrates drank hemlock and died.

implies that drinking the hemlock caused Socrates to die. Again, the carrot
merely conjoins two conjuncts together, implying no more than this conjunction.
Thus, if an English conjunction implies a cause and effect relation between the
conjuncts, then the carrot will not be able to represent the meaning of the
English conjunction.

Logical Disjunction
The symbol for logical disjunction is “∨”, called “wedge” and it replaces English
disjunctions such as “or” and “unless”. For example, the compound sentence

Aristotle invented logic or Plato did.

consists of two simpler sentences “Aristotle invented logic” and “Plato invented
logic” joined together by the English disjunction “or”. These simple sentences
joined together by the English disjunction are called its “disjuncts”. After
substituting these simple sentences with the sentence constants “A” and “P ”,
I can replace the English disjunction “or” with the wedge which gives me the
logical form:
(A ∨ P )
Here the sentence constants “A” and “P ” are the disjuncts of the wedge.
Here are two more examples. The compound sentences

Aristotle invented logic unless Plato did.

and

Unless Aristotle invented logic, Plato did.

consists of the same simpler sentences. The difference in these cases is the
English disjunction “unless”. This word has different connotations than “or”,
but the difference is irrelevant from a logical point of the view. The only thing
that matters is that “unless” indicates a disjunction just as much as the word
“or” does. So, I replace “unless” with a wedge just like I replaced “or” with
the wedge. Doing so gives me the same logical form for these examples as the
previous one:
(A ∨ P )
CHAPTER 3. LOGICAL FORM 45

Like the carrot, the wedge can join not only two simple but also two com-
pound sentences. For example, the logical form

((A ∧ B) ∨ (C ∧ D))

is a compound sentence, consisting of two conjunctions linked with a wedge. In

this case, the disjuncts of the disjunction are themselves the compound sentences
“(A ∧ B)” and “(C ∧ D)”. So, the wedge can connect the logical forms of both
simple and compound sentences.

3.4 Formation Rules for the Logical Connectives

In section 3.1, I mentioned that the logical connectives combine the logical forms
of sentences into compound sentences according to certain grammatical rules.
Logicians call these rules “formation rules”. For any logical form “{X}” and
“{Y}”,

1. “X” and “Y ” by themselves are grammatically correct.

2. “(¬X)” is grammatically correct

3. “(X ∧ Y )” is grammatically correct.

4. “(X ∨ Y )” is grammatically correct.

Again, “{X}” and “{Y}” above are variables which may take the logical forms of
either simple or compound sentences. Notice that the parentheses in formation
rules (2), (3), and (4) are part of the grammar of the tail, the carrot, and the
wedge. This means that they must be included in any logical form which has
one of these logical connectives.
We can use these grammar rules to check whether a logical form is gram-
matically correct. According to (1), the logical form

is grammatically correct.
According to (2), the logical form

(¬A)

is grammatically correct, but the logical form

A¬

is not.
According to (3), the logical form

(A ∧ B)
CHAPTER 3. LOGICAL FORM 46

is grammatically correct, but the logical form

AB∧

is not. Neither is
∧AB
nor is
A∧B
You might be surprised by this last example. Isn’t it the exact same logical
form as “(A ∧ B)” which I said was grammatically correct? If this logical form
is grammatical correct, then why is “A ∧ B incorrect? The difference is the
parentheses. Remember that the parentheses are part of the grammar of the
carrot. The first example has parentheses but the last one does not. This is
why the first is grammatically correct and the last is not.
According to (4), the logical form

(A ∨ B)

is grammatically correct, but neither

∨AB

nor
AB∨
are grammatically correct according to (4).
Of course, the logical connectives can be combined with one another and
with themselves to make larger compound sentences. The formation rules can
help you determine whether or not any logical form you may encounter is gram-
matically correct. For example, the logical form

(¬(A ∧ B))

is grammatically correct according to rules (1), (2), and (3). You can check by
breaking the logical form down using one rule at a time. Similarly, the logical
form
((¬A) ∧ B)
is grammatically correct according to the same rules.
Furthermore, the logical form

(¬(A ∨ B))

is grammatically correct according to rules (1), (2), and (4). So is

(A ∨ (¬B))

and
((A ∨ B) ∧ (B ∨ C))
CHAPTER 3. LOGICAL FORM 47

though, in this last case, the example is grammatically correct because of rules
(1), (3), and (4).
In sum, the formation rules for tail, carrot, and wedge determine whether
or not a logical form containing one or more of these logical connectives is
grammatically correct. These rules, however, are not the only ones. In chapter
5, I will introduce formation rules for two additional logical connectives, the
arrow and the double arrow.

3.5 Scope
The reason parentheses must be included in any logical form that contains a log-
ical connective is because parentheses define the logical scope of the connective.
Here is a definition:
Definition (Scope). The scope of a logical connective consists of the simple or
compound sentences it joins together.

Whenever more than two sentence constants appear in a logical form of a sen-
tence, logicians use parentheses to indicate the scope of the logical connective
to avoid ambiguity:
Definition (Ambiguity). A sentence is ambiguous if it has two or more discrete
meanings.

For example, the sentence

Frege and Russell are married.

is ambiguous. It could mean

a. Russell and Frege are married to each other

b. Russell is married to someone and Frege is married to someone

Does the example mean (a) or (b)? In conversation, the context usually helps
us to determine which meaning is intended, but we have no context to work
with here. All logical analysis can do is tell us what the options are.
Incidentally, it is important to distinguish ambiguity from vagueness. As
I’ve defined it, a sentence is ambiguous if it has two or more discrete meanings.
The same could be said of an individual word. The English word “bank”, for
instance, can mean (a) a river bank, or (b) a financial institution. Individual
words can also be vague; sometimes it is clear that what the word applies to
and sometimes it is not. In such situations, the word is vague because there are
borderline cases of applicability, cases where it is not clear whether the word
applies. For example, the word “tall” is vague. Sometimes it is clear that a
person is tall and sometimes it is clear that a person is not. However, there
are cases where it is not clear whether or not a person is tall. Is a six-foot tall
CHAPTER 3. LOGICAL FORM 48

woman tall? Some people may say “yes” and some may say “no”. Examples
like this illustrates how the word “tall” is vague.
Not all words are vague. “Bank”, for instance, is not vague. There are
no borderline cases where it is unclear whether or not something is a bank.
Moreover, sentences cannot be vague. If a sentence seems to be vague, the
culprit is really some specific word in the sentence. For example, the sentence

Russell is tall.

is not vague, except insofar as the word “tall” is vague. Like ambiguity, the
context of a conversation usually helps us sort out what an individual word
means. It may be unclear whether Russell is tall until you compare Russell
with someone who is shorter than he is or someone taller than he is. In these
situations, it becomes clear whether Russell is tall and whether the word “tall”
applies to Russell or not.
In sum, only individual words can be vague while both sentences and indi-
vidual words can be ambiguous. The ambiguity of logical forms is due most
often to the scope of its logical connectives. The parentheses that accompany
logical connectives mark out the various possible interpretations of an ambigu-
ous English sentence. So knowing where to put the parentheses will indicate
these possibilities. Commas, semicolons, the words “either”, “both”, and single
predicates in English sentences are clues to help you avoid ambiguity. They can
tell you where to put parentheses in the logical form of a sentence.

The Main Operator

Often the logical forms of compound sentences contain multiple connectives.
Compound sentences need not consist only of simple sentences; they may consist
of other compound sentences as well. For example, the compound sentence

It is not true that both Plato and Aristotle are logicians.

is a negation of a conjunction. Its logical form is

(¬(P ∧ A))

where “P ” represents the simple sentence “Plato is a logician” and “A” rep-
resents “Aristotle is a logician”. In cases like this one where there are two
connectives, logicians speak of one as the main operator:
Definition (Main Operator). The connective in a logical form with the widest
scope.
Scope is a property of connectives which indicates what sentences are effected
by the connective. In the example above, the scope of “∧” is “P ” and “A”
because both of these sentence constants are under the control of “∧”. “¬”,
however, has a wider scope than “∧” because “∧” falls within the scope of “¬”.
The parentheses indicate that “¬” has wider scope by marking “(P ∧ A)” as a
CHAPTER 3. LOGICAL FORM 49

single unit under the control of “¬”. This makes the tail the main operator of
the logical form because it has the widest scope of the connectives in the logical
form.
Compare the example above with this example and note the difference:

((¬P ) ∧ A)

Here, the main operator is “∧” because it has wider scope than “¬”. How can
you tell? There are two ways. First, look at the position of the parentheses
relative to the connectives. In the first example, the “¬” falls outside the paren-
theses which, as I have said, indicates that the sentence constants “P ” and
“A” and logical connective “∧” within the parentheses are a single unit within
the scope of “¬”. In the second example, the “¬” falls within the parentheses
indicating that the logical form of the sentence is one unit. In this case, the
parentheses are superfluous and you can safely ignore them. However, they tell
you that the scope of “¬” includes only “P ” which means that the scope of “∧”
includes not only “A” but “¬P ” as well. In other words, the scope of “∧” is
the entire logical form of the sentence. So it is the main operator of the logical
form.
The significance of the main operator may not be clear at this point but its
importance will become clear in chapter 4.

3.6 Bracketing Conventions

Parentheses are necessary to clarify the scope of a logical connective and to
avoid structural ambiguity, but parentheses can also make it difficult to spot
the main operator of a logical form with many logical connectives. For example,

(¬((A ∧ B) ∨ (C ∧ D)))

This logical form has a lot of parentheses which make it hard to read and also
hard to see the main operator easily. (What do you think the main operator of
this logical form is?) By definition, the main operator of a logical form is the
connective with the most scope. In the case of the example, the main operator
is the tail. The parentheses that enclose the entire logical form are supposed to
be the clue, but there are so many parentheses in the example that it can be
hard to keep track of which connective is the main one.
To get around this problem logicians have adopted a set of conventions for
dropping or eliminating some of the parentheses from a logical form. It is safe
to leave out some of the parentheses when doing so will not change the meaning
of the form and when they don’t introduce ambiguity into the form. There are
three conventions. They are

1. The parentheses around the entire logical form can be left out.
2. The parentheses around a tail can be left out.
CHAPTER 3. LOGICAL FORM 50

3. Parentheses around a strings of carrots or strings of wedges can be left

out.

Let me give you some examples to illustrate how these conventions work.
The first convention stipulates that you can ignore the parentheses around
the entire logical form. In the example above, this means that you can drop the
parentheses around the entire form:

¬((A ∧ B) ∨ (C ∧ D))

Notice how much easier this makes it to spot the main operator of the logical
form. The main operator is the tail because it is the only connective outside the
parentheses. The only connective outside of the form is the main one. Based
on this convention, I can say that the main operator will always fall outside the
logical from.
The same example also illustrates the second convention. Because the main
operator happens to be a tail, removing the parentheses around the entire logical
form follows the second as well as the first convention. The second convention
allows you to drop any parentheses around a tail.
The second convention is sometimes misused. Consider, for instance, the
logical form:
¬(A ∧ B)
Can you apply the second convention to this form? No. The second convention
says that you can leave out the parentheses around a tail; it does not say that
you can leave out the parentheses next to a tail. If you were to leave out the
parentheses from this form, you would end up with

¬A ∧ B

which is not the same logical form. In the original example, the tail negated the
rest of the form. By removing the parentheses you have changed the meaning of
the logical form so that the tail is now negating only part of the form, namely
“A”. Thus, it is not safe to leave out the parentheses next to a tail but it is safe
to leave out the parentheses around the tail.
The third convention stipulates that you can ignore all of the parentheses
in a logical form that consists of a string of carrots or a string of wedges. For
example, the parentheses in the logical form

((P ∧ Q) ∧ ((R ∧ S) ∧ T ))

can be left out without changing its meaning:

P ∧Q∧R∧S∧T

Similarly, the parentheses in this logical form

(P ∨ (Q ∨ (R ∨ (S ∨ T ))))
CHAPTER 3. LOGICAL FORM 51

can be left out like so

P ∨Q∨R∨S∨T
In both of these examples, the order of the connectives doesn’t matter because
they are the same even though their scope differs.
The third convention does not allow you to leave out parentheses of a logical
form when the logical form includes a mix of carrots and wedges. For example,

((P ∧ Q) ∨ (R ∧ S))

does not have a string of carrots or a string of wedges in it. So, the third
convention does not allow you to remove the parentheses from this logical form.
However, the first convention does apply to this logical form and it allows you
to drop the outermost parentheses:

(P ∧ Q) ∨ (R ∧ S)

If it is not clear why this third convention works, it should become clearer
after reading chapter 4 where you will learn about when the carrot and the
wedge are truth and false.
It is important for me to stress that these rules are conventions that logicians
have agreed to. What this means is that they do not count as rules of logic
in the same way the formation rules for the logical connectives are rules of
logic. Another way to think of this point is that the conventions only exist as a
convenience for ignoring some of the parentheses in the logical form. Officially
the parentheses are there, but the conventions say that you do not need to pay
any attention to them.

3.7 Sentence Logical Form

Now that you have learned the formation rules of the logical connectives, I can
define logical form precisely:
Definition (Logical Form). An arrangement of sentence variables and logical
connectives such that the uniform substitution of simple or compound sentences
in place of the variables results in a compound sentence.
Based on this definition, a compound sentence has a certain logical form if it
can be produced by substituting simple or compound sentences in place of the
variables in that form. Recall that sentence variables are the capital letters
“X”, “Y ”, and “Z”, which take any simple or compound sentence. Here are
two example of a logical form:
(¬X)
(X ∧ Y )
Both of these examples are logical forms because substituting statements “A”
and “B” in place of “X” and “Y ”, respectively, results in the same sentences
“(¬A)” and “(A ∧ B)”.
CHAPTER 3. LOGICAL FORM 52

3.8 Sentence Logical Analysis

Using this definition of logical form, you are now equipped to describe the logical
form of many sentences in English. Here is a method for doing so:

1. Identify the simplest sentences and any English connectives in the given
compound sentence.
2. Rewrite the compound sentence to make the simplest sentences easier to
see.
3. Assign sentence constants to each simple sentence.

4. Replace each simple sentence with the sentence constant assigned to it.
5. Identify the English connectives in the compound sentence.
6. Rewrite the compound sentence to make the connectives easier to see.

7. Replace English connectives with the appropriate logical connective.

Now I will show you how to use this method to describe the logical form of three
compound English sentences. Here is the first:

Either a person’s actions are determined or they are free, but not both.

The first step of the method for describing the logical form of a compound
sentence directs you to identify the simplest sentences in the compound sentence.
In this case, they are

A person’s actions are determined.

and

A person’s actions are free.

Notice that, if I were to remove any word from each of these sentences, I would
end up with an incomplete English sentence. So, these sentences are the simplest
sentences in the compound sentence.
The second step of the method is to rewrite the compound to make the
simplest sentences explicit. This means that that I need to rephrase the com-
pound sentence so that the simple sentences that I’ve identified are as obvious
as possible.

A person’s actions are determined or a person’s actions are free,

and it is not true that a person’s actions are determined and that a
person’s actions are free.
CHAPTER 3. LOGICAL FORM 53

Rewriting the compound sentence to make the simple sentences and the connec-
tives explicit will make it easier to replace the simple sentences with sentence
constants in the third step of the method and to replace the English connectives
with the appropriate logical connectives in the last step.
The third step of the method directs you to assign sentence constants to
each of the simplest sentences. Remember that you can use any capital letter of
the alphabet as a sentence constant, except “X”, “Y ”, and “Z”. For the simple
English sentences above, let’s use “D” and “F ”:

D: A person’s actions are determined.

F : A person’s actions are free.

Now that I’ve assigned sentence constants to these two simple sentences, the
fourth step of the method directs me to replace the simple English sentences in
the compound sentences with the sentence constants that I’ve assigned to them.
Doing so will give me

D or F , and it is not true that D and that F .

The last two step of the method directs me to clarify and, then, to replace the
English connectives with the appropriate logical connectives. In this case, I have
already made English connectives explicit. So, I must now replace “or” with
the wedge, the word “not” with a tail, and the words “and” with carrots. The
result is the logical form

((D ∨ F ) ∧ (¬(D ∧ F ))

which is the logical form of the first example. Note that I added a pair of
parentheses for each logical connective to indicate the scope of the connective.
Parentheses are important because I will need to know which connective is
the main operator. In this case, the main operator is the first carrot because
it includes all of the other connectives in the logical form of the compound
sentence within its scope. The wedge does not include the first carrot in its
scope nor does it include the tail or the second carrot. These connectives also
do not include the wedge in their scope. So, none of them can be the main
operator.
Consider now a second example:

A person is neither her mind or her body.

The first step to identifying the logical form of a compound sentence like this
one is to identify its simplest sentences. Here, the two simplest sentences are

A person is her mind.

and

A person is her body.

CHAPTER 3. LOGICAL FORM 54

Notice, again, that I cannot take out any word from either one of these sentences
without turning them into incomplete sentences.
The second step of the method is to rewrite the compound sentence to make
the simple sentences explicit. One way to do this is to rewrite the compound
sentence this way

Neither a person is her mind nor a person is her body.

Rephrasing the compound sentence this way makes it sound awkward and un-
natural. However, the rewritten sentence has the same meaning as the original
so I am safe to rephrase it this way.
The third step is to assign a sentence constant to each of these simple English
sentences. Remember you can use any capital letter of the alphabet you like to
do this, except letters such “X”, “Y ”, and “Z”. For these simple sentences let’s
use the sentence constants “M ” and “B”. Here is the interpretation:

M : A person is her mind.

B: A person is her body.

Now that I have made the simple sentences easier to find, I can replace them
with the sentence constants I assigned to them. This gives me

Neither M nor B.

The fourth step is to make the logical connectives explicit. The words “Neither
. . . nor . . . ” do not appear obviously to be a disjunction or a conjunction. Yet,
the word “nor” suggests a disjunction along with a negation. Think of “nor” as
combining “not” and “or”, which suggests that “neither . . . nor . . . ” should be
written as
¬... ∨ ¬...
but this is not correct. Why? Think about how this logical form would sound
if you read it out in English. You can read it as saying

It is not true that a person is her mind or it is not true that a person is her
body.

This reading indicates that the logical form is saying that a person is not her
mind or not her body, but this is not what the original compound sentence
means. Take another look

Neither is a person her mind nor a person is her body.

This sentence means that a person is not her mind and not her body. So the
phrase “neither X nor Y ” is a conjunction of two negations, which can be
written as
((¬X) ∧ (¬Y ))
CHAPTER 3. LOGICAL FORM 55

Thus, if I substitute the sentence constant “M ” for the variable “X” and the
sentence constant “B” for the variable “Y ”, then I will get

((¬M ) ∧ (¬B))

And this is the logical form of the compound sentence “Neither is a person her
mind nor a person is her body”.
The lesson here is that the logical form of a compound sentence represents
the meaning of the sentence—not the words used in the sentence. The fact that
the words “or” appears in the compound sentence does not necessarily mean
that its logical form includes a wedge. The same point can be made about the
words “not” and “and”. English makes use of a variety of words to express
negation and conjunction, all of which will have logical forms that include the
tail and the carrot. Yet, in the right combination, the same words could be
used to express a disjunction in which case the logical form of the sentence will
include a disjunction.

