Part I / Arguments

CHAPTER 1 I

The Quarrel

IN THIS chapter we introduce the important distinction between arguments in the

broad sense and arguments in the narrow sense. We then turn our attention to an
interesting kind of argument in the broad sense, namely the quarrel, and bring the
chapter to a close with a discussion of a pair of fallacies to which quarrelsome argu­
ments are especiaUy prone.

1. Arguments, Fallacies, and Logic

One primary goal of logic is to evaluate arguments by dividing them into various
groups: the good and the bad, the reliable and the unreliable, the sound and the
unsound. As we will see, there are many different ways of making these evaluations.
As a result, a preliminary goal will be to identify various standards of good argu­
ment and to determine how best to meet them. Arguments can be noisy and stress­
ful affairs, but they aJso can be enjoyable events in which participants exchange
evidence and ideas in the hope of reaching a consensus on some (often controversial)
topic. Sometimes we keep our reasoning to ourselves. At other times we make a pub­
lic display of it. Either way, a common way of giving expression to our thinking is by
constructing and exchanging arguments.
Many different things are arguments, including quarrels, legislative debates.
diplomatic wrangles, labour negotiations, mathematical proofs, legal briefs, scien­
tific demonstrations, and so on. However, there is an important distinction that helps
us classify these different types of argument in a useful way. This is the distinction
between arguments in the broad sense and arguments in th.e narrow sense. Arguments
in the broad sense (or, as we may also say, in the social sense) are public, multi-person
exchanges of comment and counter-comment. Arguments in this sense illustrate the
principle "It takes two to tango," except here we might better have said "tangle"! It is
easy to see that quarrels, debates, and the other types of argument listed above are all
arguments in this broad sense.
Here are some typical characteristics of arguments in the broad sense. Arguments
in the broad sense

■ involve two or more arguers; it is this feature that allows us to describe

arguments in the broad sense as also being arguments "in the social sense."

■ occur in real time; normally one arguer says something, then his or her
opponent has something else to say, then the first party responds, and so on.
2 PART 1· Arguments

Arguments in the Broad Sense

As log,oans use the term, an argument is a presentation of reasons or evidence in
support of some claim. It is an attempt to build a case in favour of a conclusion.
Normally this means that arguments are social exchanges involving a series of speech
acts uttered by two or more parties. Familiar examples include quarrels, legislative
debates, diplomatic wrangles, labour negotiations, mathematical proofs, legal briefs,
scientific demonstrations. and so on. Because they involve a variety of social and
psychological, as well as logical, factors, we say that these are arguments in the
broad (or socia0 sense.

■ are governed by a set of rules or standards, although in many contexts these

may be given only implicitly; thus not every verbal exchange between two
parties in real time is an argument; sometimes, for example, it may just be an
exchange of information.
■ involve the offering of evidence, or premisses, 1 in favour of some claim or
conclusion.
■ involve the drawing or the denying of that conclusion.

In other words, arguments in the broad sense are social exchanges between two
or more parties in which premisses are offered in favour of a conclusion according t
o
a given set of rules or standards. These rules or standards will vary from context to con­
text, but typically determine whether, in a given situation, an arguer has justified (or
proved) his or her conclusion.
Let us now say something about arguments in the narrow sense. Arguments in the
narrow sense are simply sequences of propositions, one of which is the argument's
conclusion, the rest of which are the argument's premisses. This is a somewhat
technical notion of argument, abstracted from arguments in the broad sense. Of the
five distinctive features of arguments in the broad sense (na mely agency, time,
standards, premisses and conclusion) only the last three are considered when analyzing
arguments in the narrow sense. In other words, the narrow sense of argument filters
out agency and time, concentrating solely on how an argume nt's premiss es aod
conclusion are related.
Thoug