3.9 The Logical Form of Arguments

In chapter 2, you learned that a sound argument is valid and that a valid
argument is one in which it is logically inconsistent for all of its premises to be
true and its conclusion to be false. At this point, you might be wondering how
it could be impossible for some arguments to have false conclusions and true
premises while it is possible for others.
Whether it is possible or not for all of the premises of an argument to be
true and its conclusion to be false depends on the logical form of the argument.
The logical form of an argument is the structure of the relationship between
its premises and its conclusion. It is the logical form of an argument that
determines whether it is possible for the conclusion of the argument to be false
at the same time that all of premises are true. If the logical form of an argument
prevents the conclusion from being false when all of the premises are true, then
the argument is valid. However, if the logical form of the argument does not
prevent the conclusion from being false at the same time all of the premises are
true, then the argument is invalid.
So, in order to understand why some arguments are valid and some are not,
you need to be able to identify the logical form of a given argument. Since
arguments consist of more than one sentence, you must look at the logical form
of each premise and the conclusion to find the logical form of the argument as
a whole. In other words, you must look at the logical form of each individual
sentence in the argument.
Now that you have learned how to identify the logical form of individual
sentences, you can use the same method to identify the logical form of a whole
argument. Here is an example:
CHAPTER 3. LOGICAL FORM 56

Premise: Either Plato invented logic (P ∨ A)

or Aristotle did.
(¬P )
Premise: Plato did not invent logic.
Conclusion: Aristotle invented logic. A

On the left, there is an argument in English with two premises. On the right,
I have written the logical form of the conclusion of the argument and each of
its premises. The sentence constant “P ” represents the simple English sentence
“Plato invented logic” and the sentence constant “A” represents the simple
English sentence “Aristotle invented logic”. If I replace the sentence constants
with the variables they substituted, I will be left with the following pattern of
argument:

(X ∨ Y )
(¬X)
Y

This pattern is common enough to have a name. Stoic logicians named it “Dis-
junctive Syllogism”. Disjunctive syllogism is a valid pattern of argument, which
means that no matter what logical forms are substituted for the variables, the
resulting argument will be valid. In the next chapter, you will learn why some
patterns of argument such as disjunctive syllogism produce valid arguments and
why some do not.

3.10 Summary
In this chapter, I have introduced to you the concept of logical form. You
have learned sentences may be simple or compound. A simple sentence does
not consist of other sentences as its parts. A compound sentence consists of
simpler sentences and at least one connective. Connectives come in two varieties:
linguistic and logical. A linguistic connective combines words, phrases, and
sentences into more complex forms of speech. A logical connective combines
simpler sentences into compound sentence according to a set of rules called
“formation rules”. These rules govern when and how logical connectives may
be used to build compound sentences out of simpler ones.
Logicians have developed a set of symbols that represent logical connectives
in languages and simple sentences in order to describe the logical form of a com-
pound sentence. Capital letters of the English alphabet other than “X”, “Y ”,
and “Z” are used to represent simple English sentences and three special sym-
bols called “tail”, “carrot”, and “wedge” are used to represent English logical
connectives.
Chapter 4

Truth-Functionality

In chapter 3, you learned about the concept of logical form. It is the logical form
of a compound sentence that determines whether it is consistent or inconsistent
with other compound sentences. Similarly, it is thanks to the logical form of
an argument that determines whether it is valid or invalid. Up to this point,
you have to learned to identify the logical form of compound sentences and
arguments. In this chapter, you will learn about what makes a logical form
true and what makes a logical form false which you need to know in order to
determine whether one or more compound sentences are consistent and whether
an argument is valid. By the end of this chapter, you will be able to construct
truth-tables for compound sentences and arguments. Truth-tables allow you to
do many things. They allow you to determine whether a compound is always
true, always false, or sometimes true and sometimes false. They also allow you
to determine whether one or more sentences is consistent or equivalent. Finally,
truth-tables can help you decide whether an argument is valid or invalid.

Chapter Objectives: By the end of this chapter, you will be able to

• Build truth-tables for the logical form of a single compound sentence

to determine whether it is a tautology, self-contradiction, or contin-
gent sentence
• Build truth-tables for the logical form of a pair of compound sen-
tences to determine whether they are are equivalent, contradictory,
consistent, or inconsistent
• Build truth-tables for the logical form of an argument to determine
whether it is valid or invalid.

57
CHAPTER 4. TRUTH-FUNCTIONALITY 58

4.1 Functions
Unlike other kinds of speech, a sentence is an expression that can be true or
false. When is a sentence true and when is it false? In chapter 1, I introduced
you to the correspondence theory of truth which says that a sentence is true
so long as it corresponds to some fact about the world. But how a sentence
corresponds to a fact depends on whether the sentence is simple or compound.
The truth-value of a compound sentence depends on its composition. Logi-
cians call this view “the theory of compositionality”:
Definition (The Theory of Compositionality). The truth-value of a compound
sentence is a function of the truth-values of the sentences of which it is composed.
You may be familiar with the concept of a function from mathematics where
the notation “f (x)” represents a function. In the context of mathematics, a
function is a relation between two values, an input and an output. For a given
input, a function produces one—and only one—output. The equation
f (x) = x2
is a function. If I substitute a value for the variable “x”, the function will
produce an output. For example, if I replace “x” with 2 in the formula “x2 ”,
then I will have “22 ” which is equal to 4. Thus, for the input “2”, the function
produces the output “4”.
Another way that mathematicians write functions is to use the variable “y”.
The function
f (x) = x2
could also be written as
y = x2
This way of writing a function makes it clear what the input and output of the
function are. The input is the variable “x” and the output is the variable “y”.
Whatever value I input into “x” will produce an output in “y”.
One property of functions is important to remember: functions do not pro-
duce different outputs for one and the same input. What I mean is that a
function such as “f (x) = x2 ” will never generate an output other than “4”
when “x” receives a value of “2”. Functions produce one and only one output
for each input. Different inputs may produce the same output but not the other
way around.
In language, compound sentences behave the same way functions do in math-
ematics. Certain inputs produce certain outputs. In particular, the truth-value
of a compound sentence is a function of the truth-value of the simpler sentences
that compose it. This means that whatever truth-values the simpler sentences of
a compound sentence have will determine the truth-value of the entire compound
sentence. Logicians refer to this kind of functionality as “truth-functionality”:
Definition (Truth-Functionality). A compound sentence is truth-functional if
and only if its truth-value is determined exclusively by the truth-values of its
component sentences.
CHAPTER 4. TRUTH-FUNCTIONALITY 59

Not all compound sentences are truth-functional. For example, the compound
sentence

I believe that God exists.

is not truth-functional because the truth-value of the entire sentence cannot

be determined exclusively by the truth-value of its component sentence, “God
exists”. Suppose this sentence is true. If it is true that God exists, does this
make the entire sentence “I believe that God exists” true? No, it does not
because it may be that God exists but I do not believe that God exists. Thus,
the truth-value of the sentence “God exists” does not tell you whether the
compound sentence “I believe that God exists.” is true or false. The reason is
because the compound sentence is not truth-functional.
Suppose now that the sentence “God exists” is false. Does this make the
entire sentence false? Again, the answer is “no” because it may be that God
does not exist, but I may mistakenly think he does. So, I would believe that God
exists even though God does not exist. In other words, the compound sentence
“I believe that God exists” would be true while the component sentence “God
exists” would be false.
What this example shows is that the compound sentence “I believe that
X” (where “X” is a simpler sentence embedded in the compound) is not truth-
functional; the truth-value of the simpler sentence does not determine the truth-
value of the compound sentence of which it is a part in cases like this one.

4.2 Truth-Conditions
Though not all compound sentences are truth-functional, negations, conjunc-
tions and disjunctions are. Consider three examples:

1. SCC is not in Spokane Valley.

2. The United States Congress consists of a house of representatives and a
senate.
3. Coffee is often served cold or hot.

When are these sentences true? The first is a negation of the sentence “SCC is
in Spokane Valley” and it is true if—and only if—SCC is not in Spokane Valley.
The second is a conjunction of the simple sentences “The United States Congress
consists of a house of representatives” and “The United States Congress consists
of a senate” and the second example is true if—and only if—congress does not
consist of only a house or a senate. The third example is a disjunction of the
sentences “Coffee is often served cold” and “Coffee is often served hot” and it is
true if—and only if—at least one of the disjuncts is true. Based on the analysis
of these examples, we can state generally that
CHAPTER 4. TRUTH-FUNCTIONALITY 60

A negation is true when its component sentence is false. A negation

is false when its component sentence is true.
A conjunction is true when its conjuncts are true. A conjunction is
false when at least one of its conjuncts are false.
A disjunction is true when at least one of its disjuncts is true. A
disjunction is false when both of its disjuncts are false.

Logicians call these general statements of the truth and falsehood of negation,
conjunction, and disjunction their “truth-conditions”. These conditions may be
stated precisely in the logical notation I introduced in chapter 3:

1. “X” and “Y ” are any simple or compound sentence.

2. “(¬X)” is true if and only if “X” is false.
3. “(X ∧ Y )” is true if and only if “X” is true and “Y ” is true.
4. “(X ∨ Y )” is true if and only if “X” is true or “Y ” is true, or both.

To determine whether a logical form is true or false, all we have to do is see

whether or not it satisfies one of the conditions above. For example, consider
the logical form
(¬P )
The main operator of this logical form is the tail. According to the truth-
condition for tail, this logical form is true if and only if “P ” is false. Otherwise,
the logical form is false.
Here is another example:
(P ∧ (¬Q))
The main operator of this logical form is the carrot. According to the truth-
condition for carrot, this logical form is true if and only if “P ” and “(¬Q)” are
true. The main operator “(¬Q)” is of course the tail. And so, like the previous
example, “(¬Q)” is true if and only if “Q” is false. If “Q” is indeed false, then,
according to the truth-condition for tail, “(¬Q)” is true. Given the truth of
“(¬Q)” and truth of “P ”, the logical form “(P ∧ (¬Q))” must be true as well.
Here is one more example. The logical form

(¬(P ∨ Q))

is true if and only if “(P ∨ Q)” is false. Otherwise, the logical form of the
example is false. Now, the main operator of (P ∨ Q)” is the wedge which means,
according to the truth-condition for wedge, that it would be false when both
“P ” and “Q” are false. So, for “(¬(P ∨ Q))” to be true, “(P ∨ Q)” must be false
and “(P ∨ Q)” is false when both “P ” and “Q” are false.
CHAPTER 4. TRUTH-FUNCTIONALITY 61

4.3 Truth-Tables
From the truth-conditions for tail, wedge, and carrot, we can construct tables
as visual representations of every possible truth-value the logical form of a com-
pound sentence can have. Such tables are called “truth-tables”:
Definition (Truth-Table). An arrangement of truth-values that defines for ev-
ery possible situation how the truth-value of a logical form of a compound sen-
tence or argument is determined by the truth-values of its sentence constants.
Here is what a truth-table looks like:
((A ∨ (¬ B)) ∧ B)
1 T T F T T T
2 T T T F F F
3 F F F T F T
4 F T T F F F

What is going on here? Though it may not be obvious at first, this truth-table
shows that the logical form of a conjunction is true when its conjuncts are true
and false in all other cases.

((A ∨ (¬B)) ∧ B)

at the top of the table is the logical form of a conjunction. “(A∨(¬B))” and “B”
are the logical forms of conjuncts of the conjunction and the carrot between them
is the main operator. Because a conjunction is a compound sentence, its truth-
value is determined ultimately by the truth-values of its simplest sentences.
The sentence constants “A” and “B” in the logical form represent the simplest
sentences in conjunction and each of them have one of two truth-values: either
True, “T”, or False, “F”, which are written in the columns underneath “A” and
“B” (Notice that “B” occurs twice in the logical form but the truth-values in
the columns under both occurrences of “B” are the same). Each row of the
truth-table represents a possible situation in which “A” and “B” have certain
truth-values. For example, the first row represents the situation where “A”
and “B” are both true; and the second represents the situation where “A” is
true and “B” is false. There are four rows in this table because there are four
possible combinations of truth-values that “A” and “B” could have. Since “A”
could either be true or false and since “B” could be true or false, it is possible
for them to be

1. true at the same time

2. false at the same time
3. one true and the other false

4. one false and the other true

CHAPTER 4. TRUTH-FUNCTIONALITY 62

So, “T”s and “F”s are written into each row in the columns under “A” and
“B” to indicate not only what their truth-values are but also to represent every
possible combination of truth-value they could have. If the logical form at the
top of the truth-table had only one sentence constant like “A”, then the table
would have only two rows due to the fact that “A” can only be true or false.
Based on the truth-values of “A” and “B” in each row, then truth-table lays out
what the truth-values of each connective are according to the truth-conditions
I described in section 4.2. This includes the truth-values of the main operator
of the logical form, the carrot, which I have marked in bold.
Building a truth-table for the logical form of a compound sentence requires
you to identify the total number of rows and columns needed based on the total
possible combinations of truth-values the sentence constants of the logical form
may have. The total number of rows in a truth-table is equal to the number of
possible combinations of truth-values for the sentence constants in the logical
form. The relationship between the number of different sentence constants and
the number of rows in the truth-table is expressed by the formula “R = 2n ”,
where “R” designates the number of rows in the truth-table and “n” the number
of different sentence constants. Except for the top, the truth-table above has
four rows. The reason is because there are two different sentence constants in
the logical form: “A” and “B”. If I substitute “2” for “n” in “2n ”, then I will
end up with “22 ” which is equal to 4, the number of rows in the truth-table.
Here is a simple method of constructing a truth table for any compound
sentence:
1. Identify “n”, the number of sentence constants in the given logical form.
2. Solve for “R” in the formula “R = 2n ” using “n”.
3. Draw rows equal to R under the given logical form.
4. Divide R in half, and enter “T” for “True” on the half of R rows under-
neath the first sentence constant and “F” for “False” on the remaining
half.
5. Divide R in half again, entering “T” on the number of rows according to
the new quotient underneath the second sentence constant and “F” on the
remaining rows.
6. Repeat step 5 until truth-values are entered for every simple sentence,
duplicating entries in columns with identical sentence constants.
7. Compute the truth-values of the connectives, starting with the one having
the narrowest scope and proceeding to the one with the widest scope
according to the truth-conditions for the tail, carrot, and wedge.
8. The truth-values in the column under the main operator indicates the
truth-values of the logical form as a whole.
Following this method, you can construct simple truth-tables for each of the
logical connectives:
CHAPTER 4. TRUTH-FUNCTIONALITY 63

The Truth-Table for the Tail

According to the truth-condition for the tail, the tail is true if and only if the
logical form it negates is false. The tail is false if and only if the logical form it
negates is true. Thus,
(¬ X)
1 F T
2 T F
The truth-table for the tail shows how the truth-values of any sentence having
the logical form of a negation “(¬X)” is determined by the truth-value of the
sentence that is negated “X”. The truth-table shows that “(¬X)” is false when
“X” is true and that “(¬X)” is true when “X” is false.

The Truth-Table for the Carrot

According to the truth-condition for the carrot, the carrot is true if and only if
both of its conjuncts are true. A carrot is false if and only if one or both of its
conjuncts is false. Thus,
(X ∧ Y)
1 T T T
2 T F F
3 F F T
4 F F F
This truth-table shows that the carrot is true when its conjuncts are true and
false in all other cases.

The Truth-Table for the Wedge

Finally, according to the truth-condition for the wedge, the wedge is true if and
only if one or both of its disjuncts is true. The wedge is false if and only if both
of its disjuncts are false. Thus,
(X ∨ Y)
1 T T T
2 T T F
3 F T T
4 F F F
The truth-table indicates that the wedge is true when one of the disjuncts is
true. Otherwise it is false.
These are the truth-tables for tail, carrot, and wedge. I have put them on a
single page in Appendix C in the back of this book so that you can consult them
easily whenever you need to. I encourage you to use this page as a reference
to help you remember the truth-conditions of these connectives when you build
truth-tables yourself.
CHAPTER 4. TRUTH-FUNCTIONALITY 64

4.4 Building Truth-Tables

The same method of building truth-tables described in the previous section
can be used to construct truth-tables for the logical forms of any compound
sentences. To illustrate how this method work, here is the example that I did
at the beginning of section 4.3:

(((A ∨ (¬B)) ∧ B)

This logical form has two different sentence constants: “A” and “B”. So, in
this case, n = 2. Now, solve for R in the equation:

R = 22
4=2x2

Since R = 4, the truth-table for the logical form will have four rows. I add an
extra row at the top where I can write out the logical form that I am working
on and each of its sentence constants. The total number of columns will be 6,
plus one for “A” and one for “B”. The result is the table

((A ∨ (¬ B)) ∧ B)
1
2
3
4

Now, I enter truth-values into the table. There are a variety of ways of doing
this, but I recommend that you start by writing alternating truth-values in each
row in the column under the right most sentence constant:

((A ∨ (¬ B)) ∧ B)
1 T T
2 F F
3 T T
4 F F

Notice that I have duplicated the truth-values I entered in the column under
“B” in all of the other columns for“B” in the table. The truth-values under a
sentence constant must be identical across each row.
Next, I write truth-values alternating every two rows in the column under
the next different sentence constant to the left of the first:

((A ∨ (¬ B)) ∧ B)
1 T T T
2 T F F
3 F T T
4 F F F
CHAPTER 4. TRUTH-FUNCTIONALITY 65

Again, I duplicated the truth-values written in the column under “A” under all
of the other “A” columns because the truth-values for “A” must be identical
across each row. It is worth pointing out that if there were another sentence
constant to the left of “A” in the table above, then you would write truth-values
alternating every four rows, though in this case the table would have eight rows
instead of four because the value of “R” would be 3 instead of 2. (23 = 8.)
Next, I enter truth-values in the columns under each connective according to
either the truth-conditions for the appropriate connective or the simple truth-
table for the connective presented in the previous section. To do this correctly,
I must start with the connective that has the least scope then proceed to the
connective that has the next least scope until I reach the connective that has the
widest scope, the main operator. In this case, the connective with the narrowest
scope is the tail. So, I begin with it. Following the truth-conditions for tail or
the simple truth-table for tail in section 4.3, I enter:

((A ∨ (¬ B)) ∧ B)
1 T F T T
2 T T F F
3 F F T T
4 F T F F

According to the truth-conditions for tail, the truth-values of the tail are the
contradictory of the truth-values of the sentence constant without the tail. For
example, since “B” is true in the first row, “¬” is false in that same row.
The connective with the next narrowest scope is the wedge. Following the
truth-conditions for wedge in section 4.3, I enter

((A ∨ (¬ B)) ∧ B)
1 T T F T T
2 T T T F F
3 F F F T T
4 F T T F F

According to the truth-conditions for wedge, a wedge is true when at least one
of its disjuncts is true. Otherwise it is false. In this case, “A” and “(¬B)” are
the disjuncts of the wedge. Since “A” is true in the first row, the wedge is true.
However, both “A” and “(¬B)” are false in the third row, which makes the
wedge false.
The remaining connective has the widest scope. So, it is the main operator
and tells us the truth-values of the entire logical form. It is a carrot. Following
the truth-conditions for carrot, I enter

((A ∨ (¬ B)) ∧ B)
1 T T F T T T
2 T T T F F F
3 F F F T F T
4 F T T F F F
CHAPTER 4. TRUTH-FUNCTIONALITY 66

The truth-values in the column underneath the carrot are bold to indicate that
these are the truth-values of the logical form’s main operator. According to the
truth-conditions for carrot in section 4.3, a carrot is true if both of its conjuncts
are true. Otherwise, it is false. “(A ∨ (¬B)) and “B” are the conjuncts in this
case. Thus, the carrot is true in the first row because both “(A ∨ (¬B)) and
“B” are true in that row as well. In all of the other rows, the carrot is false
because either “(A ∨ (¬B)) or “B” is false.
Here is an example with eight rows:
((A ∨ (¬ B)) ∧ C)
1 T T T
2 T T F
3 T F T
4 T F F
5 F T T
6 F T F
7 F F T
8 F F F

4.5 Contingency, Tautology, and Contradiction

What exactly have you learned by building truth-tables for the logical form of a
compound sentence? Quite a lot actually. Building a truth-table for the logical
form of a compound sentence can tell us whether the sentence is a
1. Contingency: Sometimes true and sometimes false
2. Tautology: Always true
3. Self-Contradiction: Always False
This might not seem to be very much. However, knowing whether a compound
sentence is contingent, tautological, or contradictory, can tells you whether it is
possible to prove that it is true. Neither a tautology nor a contradiction can be
directly proved or disproved, but a contingent sentence can be directly proved.
Let’s look at each type of sentence in turn.

Contingency
A contingent sentence is one that is true or false depending on the truth-value of
its simpler sentence. If at least one row in the column under the main operator
of a logical form is true and at least one false, then the logical form is contingent.
This logical form, for example, is contingent:
((A ∨ (¬ B)) ∧ B)
1 T T F T T T
2 T T T F F F
3 F F F T F T
4 F T T F F F
CHAPTER 4. TRUTH-FUNCTIONALITY 67

What this truth-table shows is that in some cases the logical form “((A∨(¬B))∧
B)” is true and in other cases it is false. Thus the logical form is contingent.

Tautology
A compound sentence is a tautology if it is true regardless of the truth-values of
its simpler sentence. In terms of a truth-table, this means that if every truth-
value in the column under the main operator of a logical form is true, then the
logical form is a tautology. This logical form, for example, is a tautology:

(¬ (A ∧ (¬ A)))
1 T T F F T
2 T F F T F

Notice that no matter what the truth-value of “A” is, whether it is true or false,
the logical form “(¬(A ∧ (¬A)))” is always true. Thus it is a tautology.
This particular tautology is an instance of the “law of non-contradiction”,
which states that no sentence can be both true and false at the same time. This
law is a fundamental law of logic.
A related tautology is worth knowing:

(A ∨ (¬ A))
1 T T F T
2 F T T F

This logical form is an instance of the “law of excluded middle” which states
that a sentence is either true or false, never both. Notice that “(¬A)” is false if
“A” is true and “(¬A)” is true if “A” is false. There are no rows in the truth-
table where “A” is both true and false. This possibility is “excluded”, hence
the name.
The law of excluded middle follows from the law of non-contradiction.

Self-Contradiction
A self-contradictory sentence is false regardless of the truth-values of its simpler
sentences. If every truth-value in the column under the main operator of a
logical form is false, then the logical form is a self-contradiction. This logical
form, for example, is a self-contradiction:

(A ∧ (¬ A))
1 T F F T
2 F F T F

This logical form is an instance of the opposite of the law of non-contradiction.

You can tell by comparing the truth-values in the columns under the main
operators of the two logical forms:
CHAPTER 4. TRUTH-FUNCTIONALITY 68

(¬ (A ∧ (¬ A))) (A ∧ (¬ A))
1 T T F F T 1 T F F T
2 T F F T F 2 F F T F

These logical forms are contradictory because they have opposite truth-value in
every row under their main operators and this is, in general, what it means for
two or more sentences to be contradictory. To be precise:
Definition (Logical Contradiction). The logical forms of two compound sen-
tences are contradictory if and only if they have opposite truth-values on each
and every row under the columns of their main operators.

4.6 Logical Consistency

In chapter 1, I explained that two or more sentences are consistent if they can be
true at the same time and inconsistent if they cannot be true at the same time.
In the case of a truth-table, if there is a row in which the truth-value of two
or more logical forms turn out to be true, then the logical forms are consistent.
If there is no row in which the main operators of both logical forms are true,
then the logical forms are inconsistent. These logical forms, for example, are
consistent:
(P ∨ Q) (P ∧ Q)
1 T T T T T T
2 T T F T F F
3 F T T F F T
4 F F F F F F

The truth-values of the main operators of both logical forms are true in the first
row of their respective tables. This makes them consistent.
On the other hand, these logical forms are logically inconsistent.

((¬ P) ∨ Q) (P ∧ (¬ Q))
1 F T T T T F F T
2 F T F F T T T F
3 T F T T F F F T
4 T F T F F F T F

In none of the rows of this table are the truth-values of the main operators of
these logical forms both true. This fact makes them logically inconsistent.
Another way to determine whether two or more sentences are logically con-
sistent is by joining the sentences together using a conjunction. In this case, the
conjunction of a group of inconsistent sentences will always be contradictory.
For example,
CHAPTER 4. TRUTH-FUNCTIONALITY 69

(((¬ P) ∨ Q) ∧ (P ∧ (¬ Q)))
1 F T T T F T F F T
2 F T F F F T T T F
3 T F T T F F F F T
4 T F T F F F F T F

The fact that this logical form is contradictory shows that its conjuncts cannot
both be true at time. In other words, the conjuncts are logically inconsistent.
The conjunction of a group of consistent sentences, on the other hand, will
never be contradictory:

((P ∨ Q) ∧ (P ∧ Q))
1 T T T T T T T
2 T T F F T F F
3 F T T F F F T
4 F F F F F F F

Both of the conjuncts of the logical form are true in the first row of the table.
Both “(P ∨ Q)” and “(P ∧ Q)” are true. Thus, the conjuncts are logically
consistent.

4.7 Logical Equivalence and Contradiction

In addition to comparing the logical forms of two or more compound sentences to
determine whether they are logically consistent or inconsistent with one another,
you can also compare the logical forms of two or more sentences to determine
whether they are logically equivalent. Here is a definition of logical equivalence:
Definition (Logical Equivalence). Two sentences are logically equivalent if and
only if they always have the same truth-value.
In terms of a truth-table, this definition tells you that the logical forms of two
compound sentences are logically equivalent when they have the same truth-
value in every row of the columns under their main operators. These logical
forms, for example, are equivalent:

(A ∧ B) (¬ ((¬ B) ∨ (¬ A)))
1 T T T T F T F F T
2 T F F F T F T F T
3 F F T F F T T T F
4 F F F F T F T T F

If you look at the truth-values in the column under the main operator of each
logical form, you’ll see that they match. In the first row of the table, the
truth-values in the column under the main operator of each form is “T”. In the
second, third, and fourth rows, the truth-values are both “F”. The fact that
CHAPTER 4. TRUTH-FUNCTIONALITY 70

these truth-values match in every row means that the two forms are logically
equivalent.
The contrary of logical equivalence is logical contradiction. As I mentioned
in section 4.6,
Definition (Logical Contradiction). Two sentences are logically contradictory
if and only if they always have opposite truth-values.
According to this definition, when the truth-table of the logical forms of two or
more compound sentence has opposite truth-values in every row of the columns
under their main operators, then the logical forms are logically contradictory.
For example, these logical forms are logically contradictory:

(A ∧ B) ((¬ B) ∨ (¬ A)))
1 T T T F T F F T
2 T F F T F T F T
3 F F T F T T T F
4 F F F T F T T F

Take a look at the truth-values in the column under the main operator of each
logical form. Notice that the truth-values of each logical form in every row are
the opposite of one another. In the first row of the table, the truth-values in the
column under the main operator of the first form is “T”, but the truth-value in
the column under the main operator of the second is “F”. In the second, third,
and fourth rows, the truth-values in the column under the main operator of the
first logical form are all “F” while the truth-values in the column under the
main operator of the second logical form are all “T”. Thus, the logical forms
are contradictory.

4.8 Arguments
In chapter 2, I defined logical validity as

Definition (Logical Validity). An inference is logically valid if and only if

the falsehood of its conclusion is logically inconsistent with truth of all of its
premises.
Truth-tables can be used to determine whether an argument is valid or invalid.
Building a truth-table for the logical form of an argument is a little different
from building a truth-table for the logical form of a single sentence. Because
an argument consists of more than one sentence, its truth-table must include
its premises and its conclusion in the same table—not in separate tables. For
example, the argument

Premise: ((P ∧ Q) ∨ Q)
Premise: (¬(P ∧ Q))
CHAPTER 4. TRUTH-FUNCTIONALITY 71

Conclusion: Q

has three separate logical forms, one for the conclusion and one for each premise.
So, its truth-table will look like this:

Premise Premise Conclusion

((P ∧ Q) ∨ Q) (¬ (P ∧ Q)) Q
1 T T T T T F T T T T
2 T F F F F T T F F F
3 F F T T T T F F T T
4 F F F F F T F F F F

Notice that I have put the logical form of the premises and the conclusion
together into the same table. Once I’ve finished building the truth-table, I can
use it to determine whether the argument is valid or invalid. According to the
definition of logical validity, if there is a row in the truth-table where the logical
forms of the premises are true and the conclusion false, then the argument is
logically invalid. If there is no such row, then the argument is logically valid. In
the truth-table above, there is no row where the main operators of both premises
are true and the conclusion is false. So the argument is valid.
Here is another example:

Premise: (P ∨ Q)

Premise: (¬Q)
Conclusion: P
Premise Premise Conclusion
(P ∨ Q) (¬ Q) P
1 T T T F T T
2 T T F T F T
3 F T T F T F
4 F F F T F F

If you look at each of the four rows of this truth-table, you will noticed that
there is no row in which the logical forms of both premises are true and the
conclusion is false. This means that this argument is valid. In other words, it
is impossible for the conclusion of this pattern of argument to be false while all
of the premises are true.

4.9 Counterexamples
Back in section 2.5, I introduced you to an informal way of proving that an
argument is invalid. This was the method of finding a counterexample to the
inference from the premises of an argument to its conclusion. A counterexample
to an argument shows that it is possible for its conclusion to be false at the
CHAPTER 4. TRUTH-FUNCTIONALITY 72

same time all of its premises are true. If an argument is valid, then there can
be no counterexample to the argument. But, if an argument is invalid, then
it is possible to come up with a example that describes a circumstance where
premises of an argument are true but the conclusion is false. Such an example
is a counterexample (see the definition in section 1.4).
It is possible to find a counterexample in a truth-table for an invalid argu-
ment. If a truth-table shows that the logical form of an argument is invalid,
then there will be a row in that table where the main operators of the logical
forms of the premises are true and the logical form of the conclusion is false.
For example,

Premise Premise Conclusion

(P ∨ Q) Q P
1 T T T T T
2 T T F F T
3 F T T T F
4 F F F F F

Look at the third row of the table. In this row, the premises “(P ∨ Q)” and
“Q” are both true and the conclusion “P ” is false. From this row, I can extract
a counterexample which proves that the logical form of the argument is invalid.
The counterexample is
(¬P ) ∧ Q
What does this mean? It means that when “P ” is false and “Q” is true, then the
premises of the argument will be true and the conclusion will be false. Since “Q”
is the second premise in the argument, the counterexample indicates that “Q” is
true. Similarly, since “P ” is the conclusion, the counterexample indicates that
the conclusion is false. What about the first premise “(P ∨ Q)”? You learned
earlier in section 4.2 that the wedge is true when at least one of the logical
forms it joins is true. Since “Q” is true in the third row of the truth-table, then
“(P ∨ Q)” must likewise be true.
How did I come up with this counterexample? To find a counterexample
in a truth-table for an invalid argument, all you need to do is look at a row
where the premises of the argument are true and the conclusion is false and
write down as a conjunction all of the sentence constants in the logical forms of
the premises and the conclusion, adding a tail to any of the sentence constants
that are false in that row. Here is one last example to make it is clear how to
find a counterexample in a truth-table:

Premise Conclusion
((P ∨ (¬ Q)) (((¬ Q) ∨ P )) ∧ Q)
1 T T F T F T T T T T
2 T T T F T F T T F F
3 F F F T F T F F F T
4 F T T F T F T F F F
CHAPTER 4. TRUTH-FUNCTIONALITY 73

Take a look at the second and fourth rows of this truth-table. In both of these
rows, the truth-value of the main operator of the premise is true and the truth-
value of the main operator of the conclusion is false. Consequently, I can extract
the counterexample
P ∧ (¬Q)
from the second row the counterexample

(¬P ) ∧ (¬Q)

from the fourth row. The first counterexample indicates that the logical form
of the premise is true and the logical form of the conclusion is false when “P ”
is true and “Q” is false. The second counterexample indicates that the logical
form of the premise is true and the logical form of the conclusion is false when
both “P ” and “Q” are false. This is because, according to the definition of
truth-functionality, the truth-value of the logical form of a compound sentence
depends exclusively on the truth-values of its sentence constants.

4.10 Summary
In this chapter, I have introduced you to the concept of truth–functionality, the
idea that the truth–value of a compound sentence is a function of its simplest
sentences. I explained how there are three main types of compound sentences
that are truth-functional:

1. Negations
2. Conjunctions
3. Disjunctions

The truth–value of a negation is a function of the truth–value of its simpler sen-

tences; it is true if and only if its component sentence is false. The truth–value of
a conjunction is a function of its conjuncts; it is true if and only if its conjuncts
are both true. The truth–value of a disjunction is a function of its disjuncts; it
is true if and only if one or both of its disjuncts are true.
These truth–conditions enable you to build visual representations of all the
possible truth–values for a negation, conjunction, or disjunction called “truth-
tables”. Truth–tables are extremely useful because they can help you determine
whether an individual compound sentence is contingent (sometimes true and
sometimes false), tautological (always true), or contradictory (always false).
They can also help you find out whether two or more compound sentences are
logically consistent (true at the same time) or logically equivalent (always has
the same truth–value. Finally, truth–tables can be used to determine whether
an argument is valid or invalid.
Chapter 5

Conditionals

In this chapter, I introduce to you the logic of conditional sentences. Conditional

sentences set a condition on when a sentence is true and when it is false. I will
explain what a conditional sentence is, show you how conditional sentences are
another type of compound sentence, and describe their parts. I will introduce
you to two new symbols that logicians use to represent conditional sentences:
the arrow and the double arrow. I will also describe for you the truth-conditions
for these symbols and show you how to build truth-tables for them.

Chapter Objectives: By the end of this chapter, you will be able to

• Use the arrow and the double arrow to write out the logical form of
conditional and biconditional sentence
• Build truth-tables for logical forms with the arrow or double ar-
row to determine whether the logical forms are tautologies, self-
contradictions, or contingencies

• Build truth-tables for the logical form of a pair of conditional sen-

tences to determine whether they are are equivalent, contradictory,
consistent, or inconsistent
• Build truth-tables for the logical form of an argument consisting of
one or more conditional sentences to determine whether it is valid or
invalid.

5.1 Conditional Sentences

A conditional sentence is a compound sentence with a conditional clause. In
English, a conditional clause is introduced by the word “if”. For example, the
sentence

74
CHAPTER 5. CONDITIONALS 75

If the sun rose every morning in the past, then the sun will rise tomorrow
morning.

is a conditional sentence because it contains the conditional clause “If the sun
rose every morning in the past”. By itself the conditional clause is an incomplete
sentence. It is sometimes called a “dependent clause” because it grammatically
depends on another clause which is “independent” to be a complete sentence.
An independent clause is another name for a complete sentence. In the example
above, the independent clause is “The sun will rise tomorrow morning”. With-
out the conditional clause, this independent clause could be a complete sentence
on its own but the conditional clause could not be.
Here is another more controversial example. The sentence

God is not benevolent if evil exists in the world.

is also a conditional sentence. Whether you agree with the conditional sentence
or not, the conditional clause is “If evil exists in the world” and the independent
clause is “God is not benevolent”.
Logicians use different terms to refer to the independent and dependent
clauses of conditional sentences. They call the dependent, conditional clause,
the part immediately after the word “if”, the “antecedent” and the independent
clause, the part that is not immediately after “if”, the “consequent”. Thus, the
antecedent of the conditional

If the sun rose every morning in the past, then the sun will rise tomorrow
morning.

is “the sun rose every morning in the past”. And the consequent of the con-
ditional is “the sun will rise tomorrow morning”. Notice that by dropping the
“if” from the conditional clause, we turn it into a complete sentence. This is
the antecedent of the conditional sentence.
Moreover, the antecedent of the conditional

God is not benevolent if evil exists in the world.

is “evil exists in the world”; and the consequent is “God is not benevolent”.
Notice that whether the word “if” is at the beginning of a sentence or at the
end, the antecedent always immediately follows “if”. There is one exception to
this which I discuss in section 5.4.

5.2 Necessary and Sufficient Conditions

Conditional sentences express necessary and sufficient conditions. Here is a
definition of each:
Definition (Sufficient Condition). A conditional sentence is a sufficient condi-
tion if and only if the truth of the antecedent guarantees the truth of the conse-
quent.
CHAPTER 5. CONDITIONALS 76

Definition (Necessary Condition). A conditional sentence is a necessary con-

dition if and only if its consequent cannot be true unless its antecedent is also
true.
Here is an example of a sufficient condition:

If Socrates is a human being, then Socrates is an animal.

In this example, the antecedent is a sufficient condition for the consequent.

What this means is that the truth of the antecedent is enough or “sufficient”
to guarantee that the consequent is true. If Socrates is a human being, then he
must also be an animal. Notice that saying the antecedent is “sufficient” to make
the consequent true does not mean that the only way the consequent is true will
be when the antecedent is true. If the antecedent is false, the consequent could
still be true for some other reason. Assuming the example is true, it may be
that Socrates is not a human being but an animal nonetheless; he may just be a
different kind of animal. In this case, the antecedent is false but the consequent
is true, though the antecedent is sufficient for the consequent.
Here is an example of a necessary condition:
S ocrates is a human being only if he is an animal.
In this case, being an animal is a necessary condition of Socrates being a human.
This is not to say that Socrates must be a human being because he is an animal.
Instead, all this sentence indicates is that Socrates cannot be a human being
without also being an animal.
By itself the word “if” indicates that a conditional sentence is a sufficient
condition. Putting the word “only” in front of “if” changes a conditional sen-
tence from a sufficient condition to a necessary one. Another way of putting it,
is that the word “only” modifies the condition in such a way that the antecedent
becomes the consequent. For instance, consider this conditional sentence:

Socrates is an animal if he is a human being.

Ordinarily, the antecedent of a conditional sentence is the conditional clause,

that is, the part of the sentence immediately after the word “if”; and in this
case the antecedent is the conditional clause after the word “if”. Yet, when we
add the word “only”, placing it in front of “if”, we get

Socrates is an animal only if he is a human being.

Think about what this sentence means and how it compares with the example
without the word “only”. The difference is that, in the first case, it will be true
that Socrates is a animal provided that it is true that he is a human being.
Being a human being is enough to make Socrates an animal. In the second case,
it will be true that Socrates is an animal only when he is a human being. In
other words, if it is true that Socrates is not a human being, then he cannot
also be an animal. You might think that this is just wrong: Socrates may not
CHAPTER 5. CONDITIONALS 77

be a human being, but he could still be an animal, and you are right. However,
thinking this way is to think that being a human being is a sufficient condition
for being an animal, not a necessary one.
Here is another example:

If you smoke, you will eventually get cancer.

As you now know, this conditional sentence cannot be divided into

(a) “If you smoke”, and

(b) “You will eventually get cancer”

because (a) is not a complete sentence, though (b) is. (a) is a sentence fragment.
Instead of dividing the sentence into (a) and (b) as I did above, it must divide
it into

(b) “You will eventually get cancer”, and

(c) “You smoke”

both of which are complete sentences. Dividing the example in this way leaves
out the words “If” and the word “then” which I explain below.
It is possible to combine a necessary and a sufficient condition. A combined
necessary and sufficient condition is called a “biconditional”. For instance, if I
take the sufficient condition

Socrates is a animal if he is a human being

and combine it with the necessary condition

Socrates is an animal only if he is a human being.

I get

Socrates is a animal if and only if Socrates is a human being.

Notice that I created a conjunction out of the conditional clauses in the sufficient
condition and in the necessary condition. I took the clause “if Socrates is a
human being” and the clause “only if Socrates is a human being” and put them
together with “and”. This biconditional sentence means being a human being is
both a necessary and sufficient condition for Socrates to be an animal. In other
words, being a human being is enough to guarantee that Socrates an animal
and a requirement of Socrates being an animal is that he is a human being.
You may be thinking here that it is false to say that Socrates must be a human
being in order to be an animal. After all, there are plenty of other things that
are not human beings but are animals. But, whether it is true or not that being
a human being is a necessary condition of being an animal is not the point here.
What matters is just what the conditional means and, whether it is true or not
CHAPTER 5. CONDITIONALS 78

that being an animal requires being a human being, the biconditional sentence
means that being a human being is required for Socrates to be an animal.
In sum, the phrase “if and only if” is an important logical phrase and you
will see it often because it is the one of the most common English phrases used
to captures biconditional sentences. When you see the phrase “if and only if”,
you can be sure that the sentence is a biconditional, a combined necessary and
sufficient condition.

5.3 The Arrow and Double Arrow

So, there are two kinds of conditions, necessary and sufficient, and they are ex-
pressed in English in conditional sentences which contain a conditional clause.
A necessary and a sufficient condition can also be combined into one bicondi-
tional sentence that expresses a combined necessary and sufficient condition.
Like negations, conjunctions, and disjunctions, conditional sentences are com-
pound sentences that consist of simpler sentences. This means that conditional
sentences have a logical form just like other compound sentences do.
Logicians use two symbols to represent conditional sentences. They are the
arrow “(→)” and the double arrow (“↔”). The arrow is used to represent a
sufficient condition and the double arrow is used to represent a combined nec-
essary and sufficient. There is no special symbol used to represent a necessary
condition. Instead, logicians use the arrow and switch the antecedent and the
consequent in a sufficient condition to represent a necessary condition. I’ll ex-
plain how this works in section 5.4. For now, let me show you how the arrow
and the double arrow work in greater detail.
The arrow represents English conditional phrases such as

• “if . . . , . . . ”
• “. . . only if . . . ”
• “. . . in case . . . ”

• “. . . provided that . . . ”
• “. . . given that . . . ”
• “. . . on condition that . . . ”

• “. . . implies . . . ”

.
The arrow joins the logical forms of no more than two simple or compound
sentences. The logical form that comes before the arrow is the antecedent. The
logical form that comes after the arrow is the consequent.
The arrow represents a sufficient condition. Consider again the conditional
sentence we look at before
CHAPTER 5. CONDITIONALS 79

If Socrates is a human being, then Socrates is an animal.

This sentence is an instance of a sufficient condition. Its antecedent is “Socrates
is a human being” and its consequent is “Socrates is an animal”. Because they
are simple sentences, we can use sentence constants to represent them. Let’s
use “H” for “Socrates is a human being” and “A” to represent “Socrates is an
animal”. Substituting these sentence constants for the simple English sentences
they represent, we get
If H, then A.
In section 5.2, I explained that “if” by itself represents a sufficient condition and
I mentioned above that logicians use the arrow to represent sufficient conditions.
So, you can use the arrow to represent the phrase “if . . . , then . . . ”. Thus,
(H → A)
Here, I must emphasize that the arrow does not simply replace the word “if”. In
English, the antecedent of a conditional sentence comes after the word “If”. The
simple sentence “Socrates is a human being” is the antecedent in the example
above because it follows the word “if”. In logical notation, on the other hand,
the antecedent of a conditional comes before the arrow. “H” is the antecedent
in the logical form above. This should make sense because “H” represents the
simple English sentence “Socrates is a human being” which is the antecedent of
the English conditional sentence. If the word “if” simply replaced the arrow,
then it would make sense to write this: “H if A”. Yet, this logical form is not
the form of the English sentence. In this example, “A” is the antecedent because
it comes after “if”. But, in the English sentence, the antecedent is “Socrates is
a human being” which “H” represents. By replacing the arrow with the word
“if”, I have inadvertently changed the consequent of the conditional sentence
into the antecedent. If I were, then, to substitute the simple English sentences
sentence constants for that represent them, I would get
Socrates is a human being if Socrates is an animal.
which does not have the same meaning as the original sentence. It may be easier
to see the difference if I wrote this conditional sentence another way:
If Socrates is an animal, Socrates is a human being.
Note the change in meaning. In the original sentence, the antecedent was
“Socrates is a human being”. In this sentence, it is “Socrates is an animal”. By
simply replacing the arrow with the word ‘if”, I have changed the condition of
the sentence, making the sentence altogether different.
Ultimately, when the arrow is used to represent the “if” in a sufficient con-
dition, it must not simply replace the word. The logical form of a conditional
sentence does not just capture the word order of a sentence; it captures the
meaning of the sentence. The arrow must be used in such a way that whatever
appears in front of it is the antecedent of the condition that the arrow represents.
Here is an another example to illustrate the point:
CHAPTER 5. CONDITIONALS 80

God is not benevolent if evil exists in the world.

As you know, this is an example of a sufficient condition. The simple English

sentences here are “God is benevolent” and “Evil exists in the world”. We can
use sentence constants to represent these simple sentences. Let’s use “B” for
“God is benevolent” and “E” for “Evil exists in the world”. If we replace the
simple English sentences with these sentence constants, we will get

It is not true that B if E.

There are two connectives left. The phrase “it is not true . . . ” is an English
phrase for a negation. So, we can use the tail to represent it:

(¬B) if E.

The word “If” represents a sufficient condition as does the arrow. So, simply
replacing the word “if” with an arrow would seem to make sense, as in this case

((¬B) → E)

But, doing so would be incorrect. The antecedent of the arrow in the logical form
above is “(¬B)” which represents the English sentence “God is not benevolent”.
However, in the English sentence, the antecedent is “evil exists in the world”
which “E” represents. By simply replacing the word “if” with an arrow, I have
changed the meaning of the English sentence in such a way that the logical
form does not represent it. Fixing this problem involves making sure that the
antecedent in the English sentence remains the antecedent in the logical form.
Since “E” represent the antecedent, it needs to go in front of the arrow. The
logical form of the conditional sentence is, therefore,

(E → (¬B))

5.4 The Arrow and Necessary Conditions

Earlier I mentioned that there is no special symbol that logicians use to represent
necessary conditions such as

Socrates is an animal only if he is a human being.

You might think that I could use the double arrow to represent a necessary
condition, but using a double arrow this way would be incorrect because, as you
will learn in the next section, the double arrow represents a combined necessary
and sufficient condition; it cannot be used to represent a necessary condition
alone.
How, then, do you go about writing out the logical form of a necessary
condition? Logicians use the arrow to represent necessary conditions as well
as sufficient conditions, but it requires them to switch the antecedent with the
consequent of the conditional sentence. Consider the example
CHAPTER 5. CONDITIONALS 81

Socrates is an animal only if he is a human being.

Here, the conditional clause “only if he is a human being” expresses a necessary

condition. What the conditional sentence means is that whether or not it is true
that Socrates is an animal depends only on whether it is true that he is a human
being. Being a human being, in other words, is a requirement of Socrates being
an animal. There is another way that we could write this conditional sentence:

If Socrates is an animal, then he is a human being.

This conditional sentence may appear completely different from the previous
one, but it really has the same meaning. This conditional sentence indicates
that being an animal is sufficient to make Socrates a human being. If Socrates
was not a human being, in other words, then he couldn’t also be an animal
according to this sentence.
What is important to understand about this last example is that a suffi-
cient condition can be logically equivalent to a necessary condition when the
antecedent and consequent are reversed. All I have done here is removed the
word only and changed the order of the simple sentences in the first example.
Instead of

Socrates is an animal only if he is a human being.

I have

If Socrates is an animal, then he is a human being.

and there is no change in the meaning of these sentences; they have the very
same meaning. This suggests two related things. First, the fact that I can do this
indicates that the word “only” modifies the “if” in a specific way in a necessary
condition. What “only” does is changes the antecedent of the condition to the
consequent. As you know, the antecedent of a conditional in English always
comes after the word “if”. So, you would expect the sentence “Socrates is a
human being” in the necessary condition above to be the antecedent of the
conditional. But it turns out that “only” turns it into the consequent, which
is why it is possible to write the necessary condition as the sufficient condition
in the way that I have done. This leaves the sentence “Socrates is an animal”,
which you would expect to be the consequent given that the antecedent of a
conditional follows “if”, as the antecedent. Thus, in the example,

Socrates is an animal only if he is a human being.

the antecedent of the conditional is the sentence “Socrates is an animal” and

the consequent is “Socrates is a human being”. This shows that it is possible
to write a necessary condition as a sufficient condition

If Socrates is an animal, then he is a human being.

CHAPTER 5. CONDITIONALS 82

Knowing that we can rewrite a necessary condition as a sufficient condition,

I can use the arrow to represent necessary conditions so long as I switch the
antecedent with the consequent. Going back once more to the example

Socrates is an animal only if he is a human being.

understanding that it is equivalent to

If Socrates is an animal, then he is a human being.

I may write out its logical form as

(A → H)

provided “A” represents the simple sentence “Socrates is an animal” and “H”
represents “Socrates is a human being”.

5.5 The Double Arrow

Logicians use the double arrow to represent biconditional sentences such as the
English sentence

Socrates is a animal if and only if Socrates is a human being.

In this example, the phrase “. . . if and only if . . . ” indicates that the sentence is
a biconditional. Other English phrases used to signal a biconditional sentence
are “. . . precisely if . . . ”, and “. . . exactly if . . . ”.
The double arrow joins no more than two simple or compound logical forms.
For example, the double arrow can join two sentence constants

(A ↔ B)

or compound logical forms such as

((¬A) ↔ (B ∨ C))

etc.
The double arrow is used to represent the combination of a necessary and
a sufficient condition. Because of this, whatever a double arrow joins together
is not only the antecedent; it is also the consequent. Suppose you are given
the logical forms “(A → B)” and “(B → A)”. These logical forms are the
same except for which sentence constant is the antecedent and which one is the
consequent. This implies that one represents a necessary condition while the
other represents a sufficient condition. If I combine these forms in one using a
carrot to conjoin them, I get

((A → B) ∧ (B → A))
CHAPTER 5. CONDITIONALS 83

and, in this case, “A” is the antecedent of one condition and the consequent
of the other. Similarly, “B” is the consequent of the first conditional and the
antecedent of the second. Each one is both an antecedent and a consequent, just
in difference logical forms. In short, the order of the antecedent and consequent
no longer matters, which is why the arrows point from one to the other. So, I
can combine these logical forms into

(A ↔ B)

using the double arrow instead of a carrot conjoining two arrows. In this way,
you can represent the necessary and the sufficient condition as a combined nec-
essary and sufficient conditional sentence or, in other words, a biconditional
sentence.
The table below summarizes what I’ve said about the arrow and the double
arrow:
Connective Name Function Represents
→ Arrow Conditional “If”
↔ Double Arrow Biconditional “if and only if”

5.6 Formation Rules for Arrow and Double Ar-

row
Up to this point, you have learned that logicians use sentence constants such as
“P ”, “Q”, or “R” to represent simple English sentences and logical connectives
such as the tail, the carrot, and the wedge to represent English logical connec-
tives. You learned in chapter 3 that there are rules for each of these connectives
that govern how they must be used in writing out the logical form of an English
sentence. These rules are called “formation rules”.
Like the tail, carrot, and wedge, there are formation rules for the arrow and
the double arrow. They are

1. “(X → Y )” is grammatically correct.

2. “(X ↔ Y )” is grammatically correct.

Adding these two rules to the ones you have already learned will give you a total
of six formations rules governing the use of these five logical connectives:

1. “X” and “Y ” by themselves are grammatically correct.

2. “(¬X)” is grammatically correct
3. “(X ∧ Y )” is grammatically correct.
4. “(X ∨ Y )” is grammatically correct.
5. “(X → Y )” is grammatically correct.
CHAPTER 5. CONDITIONALS 84

6. “(X ↔ Y )” is grammatically correct.

“X” and “Y ” above are, of course, variables which may take the logical form of a
simple sentence or the logical form of a compound sentence. Like the formation
rules for the tail, carrot, and wedge, the parentheses in grammatical rules (5)
and (6) above are also part of the grammar of arrow and double arrow. This
means that they must be included in any logical form which contains one of
these logical connectives.
You can use these rules to check whether a logical form is grammatically
correct. According to rule 1, the logical form

is grammatically correct.
According to rule 5, the logical form

(A → B)

is grammatically correct, but the logical form

→ AB

is not because it violates rule 5. The logical form

A→B

also violates rule 5 because it lacks parentheses which are needed to indicate
the scope of the arrow.
According to rule 6, the logical form

(A ↔ B)

is grammatically correct, but the logical form

AB ↔

is not because it violates rule 6. Neither is

↔ AB

nor is
A↔B
because it is missing a set of parentheses to indicate the scope of the double
arrow.
Like the tail, carrot, and wedge, the arrow and double arrow can be combined
with one another and with the other logical connectives to make the logical form
of a larger compound sentence. The formation rules can help you determine
CHAPTER 5. CONDITIONALS 85

whether any logical form you encounter is grammatically correct. For example,
the logical form
(A → (B ∧ C))
is grammatically correct according to rules 1, 3, and 5. You can check by
breaking the logical form down according to its main operator, checking to make
sure that it is formed according to the rules. For instance, the main operator of
the logical form
(A → (B ∧ C))
is the arrow. So the logical form consists of “A” and “(B ∧ C)”. “A” is gram-
matically correct according to rule 1. Rule 3 indicates that “(B ∧ C)” is correct.
The main operator of this logical form is the carrot and the carrot joins “B”
with “C”. Each of these sentence constants are grammatically correct according
to rule 1.
Similarly, the logical form

((¬A) ↔ B)

is grammatically correct according to rules 1, 2, and 6 because the main operator

is the double arrow and rule 6 indicates that the entire logical form is correct. If
it broken up into “(¬A)” and “B”, rule 2 indicates that “(¬A)” is correct and
rule 1 indicates that “B” is correct. “A” is also correct according to rule 1.
Furthermore, the logical form

(¬(A ∨ B))

is grammatically correct according to rules 1, 2, and 4. So is

(A ∨ (¬B))

and
((A ∨ B) ∧ (B ∨ C))
is grammatically correct because of rules 1, 3, and 4.
In sum, the formation rules for the arrow and the double arrow determine
whether or not a logical form containing one or more of these logical connectives
is grammatically correct.
It is worth remembering that the logical forms of compound sentences of-
ten contain multiple connectives. Whenever more than two sentence constants
appear in a logical form, you need to include parentheses to indicate the scope
of the logical connective that joins them. Remember from chapter 3 that the
scope of a logical connective consists of the the logical forms of the simple or
compound sentences it joins together. Commas, semicolons, words “either”,
“both”, and single predicates are clues to help you avoid ambiguity. The scope
of the main operator in a logical form consists of everything else in that logical
form. If there are no parentheses in the logical form, the main operator will lie
outside all parentheses. If there is more than one such connective, the main one
will not be a tail.
CHAPTER 5. CONDITIONALS 86

5.7 The Truth-Conditions of Conditionals

In chapter 4, I introduced you to the concept of truth-functionality. There,
I said that a compound sentence is truth-functional if and only if its truth-
value is determined exclusively by the truth-values of its parts. In other words,
the truth-value of a compound sentence is a function of the truth-values of its
simpler sentences.
Conditional sentences are compound sentence just like negations, conjunc-
tions, and disjunctions. And they are also like negations, conjunctions, and
disjunctions in that they are truth-functional: their truth-values can be deter-
mined exclusively by the truth-values of their parts. In the case of a bicondi-
tional sentence, its truth-value depends on the truth-value of the sentences it
combines; and, in the case of a conditional sentence, its truth-table depends on
the truth-values of its antecedent and consequent.
When is a conditional sentence true? A conditional sentence is true when
its antecedent is false or when its consequent is true or both. For example, the
conditional sentence

Socrates is an animal if he is a human being.

is true if it is either false that he is a human being or true that he is an animal,

the antecedent being “Socrates is a human being” and the consequent “Socrates
is an animal”.
On the other hand, a conditional sentence is false when its antecedent is true
and its consequent is false. The same example is false if it is true that Socrates
is a human being but false that he is an animal.
Notice that the example is not necessarily false when the antecedent is true.
It could be false when the antecedent is true, but it depends on whether the
consequent is true or false. Similarly, the example is not necessarily true when
the consequent is is false. It could be, but whether it is or not depends on
whether the antecedent is true or false.
What about biconditional sentences? When is a biconditional true? A bi-
conditional sentence is true when both of the sentences it combines have the
same truth-value. The biconditional sentence

Socrates is an animal if and only if he is a human being.

is true when it is true that Socrates is an animal and a human being or when it is
false that Socrates is a human being and that Socrates is an animal. Otherwise,
the example is false. If it turns out that Socrates is a human being but not an
animal, the biconditional sentence will be false; or, if Socrates turns out to be
an animal but not a human being, then the biconditional will be false. In short,
a biconditional sentence is false when its component sentences have different
truth-values.
Put in terms of the arrow and the double arrow, their truth-conditions are
1. “(X → Y )” is true if and only if “X” is false or “Y ” is true, or both.
CHAPTER 5. CONDITIONALS 87

2. “(X ↔ Y )” is true if “X” and “Y ” have the same truth-value—whether

true or false.
Remember that “X” and “Y ” are variables that can take the logical forms of
both simple and compound sentences. These truth-conditions indicate what the
truth-value of a logical form will be when the main operator of the form is an
arrow or a double arrow. For the logical form,

(A → B)

with an arrow as its main operator, whether it is true or not depends on the
truth-values of “A” and “B”. According to the truth-condition for the arrow,
if “A” is false or if “B” is true, then “(A → B)” will be true.
For the logical form,
(A ↔ B)
which has a double arrow as its main operator, whether it is true or not, again,
depends on the truth-values of “A” and “B”. According to the truth-condition
for the double arrow, if “A” and “B” have the same truth-value, whether they
are both true or both false, then “(A ↔ B)” will be true.

5.8 Truth-Tables for the Arrows

Based on the truth-conditions for the arrow and double arrow in the previous
section, you can build truth-tables for logical forms which includes arrows and
double arrows. Recall from chapter 5, that a truth-table is an arrangement of
truth-values that defines for every possible situation how the truth-value of a
compound sentence is determined by the truth-values of the simpler sentences
out of which the compound sentence is built. Here is the truth-table for the
arrow:
(X → Y)
1 T T T
2 T F F
3 F T T
4 F T F

A logical form with an arrow as its main operator is true when either the an-
tecedent is false or the consequent is true. In the table above, “X” is the
antecedent and “Y ” is the consequent. According to the truth-conditions for
the arrow, the arrow is true in the first row because the consequent, “Y ”, is
true. In the second row, the arrow is false because the antecedent, “X”, is true
and the consequent, “Y ”, false. In other words, the antecedent is not false and
the consequent is not true, which is contrary to the truth-conditions for the ar-
row. In the third row, the arrow is true because the antecedent is false and the
consequent is true, which is in accord with the truth-conditions for the arrow.
Finally, in the fourth row, the arrow is true because the antecedent is false. In
CHAPTER 5. CONDITIONALS 88

sum, what this table shows that the arrow is false when the antecedent is true
and the consequent false and is true in all other cases.
For the most part the truth-table for the arrow should make sense. But
sometimes people find the third and fourth rows strange. Why is the arrow
true in these rows? To answer this question, let’s take a look at a conditional
sentence in English. Suppose I make you the following promise

If you earn a perfect score on the final exam for this course, I will
give you $100.

The antecedent of this conditional sentence is “you earn a perfect score on the
final exam for this course” and the consequent is “I will give you $100”. Assume
that this conditional sentence is true. What would make it true? Suppose that
you do earn a perfect score on the final exam and I do give you $100. Then,
my promise to you that I will give you $100 if you earn a perfect score on the
final exam would be true. This situation is what the first row of the truth-table
represents.
Now, suppose instead that you do earn a perfect score on the final exam but
I don’t give you $100. What becomes of my promise to you? It is false because I
promised to give you $100 if you earned a perfect score and I didn’t. So I didn’t
keep my promise to you. This is the situation the second row of the truth-table
represents.
Next suppose that you do not earn a perfect score on the final exam and I
don’t give you $100. What does this tell you about my promise? It means that
my promise is true. The reason I didn’t give you $100 is because you didn’t
satisfying the condition for get it, namely, you didn’t earn a perfect score. The
fourth row of the truth-table represents this situation.
Finally, suppose that you do not earn a perfect score on the final exam but
I do give you $100. Is my promise true in this case? The answer is “yes”. My
promise to you is a sufficient condition. What this means is that earning a
perfect score on the final exam is one way to get $100 from me. But it isn’t the
only way to do it. Perhaps, I hire you to mow my lawn and pay you $100 for
doing so. This is what the third row of the truth-table represents. The arrow
is true in this row because the arrow represents a sufficient condition, which
means that the consequent could be true for some other reason, though the
antecedent may be false. If you looked at the third row and didn’t understand
why the arrow should be true when the antecedent is false and the consequent is
true, then you were probably thinking of the arrow as representing a necessary
condition. But, it doesn’t; it represents a sufficient condition.
Now, here is the truth-table for the double arrow.

(X ↔ Y)
1 T T T
2 T F F
3 F T F
4 F F T
CHAPTER 5. CONDITIONALS 89

The double arrow represents a combined necessary and sufficient condition. The
truth-conditions for the double arrow state that the double arrow is true when
the logical forms it joins have the same truth-value. Otherwise, the double
arrow is false. The truth-table above shows just that. The double arrow is true
when the two logical forms it joins “X” and “Y ” are both true as they are in
the first row and both false like they are in the fourth row. In the second row,
“X” and “Y ” have different truth-values. “X” is true while “Y ” is false. This
is why the double arrow is false in the second row. In the third row, “X” and
“Y ” again have different truth-values. “X” is false and “Y ” is true, which is
why the double arrow is false in this row.
You will need to use the truth-conditions for the arrow and the double arrow
or these truth-tables to build truth-tables for compound sentences involving the
arrow or double arrow. I explain step-by-step how to build a truth-table for a
compound sentence in section 4.4.
I have put these basic truth-tables for the arrow and double arrow together
with the truth-tables for the other connectives in appendix C in the back of this
book. I encourage you to consult this page whenever you need help building
truth-tables for yourself.

5.9 Building Truth-Tables for Arguments

As with any argument, you can build a truth-table for an argument with condi-
tional sentences as either its conclusion or its premises to determine whether the
argument logically valid or logically invalid. I explained how build a truth-table
for an argument in section 4.8 in chapter 4. The same procedure applies to
arguments which contain conditional sentences.
Conditional sentences offer another way to determine whether an argument is
logical valid or invalid. It is possible to write an argument as a single conditional
sentence where its conclusion is the consequent of the conditional and all of the
arguments premises are conjoined as the antecedent. For example, I can write
the argument
Premise: Socrates is either a philosopher or a logician.
Premise: Socrates is not a a logician.
Conclusion: Socrates is a philosopher.
as the conditional
If Socrates is either a philosopher or a logician and Socrates is not a logician,
then Socrates is a philosopher.
Notice that all I did here is take the premises of the argument, put them together
as conjunction with the word “and”, and made it the antecedent of a conditional
with the conclusion as the consequent.
I can do the same with the logical form of the argument. Given the inter-
pretation:
CHAPTER 5. CONDITIONALS 90

L: Socrates is a logician.
P: Socrates is a philosopher.

I can write the logical form of the premises and conclusion of the argument:

Premise: (L ∨ P )
Premise: (¬L)
Conclusion: P

Alternatively, I can use the arrow to write the logical form of the argument as
the logical form of a conditional

(((L ∨ P ) ∧ (¬L)) → P )

What is the advantage of rewriting an argument as a conditional sentence.

According to its truth-conditions, a conditional sentence is false only when its
antecedent is true and its consequent false. Notice that truth-conditions of a
conditional sentence are similar to when an argument is invalid. In chapter 2,
you learned that an argument is logical invalid when it is possible for all of its
premises to be true at the same time that its conclusion is false. Thus, if an
argument is valid, the conditional sentence whose antecedent is the premises
of the argument and whose consequent is the arguments conclusion will be a
tautology, that is, it will always be true. You can see this by writing out a
truth-table for the logical form of the argument

Premise Premise Conclusion

(L ∨ P) (¬ L) P
1 T T T F T T
2 T T F F T F
3 F T T T F T
4 F F F T F F

and comparing it to the truth-table of the logical form of the conditional sentence
that corresponds to the argument:

(((L ∨ P) ∧ (¬ L)) → P)
1 T T T F F T T T
2 T T F F F T T F
3 F T T T T F T T
4 F F F F T F T F

Did you notice that there is no row in the first truth-table where the conclusion
is false and both premises are true? The fact that the truth-table has no such
row means that the argument is logically valid. In the second truth-table, all
of the truth-values in the column under the arrow are true. This means that
the conditional sentence is a tautology. So, if an argument is valid, then the
CHAPTER 5. CONDITIONALS 91

conditional made up of argument’s premises and the argument’s conclusion will

be a tautology.
For contrast, a conditional sentence whose antecedent consists of a conjunc-
tion of the premises of a logically invalid argument and whose consequent is the
conclusion of that argument will not be a tautology. It may be either a self-
contradiction or a contingency. Here is an example. Consider this argument:

Premise: (A → B)

Premise: (¬A)
Conclusion: (¬B)

The truth-table for this argument is

Premise Premise Conclusion

(((A → B) (¬ A)) (¬ B)
1 T T T F T F T
2 F T T T F F T
3 T F F F T T F
4 F T F T F T F

Notice that this truth-table proves that the argument is invalid. In the last row,
the conclusion is false and both premises are true. So, there is a counterexample
to this argument
(¬A) ∧ (¬B)
Now, the corresponding conditional for the argument is

(((A → B) ∧ (¬A)) → (¬B))

Since the argument is invalid, you should expect that the truth-table for this
conditional sentence will not be a tautology—and it isn’t

(((A → B) ∧ (¬ A)) → (¬ B)
1 T T T F F T T F T
2 F T T T T F F F T
3 T F F F F T T T F
4 F T F T T F T T F

it is in fact a contingency, as the truth-table shows. Thus, if an argument is

invalid, then the conditional sentence made up of that argument’s premises and
conclusion will be not be a tautology; it could be either a self-contradiction or
a contingency.
CHAPTER 5. CONDITIONALS 92

5.10 Logical Equivalence and the Double Arrow

In chapter 4, I explained that you can use truth-tables to determine whether
two or more sentences are logically equivalent or contradictory by comparing
the truth-values in the column under their main operators. Two sentences are
logically equivalent if and only if they have the same truth-value on each line
under their main connectives. In a truth-table, the truth-values in the column
under the main operators of two sentences are the same in every row, then the
sentences are logically equivalent. If the truth-values are different in every row,
then the sentences are logically contradictory. These two sentences, for example,
are logically equivalent:

(P → Q) ((¬ Q) → (¬ P ))
1 T T T F T T F T
2 T F F T F F F T
3 F T T F T T T F
4 F T F T F T T F

Now, you have learned that the double arrow represents logical equivalence.
So, for any two logically equivalent sentences, the logical form created by join-
ing them together with a double arrow is a tautology. Thus, joining the two
sentences above with a double arrow yields

((P → Q) ↔ ((¬Q) → (¬P )))

which is a tautology. The following truth-tables proves it:

((P → Q) ↔ ((¬ Q) → (¬ P ))
1 T T T T F T T F T
2 T F F T T F F F T
3 F T T T F T T T F
4 F T F T T F T T F

This logical form is a tautology, which implies that the logical forms joined
by the double arrow are equivalent. Why is this? Remember the definition of
logical equivalence: two sentences are logically equivalent if and only if their
truth-values are always identical. Thus, if a logical equivalence is true, then
its component sentences will have the same truth-value whether that value is
“true” or “false”. On the other hand, if a logical equivalence is false, then its
component sentences will have opposite or contradictory truth-values; one will
be true while the other will be false.
Again, you learned in chapter 4, logical contradictions are contrary to log-
ically equivalent sentences. Two sentences are logically contradictory if and
only if they always have opposite truth-values. This definition implies that two
logical forms are logically contradictory when their truth-tables have opposite
truth-values on every row in the column under their main operators. For exam-
ple, these two sentences are contradictory:
CHAPTER 5. CONDITIONALS 93

(P → Q) (P ∧ (¬ Q))
1 T T T T F F T
2 T F F T T T F
3 F T T F F F T
4 F T F F F T F

The logical forms “(P → Q)” and “(P ∧ (¬Q))” are contradictory because the
truth-table shows that they always have opposite truth-values. The fact that
they always have opposite truth-values implies that the biconditional sentence
consisting the two sentences is a logical contradiction:

((P → Q) ↔ (P ∧ (¬ L))
1 T T T F T F F T
2 T F F F T T T F
3 F T T F F F F T
4 F T F F F F T F

This logical form of this biconditional sentence is a logical contradiction, which

means that the logical forms joined by the double arrow are logically contradic-
tory.

5.11 Summary
In this chapter, I have introduced you to the logic of conditionals. A conditional
is a compound sentence with a conditional clause and an independent clause.
In English, the word “if” introduces a conditional clause.
A conditional clause may be either a sufficient condition or a necessary con-
dition. A sufficient condition is one that, if satisfied, makes the independent
clause true. To put it precisely, if the conditional clause is true, then it guaran-
tees that the independent clause will also be true.
A necessary condition is one that must be true when the independent clause
is true. In other words, if the independent clause of a conditional sentence is
true, then the conditional clause must also be true.
A special kind of conditional sentence is called a “biconditional”. A bicondi-
tional is a compound sentence whose simpler sentences are conditional clauses
for each other. In English, the phrases “. . . if and only if . . . ” and “precisely if”
are commonly used in biconditional sentences. Outside of mathematics, logic,
and philosophy, it is unlikely that you will encounter a biconditional sentence.
Like negation, conjunction, and disjunction, conditional and biconditional
sentences are truth-functional, that is, their truth-value is a function of the
truth-value of its parts. With regard to conditional sentences, this means that
the truth-value of the conditional sentence as a whole is a function of the truth-
values of its conditional and independent clauses. If the conditional clause is
false or the independent clause is true, then the entire conditional sentence will
be true. For a biconditional sentence, if its component sentences have the same
truth-value, then the biconditional sentence as a whole will be true.
CHAPTER 5. CONDITIONALS 94

The single arrow is the logical connective that represents conditional sen-
tences with a sufficient condition. The logical form that appears in front of a
single arrow is called “the antecedent”. The antecedent represents the condi-
tional clause in a conditional sentence. The logical form that appears after a
single arrow is called “the consequent” and it represents the independent clause
of a conditional sentence. The double arrow is the logical connective that rep-
resents a biconditional sentence. Because the logical forms on either side of a
double arrow represent conditional clauses for the other clause, there are no
special names for them
There is no logical connective that specifically represents a necessary con-
dition. However, if a conditional sentence with a necessary conditional is rear-
ranged in the right way, the single arrow can be used to represent a necessary
condition.
Chapter 6

Simple Sentences

Here is a famous but trivial argument:

Premise: Socrates is a human being.

Premise: All human beings will eventually die.

Conclusion: Socrates will eventually die.

Is this argument valid? Let’s apply the skills we’ve learned so far to this argu-
ment. First, let’s write out the the logical form of the argument. Both of the
premises and are simple sentences. So, if we assign the sentence constants “H”
to the first premise and “D” to the second premise, we will have:

Premise: H
Premise: D

The conclusion is likewise a simple sentence. So, it too will be represented by a

sentence constant. Using “S” for the conclusion, we will have:

Conclusion: S

Thus the logical form of the argument is

Premise: H
Premise: D
Conclusion: S

Now let’s build a truth-table for this logical form to find out whether it is valid
or invalid. Here is the finished table:

95
CHAPTER 6. SIMPLE SENTENCES 96

H D S
1 T T T
2 T T F
3 F T T
4 F T F
5 T F T
6 T F F
7 F F T
8 F F F

According this truth-table, the logical form of the argument is invalid. The
second row of truth-table shows that it is possible for all of the premises to be
true and the conclusion to be false at the same time. A counterexample to the
validity of the argument is
(¬S) ∧ D ∧ H
because “D” and “H” are true in the second row and “S” is false. Does this
seem right to you? Look again at the argument in English:

Premise: Socrates is a human being.

Premise: All human beings will eventually die.
Conclusion: Socrates will eventually die.

Is this argument really invalid? It looks valid. After all, if it is true that
all human beings will eventually die and if it is true that Socrates is a human
being, then surely Socrates will eventually die too. If the argument is valid, then
something has gone wrong in the process of testing the argument for validity.
There are two possibilities: Either there is a mistake in the notation for the
logical form of the argument or a mistake in the truth-table. Yet, there is no
mistake in either one. So, what is the problem? What we are faced with here
is a conflict between our intuition about the validity of the argument and the
results of our logical analysis of that argument. This means that there are only
two possibilities: either our intuition is incorrect and the argument really is
invalid; or the argument is valid and the analysis is incorrect. But which is it?
The problem is not our intuition; the argument is valid. The problem is
the logical analysis. Take another look at the argument and compare it with its
logical form:

Premise: Socrates is a human being. H

Premise: All human beings will even-
tually die. D
Conclusion: Socrates will eventually
die. S
CHAPTER 6. SIMPLE SENTENCES 97

What do you notice? Do you see a problem? If you don’t, compare the English
premises with their logical forms and the English conclusion with its logical form.
In the English version of the argument, there is a relationship between each of
the premises and the conclusion. The first premise and the conclusion are both
about “Socrates”. The second premise and the conclusion assert that something
will eventually die; and the first and second premise have something to say about
“human beings”. So, each sentence of the English argument is related. Now look
at the logical forms. Do they indicate that there is a relationship between each of
the premises and the conclusion? They do not. This is the problem: the English
version of the argument indicates that there is a relationship between each of
its sentences, but the logical version does not. These relationships between the
sentences of the argument are part of its logical form. Yet the logical notation
fails to capture this feature of the logical form of the argument.
What all of this suggests is that the notation you have learned so far is
inadequate; it describes the logical form of the argument too superficially to
capture all of its relevant logical structure. What we need is a notation that can
capture the logical form of arguments like the example above, a notation that
captures the relationships between each simple sentence in the argument.
The kind of notation we need to describe the logical form of arguments like
the one above logicians call “predicate logical notation”.
In order to develop the kind of notation we need to describe the logical
form of arguments like the example above, we need to distinguish between three
different kinds of simple sentences:

• Singular Sentences

• Universal Sentences
• Particular Sentences

In the rest of this chapter, I explain each of these kinds of simple sentences and
the kind of notation required to capture the logical form of each one.

Chapter Objective: By the end of this chapter, you will be able to write
out the logical form of singular, universal, and particular sentences.

6.1 Singular Sentences

A singular sentence asserts that a specific person, place, thing, or time, has a
specific property. A property is a characteristic, trait, or attribute of an object
or a group of objects. The subject of a singular sentence is a proper noun that
refers to an individual person, place, object, or time, and the predicate is an
adjective which refers to the property attributed to the subject. Here is an
example:

Socrates is a human being.

CHAPTER 6. SIMPLE SENTENCES 98

The proper noun “Socrates” is the subject of this sentence and refers to the
person named “Socrates”. The phrase “. . . is a human being” is the predicate
of this sentence and refers to a group which has the property of being human.
(The three dots are called an “ellipsis” and just mean that I have left out the
subject of the sentence to make it easier to focus on the predicate which is the
part of the sentence that is leftover.) So, the example asserts that Socrates is a
member of the group of things which have the property of being human.
Logicians use lowercase letters (other than “x”, “y”, and “z”) to represent
proper nouns and capital letters (other than “X”, “Y ”, or “Z”) to represent
the predicates. These lowercase letters are called “individual constants” because
they are constants that represent the names of individual objects. The capital
letters are called “predicate constants”. Thus, I can write out the logical form
of the example above by assigning an individual constant such as “s” to the
proper noun “Socrates” and a predicate constant such as“H” to the phrase
“. . . is a human being”, which gives me the logical form:

Hs

You are probably wondering why I put the predicate constant in front of the
individual constant when, in English, the subject of the sentence comes before
the predicate. The reason is that a predicate is a function which can take
different values. I talked about functions in chapter 4 when I introduced you to
the logical concept of truth-functionality. You might remember that a function
in mathematics looks like this:
f (x)
Here, the function is labelled “f ” and “x” represents the value that the function
takes. Predicates are also functions, which is why logicians write out predicates
in the way that they do with the variable after the predicate constant:

Hs

The predicate constant “H”, in this case, is the function and the individual
variable “x” represents the value that the function takes. I will talk about
individual variables more in the next section.
Before I move on, let me show you another example:

Socrates will eventually die.

is a singular sentence. “Socrates” is again the subject of the sentence and I use
an individual constant “s” to represent it. So, “s” represents the proper noun
“Socrates”. The phrase “. . . will eventually die” is the predicate of the sentence
and I use a predicate constant “D” to represent it. Thus, “D” represents the
phrase “. . . will eventually die”. Using these constants, I can write out the logical
form of the example as
Ds
CHAPTER 6. SIMPLE SENTENCES 99

6.2 Universal Sentences

A universal sentence asserts that every member of a group is also a member of
a group of objects with a specific property. For example, the sentence

All human beings will eventually die.

is a universal sentence because it asserts that every member of the group of hu-
man beings are also members of the group that share the property of eventually
dying.
In a universal sentence, the subject is a common noun or adjective which
refers to a group of objects and the predicate is a common noun or adjective
which refers to a specific property. In the example above, for instance, the
subject is “human beings”. The word “human beings” is a common noun that
refers to a group of objects. The predicate, in this case, is “. . . will eventually
die”. This predicate is an adjectival phrase that refers to the specific property
of eventually dying. Because a universal sentence asserts that every member
of a group of objects has some specific property, the subject of the sentence
is marked with a universal “quantifier”. In the example, the English universal
quantifier “All” accompanies the common noun “human beings”. “All” is one
of several universal quantifiers in English. Other English universal quantifiers
include “every”, “no”, and “any”. English also routinely leaves out the universal
quantifier of universal sentences. For example, the sentence

Human beings will eventually die.

is commonly understood to be equivalent to

All human beings will eventually die.

So, if the subject of an English sentence is a common noun without a quantifier,

then it is safe to assume that it is a universal sentence and a universal quantifier
is implied.
Because both the subject and the predicate of a universal sentence are com-
mon nouns or adjectives, I use predicate constants to represent them. Thus, the
predicate

. . . will eventually die.

can be written with the predicate constant

What comes after “D”? In the singular sentence “Socrates will eventually die”,
“s” for “Socrates” followed “D” for “. . . will eventually die”. And, as I’ve said,
an individual constant or variable must always follow a predicate constant. Yet
here the subject of the sentence is not a proper noun. So, I cannot use an
individual constant such as “s” after the predicate constant because that would
give us the English sentence
CHAPTER 6. SIMPLE SENTENCES 100

Socrates will eventually die.

which is not the sentence whose logical form I am trying to write out.
I don’t know what the subject is at this point. Consequently, I will use an
individual variable such as “x” in this case to indicate that the subject has yet
to be identified. Individual variables differ from individual constants in that
the values of an individual variable are not fixed; their values are variable. The
value of individual constants, on the other hand, is fixed; they always specify
an individual person, place, or thing. Logicians conventionally reserve the last
three letters of the alphabet (“x”, “y”, and “z”) as individual variables.
I may not know much about the subject of the predicate “. . . will eventually
die”, but I do know that the subject is a member of the group “human beings”.
In other words, “x is a human being”. Like the predicate “. . . will eventually
die, which is written with the predicate constant “D”, the word “human beings”
is a common noun. So, it too can be written with a predicate constant. In this
case, I will use “H”. What comes after “H”? As before, the subject is not a
proper noun. So, an individual constant such as “s” cannot follow “H” in this
situation. Instead, I will use an individual variable to mark the fact that the
subject is not a specific person, place, thing, or idea. This gives me the logical
form “Hx”. Since the subject of “Dx” is restricted to the subject of “Hx”, this
gives me the logical form
(Hx → Dx)
In English, this logical form is read “If it is a human being, then it will eventually
die”. This doesn’t mean much because the subject, the word “it”, doesn’t refer
to anything. In other words, we don’t know who or what the logical form is
talking about. The problem is that the logical form is incomplete. Logicians
call expressions of this sort “formulae” because they have no definite subjects.
Logical formulae are just like mathematical formulae. In mathematics, a
formula is a mere pattern or structure that a calculation can take. Until a
number is put into the variable, no calculation can be done according to a
formula. Similarly, logical formulae are merely patterns that a logical forms
could take. Until it has a definite subject, formulae have no truth-value. I have
more to say about formulae in section 6.3.
The way to solve this problem is to recognize that the English sentence
“All human beings will eventually die” is not trying to saying anything about a
definite subject. Instead, it is talking about all human beings. So, like English,
we need a way to indicate that we are speaking indefinitely about every member
of a group of objects. The solution is to “quantify” the variable, to indicate
how many things the variable could take. Logicians say that the variables in a
formula must be “bound” to a logical quantifier to have a truth-value. There are
two logical quantifies: the universal quantifier and the existential quantifier. Let
me show you how the universal quantifier works before turning to the existential
quantifier.
CHAPTER 6. SIMPLE SENTENCES 101

The Logical Universal Quantifier

The logical universal quantifier is formed by placing an individual variable after
an upside-down “A”: ∀x. “∀x” is read as “for any x” or “for all x”. The symbol
“x” beside the universal quantifier here is not technically a variable. Instead, it is
a marker that indicates which individual variables in the rest of the logical form
are governed by the universal quantifier. Combining the universal quantifier
with the predicate constant “H” for the predicate “. . . is a human being” and
the predicate constant “D” for the predicate “. . . will eventually die”, we get
the logical form
(∀x(Hx → Dx))
This logical form says that “for all values of x, if x is a human being, then x will
eventually die”. Remember that the reason I used an arrow in this logical form
is because I wanted to restrict the subject of the predicate “D” to the subject
of the predicate “H”. In other words, the only things that will eventually die
are human being. There is a lesson here. In general, universal sentences are
conditionals. When a universal sentence predicates a property of a thing, it will
set a condition on what sort of thing its subject can be. Thus, by default, we
use the arrow to write out the logical form of universal sentences. The arrow
and the universal quantifier allow us to write out the logical form of universal
sentences in any language. To summarize, the logical form of the sentence

All human beings will eventually die.

can be written
(∀x(Hx → Dx))
Here are the steps I followed to write out the logical form of an English
universal sentence.
For any universal sentence with the form

All S are P

where “S” abbreviates the word “subject” and “P” abbreviates “predicate”,
first, rewrite the sentence to make the fact that it is a conditional explicit:

If any thing is an S, then it is a P.

Second, focusing on the subjects of the simple sentences in the antecedent and
consequent of the conditional, replace the subject of each simple sentence with
an individual variable such as “x”

If any x is an S, then x is a P.

Third, take the English universal quantifier out of the antecedent of the con-
ditional and move each of the individual variables to the other side of their
respective predicate constants like this

For any x, if Sx then P x.

CHAPTER 6. SIMPLE SENTENCES 102

Finally, replace the English universal quantifier with the logical universal quan-
tifier and the English conditional with an arrow

(∀x(Sx → P x))

Et voilá! I have successfully written out the logical form of the universal sen-
tence.

6.3 Formulae
Let me return now to formulae. As I said before, a formula is a pattern that a
logical form could take. Formulae make no definite assertion about any specific
thing in the universe, have no truth-value, and cannot be written as a meaningful
English sentence. Now that I have introduced the universal quantifier, I can
define formulae more precisely:
Definition (Formula). The expression that remains when a quantifier is re-
moved from a logical form.
For example, when I remove the universal quantifier from the logical form

(∀x(Sx → P x))

I get the formula

(Sx → P x)
The individual variables that remain in the formula after the quantifier has been
removed are called “free variables” because they are not bound by any quantifier.
Logical forms in which free variables occur are sometimes called “open” instead
of “formulae”, but these terms are synonymous. In contrast, the variables that
occur in logical forms with a quantifier are called “bound variables” and the
logical forms in which bound variables occur are sometimes called “closed”. For
example, this logical form
Fx
is a formula; it has no truth-value because I do not know the quantity of the
value of variable “x”. This makes “x” a free variable and the logical form open.
On the other hand, this form

(∀xF x)

is not a formula; it is the logical form of a simple sentence which means that it
has a truth-value. I know this because the variable “x” is bound by the universal
quantifier. Hence, the variable is a “bound variable” and the logical form itself
is closed.
Two things must happen for a quantifier to bind a variable. First, the
quantifier must indicate which variable it binds. In the logical form above the
“x” which appears immediately after the universal quantifier is not actually
CHAPTER 6. SIMPLE SENTENCES 103

a variable. It is a marker that indicates which variable the quantifier binds.

The real variable is the “x” that follows the predicate “F ”. This is why the
formulae “F x” has only one variable instead of two. When adding or removing
a quantifier from an expression the real variable remains whether it is free or
bound. The marker, however, comes and goes with the quantifier.
The importance of this first condition for a quantifier to bind a variable is
not as obvious in the examples you have seen up to this point. So consider this
one:
(∀x((P x ∧ M x) → Dy))
Is this logical form open or closed? Does it contain any free variables? The
answer is “yes”. The individual variable “y” is free, making the logical form a
formula. The “x” immediately after the universal quantifier indicates that the
quantifier only binds the variable “x”. In order to solve this problem I must
introduce another quantifier to bind the variable “y”. Doing so will yield the
following closed form:

(∀x(∀y((P x ∧ M x) → Dy)))

I could also have written the logicla form this way:

(∀x((P x ∧ M x) → (∀yDy)))

One way or the other, observe that there are no free variables left in the logical
form; “x” and “y” are now each bound by a quantifier.
You might wonder whether I could have solved the problem another way by
doing something like this

(∀xy((P x ∧ M x) → Dy))

Unless this logical form really means “(∀x∀y((P x ∧ M x) → Dy))”, the answer
is “no”. A quantifier can bind only one variable at a time. Different variables
must be bound by separate quantifiers.
The second condition that must be met for a quantifier to bind a variable
is that each instance of the variable a quantifier binds must occur within the
scope of that quantifier. Like connectives, quantifiers have scope. The scope of
a quantifier refers to the parts of an expression a quantifier governs. The scope
of the universal quantifier in

(∀x(Ax → Bx))

is
(Ax → Bx)
So, the variable “x” is bound in the first case but free in the second. Notice
also that the “x” immediately after the quantifier in the first case disappears in
the second.
CHAPTER 6. SIMPLE SENTENCES 104

Like the first condition, the importance of the second condition may not be
obvious. So, consider this example:
(∀x(P x ∧ M x) → Dx)
Is this logical form closed or open? Does it contain any free variables? It does,
though it may be hard to tell. The last variable in the logical form attached to
“D” is free because it occurs outside the scope of the universal quantifier. When
people see a quantifier at the beginning of a logical form, they often assume that
it is the main operator of the logical form. But this is not always true. In the
example above, the main operator is the arrow. Its scope is larger than the
scope of the universal quantifier. Only the part of the logical form between the
parentheses is within the scope of the quantifier. The arrow includes this part
as well as the other part “Dx”. In short, the entire logical form is within the
scope of the arrow but only part of the logical form is within the scope of the
quantifier. (See my explanation of scope back in chapters 3 and 4.) In the past
this hasn’t been a issue. But now the fact that the quantifier is not the main
operator creates a problem: a variable is left free and technically becomes a
different variable when it falls outside the scope of a quantifier. Even thought
there are only “x”s in the logical form, when some of them are bound by a
quantifier and some are not we end up with two different variables. Call them
“x1 ” and “x2 ”:
(∀x((P x1 ∧ M x1 ) → Dx2 ))
The consequence of this is a logical form that remains open. It ends up being a
formula and, remember, formulae have no truth-value.
Solving this problem is straight-forward. All we have to do is make sure that
“x2 ” occurs within the scope of the quantifier:
(∀x((P x ∧ M x) → Dx))
The second set of parentheses in this case extends the scope of the universal
quantifier all the way to the end of the logical form. Now it includes “Dx”.
This “x” is no longer different from the “x”s that occur earlier in the logical
form. They are now the same variable.
Ultimately what’s important about the difference between open and closed
logical forms comes down to whether all of the individual variables in the forms
are bound by some quantifier. If they are, then the logical form has a truth-
value; if they are not, then it does not. When you are writing out the logical
form of an English sentence, your goal is to obtain a closed logical form with
no free variables in it. Since the English sentence has a truth-value, its logical
form must have a truth-value as well.
To review how to write out the logical form of universal English sentences,
let me go over another example. For the universal sentence
All philosophers are logicians.
first I rewrite the English sentence to make the fact that it is a conditional
sentence explicit:
CHAPTER 6. SIMPLE SENTENCES 105

If any thing is a philosopher, then it is a logician.

Second, I replace the subjects of the antecedent and the consequent with an
individual variable:

If any x is a philosopher, then x is logician.

Third, I bring the universal quantifier “any” out of the conditional to make it
clear that it binds the variables in both the antecedent and the consequent of
the conditional:

For all x, if x is a philosopher, then x is a logician.

Fourth, I define predicate constants for the predicates in the antecedent and
consequent of the conditional:

P : “. . . is a philosopher”
L: “. . . is a logician”

Fifth, I replace the English predicates in the antecedent and the consequent
with the predicate constants that abbreviate them,

For all x, if P x then Lx

Sixth, I replace “if . . . then . . . ” with the arrow:

For all x, (P x → Lx)

Finally, seventh, I replace the phrase “For all x” with the universal quantifier:

(∀x(P x → Lx))

Before I finish, I need to check to make sure that there are no free variables
in the logical form. The universal quantifier binds the individual variable “x”
in this example and both occurrences of “x” in the logical form are within the
scope of the universal quantifier. There are no other variables. So I am done.
“(∀x(P x → Lx))” is the logical form of the English sentence “All philosophers
are logicians.”

6.4 Particular Sentences

A particular sentence asserts that at least one member of a group of objects
is a member of a group that shares has a specific property. For example the
sentence

Some human being will eventually die.

CHAPTER 6. SIMPLE SENTENCES 106

is a particular sentence because it asserts that at least one member of the group
“human being” has the property of eventually dying.
Like universal sentences, the subject of a particular sentence is a common
noun or adjective which refers to a group of objects and the predicate is a
common noun or adjective which refers to a specific property. In the example
above, the subject is “human being”. The word “human being” is a common
noun that refers to a group of objects. The predicate, in this case, is “. . . will
eventually die”. This predicate is an adjectival phrase that refers to the specific
property of eventually dying. Because a particular sentence asserts that at least
one member of a group of objects has some specific property, the subject of the
sentence is marked with an “existential quantifier”. In the example above, the
English existential quantifier “Some” accompanies the common noun “human
being”. “Some” is the only existential quantifier in English. So, when you see
the word “some” in a sentence, the sentence is always particular.
Though “some” is the only English existential quantifier, English often uses
the indefinite article “a” or “an” as an existential quantifier. The sentence

A human being will eventually die.

is equivalent to

Some human being will eventually die.

In general, if the subject of a sentence is accompanied by the indefinite article

“a” or “an”, you should treat it as a particular sentence. I must warn you that
this does not apply to the definite article “the”. This word turns a common
noun or adjective into a proper noun. For example, the subject of the sentence

The human being will eventually die.

refers to a specific human being. We don’t happen to know the name of this
human being but, if we did, then we could substitute the name of the human
being for the phrase “the human being” which would give us a singular sentence
such as

Socrates will eventually die.

Beside the word “some” and the indefinite article “a”, English has another
way of expressing a particular sentence. For example,

There is at least one human being who will eventually die.

In fact, this example of a particular sentence in English is what logicians believe

all particular sentences really mean. Thus the sentences

Some human being will eventually die.

and
CHAPTER 6. SIMPLE SENTENCES 107

A human being will eventually die.

can both be rewritten as

There is at least one human being who will eventually die.

and it is this form of a particular sentence which explains why the word “some”
is called the “existential quantifier”. In the last example, the sentence asserts
the existence of one human being who will eventually die. If there is no such
human being (if all human being were immortal), then the sentence would be
false. If, on the other hand, at least one human being who will eventually die
exists, then the sentence is true. In fact, all human being will eventually die.
But notice that saying there is at least one human being who will eventually
die is true even if all human being will eventually die. The particular sentence
makes a weaker claim in this case than the universal sentence which makes a
much stronger claim.
Another point worth mentioning about the example is the word “who”. This
word is a relative pronoun and serves to modify the word “human being”. Back
in chapter 3, I mentioned that relative pronouns such as “who”, “whom”, and
“which” are basically conjunctions. This means that the example could have
been rewritten as

There is at least one human being and he will eventually die.

Observe that I replaced the word “who” with the English conjunction “and”
and added the pronoun “he” to give the conjunct a subject. The significance
of this is that particular sentences are really conjunctions. When you write out
the logical form of a particular sentence, you must introduce a carrot by default:

There is at least one human being ∧ he will eventually die.

Now, because both the subject and the predicate of a particular sentence
are common nouns or adjectives, you must use predicate constants to represent
them just like I did with the subjects and predicates of universal sentences.
Thus, the predicate

. . . will eventually die.

is written with the predicate constant

followed by an individual variable such as “x”, indicating that the subject has
yet to be identified. You do the same with the predicate in the subject

. . . is a human being.
CHAPTER 6. SIMPLE SENTENCES 108

which is written with the predicate constant

Hx

Thus, you have

(Hx ∧ Dx)
In English, this formula says “it is a human being and it is going to die”. This
doesn’t mean much because the subject, the word “it”, doesn’t refer to anything;
we don’t know who or what the sentence is talking about. The problem is that
the logical form is open. The way to solve this problem is to recognize that the
English sentence “There is at least one human being who will eventually die” is
not trying to saying anything about an individual. Instead, it is talking about
some human beings in general. So, like English, there must be a way to speak
generally about some member of a group. The solution is to bind the variables
in the formula to a logical existential quantifier.

The Logical Existential Quantifier

The logical existential quantifier is formed by placing an individual variable
to the right of a backwards “E”: “∃x”. As with the universal quantifier, the
individual variable accompanying the existential quantifier is not a real variable;
it is a marker indicating which variables in a logical form within the scope of
the existential quantifier are bound by that quantifier. In English, “∃x” is
read “there is at least one thing such that . . . ”. The existential quantifier is
combined with the carrot to write out the logical form of particular sentences.
Thus, adding the existential quantifier to the formula above above, we get

(∃x(Hx ∧ Dx))

This logical form says that there is at least one thing such that it is a human
being and it will eventually die, which is another way of saying “some human
being will eventually die.”
Let me show you another example to review what you’ve learned. Consider
the particular sentence

Some logicians are philosophers.

First, you rewrite the sentence to make its logical structure clearer:

There is at least one thing such that it is a logician and it is a philosopher.

Notice that here I replace the word “some” with the phrase “there is at least one
thing such that . . . ” in preparation for using the logical existential quantifier.
Doing so also forces me to rewrite the rest of the sentence so that it is clear that
it is a conjunction and that “logician” is a predicate.
Second, you assign some predicate constants to represent the predicates
“. . . is a logician” and “. . . is a philosopher”:
CHAPTER 6. SIMPLE SENTENCES 109

L: “. . . is a logician”
P : “. . . is a philosopher”

Third, you replace the English predicates in the example with their equivalents.
To do this correctly, you must replace the word “thing” in the rewritten sentence
above with an individual variable “x” as well as the word “it” which is the
subject of each predicate because the subjects of these predicates refers back to
the word “thing”:

There is at least one x such that x is a logician and x is a philosopher.

Now, you can replace the English predicates with the predicate constants as-
signed to them:

There is at least one x such that Lx and P x.

Fourth, you replace the English conjunction “and” with the carrot along with
parentheses to indicate the scope of the carrot:

There is at least one x such that (Lx ∧ P x).

Finally, you replace the phrase “There is at least one x such that . . . ” with the
existential quantifier:
(∃x(Lx ∧ P x))
Before you are finished, you need to check to make sure that all of the vari-
ables bound by the quantifier are within its scope. In this case, they are. The
existential quantifier binds the individual variable “x” and both occurrences of
“x” in the logical form are within the scope of the quantifier as indicated by
the parentheses. Now you are done. “(∃x(Lx ∧ P x))” is the logical form of the
English sentence “Some logicians are philosophers”.

6.5 The Formation Rules for Quantifiers

Here now are the formation rules for using the quantifiers together with the
formation rules for the connectives you learned in chapters 3 and 5. For any
sentences “X” and “Y ” and for any individual variable “φ”,

1. “Xx” and “Y y” by themselves are grammatically correct.

2. “(¬Xx)” is grammatically correct.
3. “(Xx ∧ Y x)” is grammatically correct.
4. “(Xx ∨ Y x)” is grammatically correct.
5. “(Xx → Y x)” is grammatically correct.
6. “(Xx ↔ Y x)” is grammatically correct.
CHAPTER 6. SIMPLE SENTENCES 110

7. “(∀φXφ)” is grammatically correct.

8. “(∃φXφ)” is grammatically correct.
9. No other expressions are grammatically correct.

“X” and “Y ” above are predicate variables that stand for any predicate con-
stants such as “P ” and “Q”. “x” and “y” are individual variables that stand
for any individual constant such as “a” and “b”. What is new here is the “φ”.
I am using “phi” as a variable that takes individual variable such as “x” and
“y”. Call it a “metavariable”, a variable that takes other variables.
The purpose of these formation rules is to tell you which expressions are
grammatically correct and which ones are not. If an expression violates one of
these grammatical rules, then it is not a grammatically correct logical form. For
example, the expression
(∀x(P x ∨ Qx))
is grammatically correct according to rules 1, 4, and 7. It is correct according
to rule 7 because the entire sentence is correct; it is correct according to rule 4
because
(P a ∨ Qa)
is correct; and it is correct according to rule 1 because

Pa

and
Qa
are grammatically correct.
On the other hand, the expression

(P x ∧ Qx)

is not grammatically correct because it violates rules 3. According to rule 3,

the sentence
(P a ∧ Qa)
is grammatically correct. Similarly, the expression

(∃x(∀y(P x ∨ (¬Qy))))

is grammatically correct according to rules 1, 2, 4, 7, and 8. It is correct

according to rule 8 because the main operator of the expression is the existential
quantifier. It is correct according to rule 7 because the part

(∀y(P a ∨ (¬Qy)))

is grammatically correct. It is correct according to rule 4 because the part

(P a ∨ (¬Qa))
CHAPTER 6. SIMPLE SENTENCES 111

is grammatically correct. It is correct according to rule 2 because

(¬Qa)

is grammatically correct. And, finally, it is correct according to rule 1 because

Pa

and
Qa
are both grammatically correct.

6.6 Summary
The logical notation you learned in chapters 3 and 5 is not sensitive enough
to capture the logical form of many arguments because it cannot describe the
logical form of simple sentences. In this chapter, you have learned a more
sophisticated kind of notation that is sensitive to the logical form of simple
sentences.
There are three kinds of simple sentence: singular, universal, and particu-
lar. Singular sentences have proper nouns as their subjects; universal sentences
have universally quantified common nouns or adjectives as their subjects; and
particular sentences have existentially quantified common nouns as their sub-
jects. A proper noun is the name of a specific individual person, place, thing,
or idea and is represented by an individual constant such as “a”. An individual
constant is any lowercase letter other than “x”, “y”, or “z” which are reserved
as individual variables. A common noun is the name of a group of persons,
places, things, or ideas and are predicates represented by predicate constants.
A predicate constant is any uppercase letter such as “A”, “B”, or “C”, except
“X”, “Y ”, and “Z”.
Because universal sentences are conditionals, writing out their logical form
requires us to use arrows. Similarly,because particular sentences are conjunc-
tions, writing out their logical form correctly requires us to use carrots.
Chapter 7

Models

By now, you are well aware that a sentence is any expression which has a truth-
value, meaning that it can be either true or false and not both. You also learned
in the chapter 3 that sentences may be simple or compound. A simple sentence
does not consist of other sentences as its parts but a compound sentence does
consist of simpler sentences as well as at least one connective that joins the
simpler sentences together.
In chapter 4, you learned that the truth-value of a compound sentence is
a function of the truth-values of its simpler sentences. In other words, the
truth-value of a compound sentence depends exclusively on the truth-values of
its simpler sentences. There, I postponed discussion of the truth-conditions
of simple sentences until I explained the different kinds of simple sentences,
namely, singular, universal, and particular sentences. Having learned about
them in chapter 6, you are now ready to learn about their truth-conditions.
The purpose of this chapter is to introduce you to the truth-conditions of
singular, universal, and particular sentences.

Chapter Objectives: By the end of this chapter, you will be able to

• use models to determine when a singular, universal, or particular
sentence is true and when it is false.
• use models to determine when a simple or compound sentence is true
and when it is false.
• build models of the logical forms of simple and compound sentences
• build models of the logical forms of singular, universal, and particular
sentences.

• construct models of invalid arguments to prove that they are invalid.

112
CHAPTER 7. MODELS 113

7.1 Models
The truth-value of a simple sentence depends on the properties and relations of
objects in the world. A simple sentence is true in some situations, but false in
others. For example, the sentence “John Watson is a physician” is true if—and
only if—the object “John Watson” has the property of “‘being a physician”. If
the object “John Watson” does not have this property, then the sentence “John
Watson is a physician” is false.
Logicians build models of the properties and relations of objects which make
a sentence true. A model defines a domain of discourse and identifies the prop-
erties and relations of members of the domain. A domain of discourse consists
of a set of individual objects which are represented by individual constants. Lo-
gicians use curly brackets, “{“ and “}”, to represent sets of objects. A domain
of discourse which includes some of the characters in Arthur Conan Doyle’s
detective stories, for example, is written as

{h, w, m, l, b}

where “h” represents “Sherlock Holmes”, “w” represents “John Watson”, “m”
represents “Professor Moriarty”, “l” represents “Inspector Lestrade”, and “b”
represents “Inspector Bradstreet”. For this reason, a domain cannot have vari-
ables in it and cannot be empty. In this chapter, I discuss models that define the
properties of the objects in the domain as well as models that define relations
between objects.
A model defines the properties of an object in the domain by writing the
logical form of a singular sentence consisting of a predicate constant which
represents the property followed by an individual constant which represents the
object. For example, the logical form

Pw

indicates that “w” has the property “P ”. Logicians usually write a conjunction
of the logical forms of several singular sentences next to the domain of discourse
in a model to define the properties of individual objects and the relations that
objects have. For example, I would write “P w ∧P h” next to the domain: {h, w}
to define the properties of of “h” and “w”. Here is an example of a complete
model:
D : {h, w}; P w ∧ P h
The cursive “D” in the model is the symbol that logicians use to mark the
domain, which is everything between the curly brackets.
To determine whether the logical form of a simple sentence is true or false,
you must find out if it satisfies the requirements of some model. Without a
model, there is no way to determine whether the logical form of a simple sentence
is true or false. In the rest of this chapter, I will go through the truth-conditions
for the logical forms of the various kinds of simple sentences as well as the
different kinds of compound sentences. I begin with the logical form of singular
sentences.
CHAPTER 7. MODELS 114

7.2 Truth-Conditions of Singular Sentences

A singular sentence is true in a model if and only if the object to which its
subject refers has the property referred to by its predicate. For example, the
logical form
Dh
is true in the model:

D : {h, w, m}; Dh ∧ Dw ∧ Cm

because “h” has the property “D” in the model; “Dh”, in other words, appears
in the conjunction of predicates in the model.
On the other hand, a singular sentence is false if and only if the object that
its subject refers to does not have the property to which its predicate refers.
The logical form
Dm
is false in the same model because “m” does not have the property of “D” in the
model; there is, in other words, no “Dm” listed in the conjunction of predicates
in the model.
Here’s another example. The logical form

Cw

is also false in the same model because “w” does not have the property “C” in
the model.
Finally, the singular logical form

Ch

is false in the same model because “h” does not have the property “C” in the
model.

7.3 Truth-Conditions of Universal Sentences

A universal sentence is true in a model if and only if all of its substitution
instances are true. What is a substitution instance? A substitution instance is
Definition (Substitution Instance). The singular sentence obtained by remov-
ing a quantifier from a universal or particular sentence and substituting an indi-
vidual constant from the domain of a model for every individual variable bound
by that quantifier.
A substitution instance of a universal sentence is a singular sentence that sub-
stitutes every individual variable bound by the universal quantifier with an
individual constant from the domain of a model. Here is an example of what I
mean. Given the domain,
D : {h, w}
CHAPTER 7. MODELS 115

the substitution instances for

(∀xP x)
are
Ph
Pw
There are no other substitution instances for “(∀xP x)” in this domain. “P h”
is a substitution instance of “(∀xP x)” because “h” is in the domain and can
be substituted for the individual variable “x” in “P x”, dropping the universal
quantifier from the logical form. “P w” is a substitution instance of “(∀xP x)”
because “w” is in the domain and can be substituted for “x” in “P x” after
removing the universal quantifier.
Is “(∀xP x)” true or false? Again, a universal sentence is true in a model
if and only if all of its substitution instances are true in that same model. So,
“(∀xP x)” is true if “P h” and “P w” are true in the model. However, the domain
given above is not a model. It doesn’t indicate what the properties and relations
of its objects are. So, it is impossible to tell whether “P h” and “P w” are true
or false and it is, therefore, unclear whether “(∀xP x)” is true or false.
If you are not sure what the problem is here, compare this example to the
last one. The logical form
Cw
is false in the model

D : {h, w, m}; Dh ∧ Dw ∧ Cm

because “Cw” is not in the list of predicates in that model. In the model

D : {h, w}

on the other hand, there is no list of predicates given after the domain. So, there
is no way to determine whether “P h” or “P w” is true or false in the model.
Now, consider this model:

D : {h, w}; Dh ∧ Dw

This model defines a domain as well as the properties of its members. Now
consider the following logical form:

(∀xDx)

Is it true or false? The answers depends on whether its substitution instances are
true. Given the domain in the model, “(∀xDx)” has two substitution instances.
They are
Dh
Dw
CHAPTER 7. MODELS 116

“Dh” is true because the model defines the object “h” as having the property
“D”. “Dw” is also true because the model defines the object “w” as having
the property “D”. Since both substitution instances are true in the model,
“(∀xDx)” is also true.
Now consider the same logical form again

(∀xDx)

but also consider a different model:

D : {h, w}; Dh

Is “(∀xDx)” true or false in this model? Its substitution instances are the same:

Dh

Dw
The model defines “h” as having the property “D”, but it does not define “w”
as having the property “D”. So, “Dh” is true in the model but “Dw” is not.
Thus, not all of “(∀xDx)”s substitution instances are true in the model, making
“(∀xDx)” false.

7.4 Truth-Conditions of Particular Sentences

A particular sentence is true in a model if and only if at least one of its substi-
tution instances is true. Consider this example:

(∃xP x)

Given the domain

D : {h, w}
the logical form “(∃xP x)” has two substitution instances. They are

Ph

Pw
There are no other substitution instances for “(∃xP x)” in this domain because
there are only two objects in the domain.
Is “∃xP x” true or false in this model? The model does not define the
properties of “h” or “w”. Technically, this means that none of the objects in
the domain have the property “P ”. Thus “P h” and “P w” are both false in this
model. None of the substitution instances for “(∃xP x)” are true which indicates
that “(∃xP x)” is false.
Here is another example. Suppose you are given the model

D : {h, w}; Dh
CHAPTER 7. MODELS 117

Is the logical form

(∃xDx)
true or false in this model? “(∃xDx)” has two substitution instances in this
domain:
Dh
Dw
“Dh” is true because the model defines “h” as having the property “D”. On the
other hand, “Dw” is false because the model does not define “w” as having the
property “D”. Now, the logical form of a particular sentence is true in a model
if and only if at least one of its substitution instances is true. This means that
“(∃xDx)” will be true in the model if at least one of its substitution instances
is true in the model as well. Since “Dh” is a substitution instance of “(∃xDx)”
and it is true in the model, “(∃xDx)” is true in the model as well. The fact
that “Dw” is false in the model does not change the truth of “(∃xDx)” because
it is the logical form of a particular sentence.
Let’s change the model and see what happens to “(∃xDx)”. Suppose you
have the model
D : {h, w}; Dh ∧ Dw
Is “(∃xDx)” true or false in this model? The domain in this model is the same
as in the previous one. So, “(∃xDx)” will have the same substitution instances:

Dh

Dw
However, in the model, both “Dh” and “Dw” are true because the model de-
fines “h” and “w” as having the property “D”. What happens to “(∃xDx)”?
Nothing. Since at least one of its substitution instances is true in the model,
“(∃xDx)” is true as well. The fact that both of its substitution instances are
true changes nothing.
Let’s look at one more example. Suppose you are given this model

D : {h, w};

Is the logical form “(∃xDx)” true or false in this model? “(∃xDx)” has two
substitution instances. They are
Dh
Dw
Neither “Dh” nor “Dw” is true because the model defines neither “h” nor “w”
as having the property “D”. In fact, since none of the objects in the domain
have the property “D”, there are no objects with the property. So, none of this
logical form’s substitution instances are true in the model.
CHAPTER 7. MODELS 118

Universal and Particular Sentences: A Compari-

son
Compare the truth-conditions of universal and particular sentences. Notice
that when a universal sentence is false, at least one corresponding particular
sentence is not true; and if the particular statement is false, then a corresponding
universal statement is true. Consider, for example, this sentence:
Everyone is a detective.
The logical form of this English sentence is
(∀xDx)
where “D” represents the predicate “. . . is a detective”.
Now suppose you have this model:
D : {h, w}; Dh
Is “(∀xDx)” true or false in this model? “(∀xDx)” has two substitution in-
stances:
Dh
Dw
According to the model, “h” has the property “D”. So, “Dh” is true. “w”,
however, does not have the property “D”. So, “Dw” is false. In order for
“(∀xDx)” to be true in the model, all of its substitution instances must be
true, but only one of them is true. Thus, “(∀xDx)” is false; at least one of its
substitution instances is false in the model. A logically equivalent way of saying
that “‘(∀xDx)” is false’ is to say that “‘(¬(∀xDx))” is true’, which, in turn is
equivalent to saying that “‘(∃x(¬Dx))” is true’.
Here is another sentence:
Someone is a detective.
The logical form of this example is
(∃xDx)
Now, suppose you have the model:
D : {h, w};
Is “(∃xDx)” true or false in this model? “(∃xDx)” has two substitution in-
stances in this domain:
Dh
Dw
Neither “h” nor “w” have the property of “D” in the model. So, neither “Dh”
nor “Dw” is true. Since a particular sentence is true only if at least one of its sub-
stitution instances is true and none of the substitution instances for “(∃xDx)”
is true, then “(∃xDx)” is false in this model. In other words, “(¬(∃xDx))” is
true and “(∀x(¬Dx))” is true.
CHAPTER 7. MODELS 119

7.5 Truth-Conditions of Compound Sentences

Recall from chapter 4 that the truth-value of a compound sentence is a function
of the truth-values of its simpler parts. In other words, the truth-value of a
compound sentence depends only on the truth-values of its simpler sentences.
Models complicate the truth-conditions of compound sentences but not a lot.
Let’s look at an example of a negation, conjunction, disjunction, conditional,
and biconditional to learn how you can use models to determine the truth-value
of compound sentences.
First, here is an example of an English negation:

John Watson is not a physician.

The logical form of this sentence is “(¬P w)”. Now suppose you have this model:

D : {h, w}; Dw ∧ Dh ∧ P w

Is “(¬P w)” true or false in this model? It is false because “P w” is true in the
model.
Second, here is an example of an English conjunction:

John Watson is a physician and a detective.

The logical form of this sentence is “(P w ∧ Dw)”. Suppose you have the same
model as before:
D : {h, w}; Dw ∧ Dh ∧ P w
Is “(P w ∧ Dw)” true or false in this model? It is true because both “P w” and
“Dw” are true in the model.
Third, here is an English disjunction:

John Watson is a criminal or a detective.

The logical form of this sentence is “(Cw ∨ Dw)”. In the model

D : {h, w, m}; Dh ∧ Cm

“(Cw ∨ Dw)” is false because both “Cw” and “Dw” are false in the model.
Fourth, here is an English conditional:

If John Watson is a physician, then he is a criminal.

The logical form of this sentence is “(P w → Cw)”. In the model:

D : {h, w}; P w ∧ P m

“(P w → Cw)” is false because “P w” is true and “Cw” is false in the model.
Finally, here is an English biconditional:

John Watson is a detective if and only if Sherlock Holmes is a detective.

CHAPTER 7. MODELS 120

The logical form of this sentence is “(Dw ↔ Dh)”. In the model:

D : {h, w}; Dh ∧ Dw ∧ Cm

“(Dw ↔ Dh)” is true because both “Dw” and “Dh” are true in the model.
What happens when a compound sentence includes quantifiers? Here is an
example of an English universal disjunction:

Everyone is either a detective or a criminal.

The logical form of this sentence is “(∀x(Dx ∨ Cx))”. Suppose you are given
the model:
D : {h, w, m}; Dh ∧ Dw ∧ Cm
Is “(∀x(Dx ∨ Cx))” true or false in this model? Because “(∀x(Dx ∨ Cx))” is
the logical form of a universal sentence, you must look at all of its substitution
instances to find out if it is true or false in the model. According to the model’s
domain, “(∀x(Dx ∨ Cx))” has three substitution instances. They are

(Dw ∨ Cw)

(Dh ∨ Ch)
(Dm ∨ Cm)
As you know, a universal sentence is true in a model if and only if all of
its substitution instances are true in the same model. In the model above,
“(∀x(Dx ∨ Cx))” is true because all of the substitution instances are true in
the model. All of the substitution instances are disjunctions. So, if at least one
disjunct is true, then the entire disjunction will be true. The first substitution
instance is true in the model because “w” has the property “D” in the model,
making “Dw” true. The second substitution instance is true in the model be-
cause “h” has the property “D” in the model, which makes “Dh” true. The
third substitution instance is true in the model because “m” has the property
“C”, making “Cm” true in the model.
Here is an example of an English existential negation:

Someone is not a detective.

The logical form of this sentence can be written as “

(∃x(¬Dx))

” Given the model

D : {h, w, m}; Dh ∧ Dw ∧ Cm
“(∃x(¬Dx))” will have three substitution instances:

(¬Dw)

(¬Dh)
CHAPTER 7. MODELS 121

(¬Dm)
Now a particular sentence is true in a model if and only if at least one of its
substitution instances is true in the same model. In the model given above,
“(∃x(¬Dx))” is true because at least one of the substitution instances is true in
the model, namely “(¬Dm)”. “(¬Dm)” is true in the model because “m” does
not have the property “D” in the model, that is, “Dm” does not appear in the
conjunction of predicates in the model. So, “Dm” is false; or, in other words,
“(¬Dm)” is true.

7.6 Working with Models

In chapters 4 and 5, you learned about truth-tables and how they can be used
to determine whether a compound sentence is a tautology, self-contradiction, or
contingency; whether a pair of compound sentences are logically consistent or
inconsistent, or equivalent or contradictory; and whether an argument is valid
or invalid.
As useful as truth-tables are, they unfortunately cannot tell you anything
about simple sentences because they are designed around the concept of truth-
functionality which I also introduced you to in chapter 4 and truth-functionality
is a property of compound sentences.
Fortunately, you have now learned about models. Like truth-tables, models
can give you useful information about simple sentences and how they relate to
one another. Models can tell you not only when the logical form of a simple
sentence is true or false; they can also tell you whether a pair of logical forms
are consistent and whether the logical form of an argument consisting of simple
sentences is invalid. Unlike truth-tables, there are certain limits to what models
can do. While a model can prove that the logical form of a pair of simple
consistent sentences really are consistent, a model cannot prove that a pair
of inconsistent simple sentences are, indeed, inconsistent. Moreover, though a
model can tell you that an argument is invalid, it cannot prove that an argument
is valid.
In this section, I will explain how to use models to determine whether the
logical form of a pair of simple sentences are consistent and whether the logical
form of an argument is invalid. Along the way, I will explain why models cannot
prove that the logical form of a pair of simple sentences are inconsistent or that
the logical form of an argument is valid.

Logical Consistency
As you know from chapter 1, one or more sentences are logical consistent if and
only if they can be true at the same time. If two or more sentences cannot be true
at the same time, then they are not logically consistent or just “inconsistent”.
In the case of simple sentence, if two or more simple sentences are consistent,
then a model will be able to demonstrate that their logical forms are consistent.
CHAPTER 7. MODELS 122

Consider, for example, the logical form of a pair of singular sentences:

Pa

Qb
It may be obvious to you at this point that these logical forms are consistent.
But the obviousness of their consistency is not proof. To demonstrate that the
pair of logical forms are consistent, I can build a model in which both of them
are true:
D : {a, b}; P a ∧ Qb
According to this model, “P a” and “Qb” are both true. Because they are true
in the same model, the model proves that it is possible for them to be true at the
same time, which according to the definition of logical consistency in chapter 1
means that they are consistent.
If the logical form of two or more universal or particular sentences are consis-
tent, then models can also prove that they are. However, in the case of universal
and particular sentences, substitution instances can make models more compli-
cated. Consider the logical forms of a pair of universal sentences:

(∀xP x)

(∀xQx)
These logical forms are consistent and a model can prove it:

D : {a}; P a ∧ Qa

According to this model, each logical form has one substitution instance. For
“(∀xP x)”, it is
Pa
and, for “(∀xQx)”, it is
Qa
These substitution instances are listed in the same model which means that they
are both true. Since all of the substitution instances for both logical forms are
true in the model, the logical forms themselves are also true in the same model.
Consequently, they are consistent.
Now, consider the logical forms of a pair of particular sentences:

(∃xP x)

(∃x(¬P x))
Though it may not look like it, these logical forms are consistent and a model
can prove it:
D : {a, b}; P a
CHAPTER 7. MODELS 123

This model indicates that each logical form has two substitution instances. For
“(∃xP x)”, the substitution instances are

Pa

Pb
and, for “(∃x(¬P x))”, they are
(¬P a)
(¬P b)
Though “P b” is false in the model, “P a” is true. So, “(∃xP x)” is true in the
model. Meanwhile, “(¬P a)” is false in the model while “(¬P b)” is true. So,
“∃x(¬P x)” is also true in the model. Since both logical forms are true in the
same model, the model proves that they are consistent.
Here is one more example that incorporates some of the logical connectives.
The logical forms
(∀x(P x → (¬Qx)))
(¬(∃x(P x ∧ Qx)))
are consistent and this model proves it:

D : {a}; P a

According to this model, each of the logical forms above has one substitution
instance. For “(∀x(P x → (¬Qx)))”, it is

(P a → (¬Qa))

and, for “(¬(∃x(P x ∧ Qx)))”, it is

(P a ∧ Qa)

According to the model above, “(P a → (¬Qa)” is true because its antecedent,
“P a”, is in the model and “Qa” is not, which makes its consequent “(¬Qa)”
true. “(P a ∧ Qa)”, on the other hand, is false in the model because “Qa” is
not in the model. However, the fact that “(P a ∧ Qa)” is false in the model
means that “(¬(∃x(P x ∧ Qx)))” is true because this logical form says that
none of its substitution instances are true, which is in fact the case. Thus,
“(¬(∃x(P x ∧ Qx)))” and “(∀x(P x → (¬Qx)))” are true in the same model,
which proves that they are consistent.
While models can be used to prove that the logical forms of two or more
consistent sentences are, in fact, consistent, models cannot prove that the logical
forms of two or more sentences are inconsistent, that they cannot be true at
the same time. The reason is because a model cannot prove that two or more
sentences are never true at the same time. If I build a model in which one
logical form of a pair is true and the other logical form is false, then at best the
model proves that one could be false at the same time the other is true. But the
CHAPTER 7. MODELS 124

model does not prove that one must always be false and the other must always
be true.
Take, for example, the logical forms

(∀xP x)

(¬(∃xP x))
These logical forms are inconsistent; they can never be true in the same model.
Here is a model that makes one of them true and the other false:

D : {a}; P a

In this model, each logical form has one substitution instance and it is the same
for both:
Pa
“P a” is true in the model. Consequently, “(∀xP x)” is true in the model while
“(¬(∃xP x))” is false. Does this model prove that it is possible for these logical
forms to have contradictory truth-values? Yes, it evidently does. The question,
however, is whether the model proves that the logical forms must always have
contradictory truth-values. The answer is “no”; The model doesn’t rule out the
possibility that they could both be true in some other model.

Logical Validity
Having learned how models can be used to prove that two or more consistent
sentences really are consistent, you are ready to learn how to use models to
prove that an argument is invalid. In chapter 1, I explained that an argument
consists of more than one sentence. One of these sentences is the conclusion
while the rest are premises that provide justification for that conclusion. In
chapter 2, I explained that arguments can be valid or invalid. An argument is
valid if and only if the truth of its premises are inconsistent with the falsehood
of its conclusion. If the falsehood of the conclusion is consistent with the truth
of the premises, then the argument is invalid.
By now, I hope the point of validity is clear. Validity is a measure of whether
the premises of an argument do their job of justifying the conclusion. In logically
valid arguments, the premises do their job when they are true; in logically invalid
arguments, they don’t. Even when the premises are true in an invalid argument,
they don’t justify the conclusion.
Since arguments consists of sentences, their premises can, by definition, con-
sist of simple sentences or compound sentences. In the case of an invalid ar-
gument with simple sentences as either its premises or conclusion, a model can
demonstrate that it is invalid. Consider the logical form of the argument:

Premise: (∃xP x)
Conclusion: P a
CHAPTER 7. MODELS 125

This argument is invalid because it is possible for the conclusion to be false at

the same time the premise is true. This model proves it:

D : {a, b}; P b

In this model, the premise “(∃xP x)” has two substitution instances:

Pa

Pb
and one of these substitution instances is true: “P b”. On the other hand, the
conclusion: “P a” is false in the same model. Thus, the conclusion is false in
the same model where the premise is true, which indicates that the argument
is invalid.
Here is another example:

Premise: (¬(∀xP x))

Conclusion: (∃xP x)

This example is also invalid and this model can prove it:

D : {a}; Qa

According to this model, the premise and the conclusion both have one and the
same substitution instance:
Pa
This substitution instance is false in the model, which makes the conclusion
false. However, it also makes the premise true because the logical form of the
premise means that not all of its substitution instances are true. In fact, none of
them are. So, the premise is true. Since the premise is true and the conclusion
is false in the same model, the model proves that the argument is invalid.
Here is one last more complicated example:

Premise: (∃xP x)
Premise: (∃xQx)
Conclusion: ∃x(¬(P x ∨ Qx)))

This example is invalid because the truth of the premises are consistent with
the conclusion being false. Here is a model that demonstrates this fact about
the argument:
D : {a}; P a ∧ Qa
Based on the domain of this model, each of the premises and the conclusion of
this argument have one substitution instance. The first premise’s substitution
instance is
Pa
CHAPTER 7. MODELS 126

The second premise’s substitution instance is

Qa

and the conclusion’s is

(P a ∧ Qa)
. All of these substitution instances are in the model. So, all of them are
true. This makes the premises true in the model since they require only one
of their substitution instances to be true in the model to be true themselves.
The conclusion, however, is a different story. While its substitution instance is
true, the conclusion itself is false because its logical form indicates that one of
its substitution instances is false. Yet, its one substitution instance is false. So,
it is not true in the model. What this means is that the conclusion is false in
the same model that its premises are true, which proves that the argument is
invalid.
I mentioned at the beginning of this section that, even though you can
use models to prove that an invalid argument is in fact invalid, you cannot
use models to prove that a valid argument is valid. The reason is somewhat
complicated. According to the definition of logical validity given in chapter 1,
an argument is logically valid if and only if it is inconsistent for its conclusion
to be false and for its premises to be true. In other words, the conclusion of
a valid argument cannot be false when all of its premises are true. In terms
of a model, this means that there is no model in which the premises of a valid
argument are true and the conclusion is false. There is, in short, literally no way
for anyone—you or me—to come up with a model that makes the conclusion of
a valid argument false and its premises true. It just cannot be done.
The fact that it is impossible to build a model that proves a valid argument
is valid creates a dilemma. Suppose that you find yourself dealing with a valid
argument but which you don’t know it is valid. You might be tempted to see
whether it is invalid by constructing a model. However, if the argument really
is valid, then there will be no way for you to build a model that proves it to be
invalid simply because it is impossible to do so. Not knowing that the argument
is valid, you might keep trying to build a model to prove that it is invalid but
you will never be able to do it. When you finally give up, can you concluded
that the argument must be valid? No, you cannot because it may be that the
argument is valid but it also may be that you simply couldn’t come up with the
counterexample. For example, the argument:

Premise: (∀x(P x → Qx))

Premise: (∀x(Qx → Rx))
Conclusion: (∀x(P x → Rx))

is valid; The truth of its premises are inconsistent with its conclusion being false.
A model could prove an argument like this one invalid by making its premises
true and its conclusion false if the argument was invalid. Since the argument
CHAPTER 7. MODELS 127

is valid, the best that any model could do for it is prove that its premises and
conclusion are consistent by making them true. In the model

D : {a}; P a ∧ Qa ∧ Ra

both premises and the conclusion are true, but it doesn’t prove that the argu-
ment is valid because it does not prove that it is inconsistent for the premises
to be true and the conclusion false. Even the premises and conclusion of an
invalid argument are consistent. Take this argument, for instance:

Premise: (∀x(P x → Qx))

Premise: (∀x(Rx → Qx))
Conclusion: (∀x(P x → Rx))

The premises and conclusion of this argument are consistent. The same model
I built above makes the premises and the conclusion of the argument true.
Nevertheless, the argument is invalid. Here is a model in which the conclusion
of the argument is false and its premise are true:

D : {a}; P a ∧ Qa

The point is that building a model to demonstrate how the premises and the
conclusion of an argument can be true at the same time does not prove that the
argument is valid because it is also possible to do so if the argument is invalid.
In the end, a model can only prove that an invalid argument is invalid; it cannot
prove that a valid argument is valid.

7.7 Summary
Logicians build models of the facts which makes a sentence true. A model
defines a domain of discourse and the properties of members of the domain
and the relations between them. A domain of discourse consists of a set of
objects which are represented by individual constants. Without a model, there
is no way to determine whether a simple sentence is true or false. Whether a
simple sentence is true or false depends on whether it is singular, universal, or
particular. A singular sentence is true in a model if and only if the referent of
its subject is a member of the group identified by the predicate. A universal
sentence is true in a model if and only if all of its substitution instances are
true. A particular sentence is true in a model if and only if at least one of its
substitution instances is true. A substitution instance of a universal or particular
sentence is a singular sentence which substitutes every variable bound by the
quantifier with an individual constant from the domain of discourse in the model.
The truth-value of a compound sentence is a function of the truth-values of its
component sentences.
Appendix A

Common Indicator Words

In English we use certain words to indicate which sentences in an argument are

its premises and which is its conclusion. Here are some of the most common
indicators words:

Premise Indicators Conclusion Indicators

“Since . . . ” “Therefore . . . ”

“Thus . . . ”
“As indicated by . . . ”
“Consequently . . . ”
“Because . . . ”
“We may infer . . . ”
“For . . . ”
“Accordingly . . . ”
“In that . . . ” “We may conclude . . . ”
“May be inferred from . . . ” “It must be that . . . ”

“As . . . ” “For this reason . . . ”

“So . . . ”
“Given that . . . ”
“Entails that . . . ”
“Seeing that . . . ”
“Hence . . . ”
“For the reason that . . . ”
“It follows that . . . ”
“In as much as . . . ” “Implies . . . ”
“Owing to . . . ” “As a result . . . ”

128
Appendix B

The Truth-Conditions of
the Logical Operators

“X” and “Y ” are any simple or compound sentence.

1. “(¬X)” is true in a model if and only if “X” is false in the same model.
2. “(X ∧ Y )” is true in a model if and only if “X” is true and “Y” is true in
the same model.

3. “(X ∨ Y )” is true in a model if and only if “X” is true or “Y” is true, or

both in the same model.
4. “(X → Y )” is true in a model if and only if “X” is false or “Y” is true, or
both in the same model.

5. “(X ↔ Y )” is true in a model if and only if “X” and “Y” have the same
truth-value—whether true or false—in the same model.
6. “∀xXx” is true in a model if and only if all of its substitution instances
are true in the same model.

7. “∃xXx” is true in a model if and only if at least one of its substitution

instances are true in the same model

129
Appendix C

Basic Truth-Tables

Tail Arrow
X (¬X)
1 T F
2 F T X Y (X → Y)
1 T T T
Carrot 2 T F F
X Y (X ∧ Y) 3 F T T
1 T T T 4 F F T
2 T F F
3 F T F
4 F F F Double Arrow
Wedge
X Y (X ∨ Y) X Y (X ↔ Y)
1 T T T 1 T T T
2 T F T 2 T F F
3 F T T 3 F T F
4 F F F 4 F F T

130
Appendix D

Building Truth-Tables

Here is a simple method of constructing a truth-table for any compound sen-

tence:
1. Identify “n”, the number of simple sentences in a given compound sen-
tence.
2. Solve for “R” in the equation “R = 2n ” using “n”.
3. Draw rows equal to R under the original compound sentence.
4. Divide R in half, and enter “T” for “True” on the half of R rows un-
derneath the first simple sentence and “F” for “False” on the remaining
half.
5. Divide R in half again, entering “T” on the number of rows according to
the new quotient underneath the second simple sentence and “F” on the
remaining rows.
6. Repeat until truth-values are entered for every simple sentence, duplicat-
ing columns for identical simple sentences.
7. Compute the truth-values of the connectives, starting with one having the
narrowest scope and proceeding to the one with the widest scope.
8. The truth-values in the column under the main operatpr (the connective
with the widest scope) indicates the truth-values of the complex sentence.
Write the simple sentences to the left of the given compound sentence, separat-
ing them by a line. Underneath the simple sentences, enter their truth-values
following the previous procedure. Enter the truth-values for each simple sen-
tence in the compound sentence according the values given in the separated
columns. Compute the truth-values of the operators, starting with the one with
the narrowest scope and proceeding to the one with the widest scope. The
truth-values in the column of the main operator indicates the truth-values of
the compound sentence.

131
Appendix E

Glossary

Ambiguity: An expression is ambiguous if it has more than one discrete mean-

ing.

Argument: An argument is a set of sentences, one or more of which gives us

reason to believe one and only one of the others.
Assumption: In the context of an argument, a general belief which a person
relies on to make their inferences logically valid.

Compound Sentences: A compound sentence is a sentence that consists of

simpler sentences and at least one connective.
Connective: A connective combines simple or compound sentences into com-
pound sentences according to certain grammatical rules called formation
rules.

Counterexample: An example that describes a situation in which a sentence

is false.
Counterexample to an Argument: An example which describes a circum-
stances in which the premises of an argument are true and the conclusion
of that argument is false.
Formula: The expression that remains when a quantifier is removed from a
logical form.
Fallacy: An invalid inference that people frequently confuse with a valid one.

Logical Consistency: Two or more sentences are logically consistent if and

only if all of them can be true at the same time.
Inference: An inference is the mental act of moving from the premises of the
argument to a conclusion.

132
APPENDIX E. GLOSSARY 133

Interpretation: A list of simple sentences and the sentence constants that

represent them.
Logical Equivalence: Two sentences are logically equivalent if and only if
they have the same truth value in all cases.

Logical Form: The logical form of a sentence is an arrangement of sentence

variables and truth-functional connectives such that the uniform substi-
tution of simple or compound sentence in place of the variables results in
a larger compound sentence.
Tautology: A compound sentence is tautology if it is true regardless of the
truth-values of its component sentences.
Logical Validity: An inference is logically valid if and only if the falsehood of
its conclusion is logically inconsistent with truth of all of its premises.
Necessary Condition: A conditional sentence is a sufficient condition for an-
other statement if and only if the truth of the antecedent is the only
requirement needed to entail the truth of the consequent.
Predicate: The part of a sentence that excludes the subject.
Premise: A premise is a sentence that gives us a reason to believe a conclusion,
which is another sentence.

Principle of Interpretive Charity: In the context of evaluating arguments,

a person ought to interpret another person’s argument as logically valid
rather than logically invalid if it is possible to do so.
Sentence: A sentence is any expression which has a truth-value, that is, it can
be true or false.

Simple Sentence: A simple sentence is a sentence that does not consist of

other sentences as its components.
Soundness: A argument is sound if and only if (a) all of its premises are true,
and (b) the inference from its premises to its conclusion is valid.

Straw Man Fallacy: The logical mistake of deliberately misrepresenting an-

other person’s point of view or opinion in such a way that it will be easily
refuted or rejected.
Sufficient Condition: A conditional sentence is a necessary condition for an-
other statement if and only if consequent statement cannot be true without
the truth of the antecedent.
Truth-Functionality: A compound sentence is truth functional if and only if
its truth-value is determined exclusively by the truth-values of its compo-
nents.
APPENDIX E. GLOSSARY 134

Truth-Table: A truth table is an arrangement of truth-values that defines for

every possible situation how the truth-value of a compound sentence is
determined by the truth-values of its components.
Index

Ambiguity, 32, 47 Mention, 7

Argument, 12
Necessary Condition, 76
Compositionality
The Theory of, 58 Philosophy, iii
Conclusion, 12 Predicate, 48
Connective Premise, 12
Definition, 38 Principle of Charity, 31
Consistency, 9
Counterexample, 8 Scope, 47
Arguments, 29 Sentence
Complete, 4
Fallacy Compound, 37
Straw Man, 30 Simple, 37
Formula, 102 Soundness, 24
Implicit Premise, 31 Substitution Instance, 114
Interpretation, 40 Sufficient Condition, 75
Symbolic logic, iii
Logic, iii
Logical Contradiction, 68, 70 Truth
Logical Equivalence, 69 The Correspondence Theory, 5
Logical Form, 51 Truth Value, 4
Logical Validity, iii, 70 Truth-Functionality, 58
Truth-Table, 61
Main Operator
Definition, 48 Validity, 25

135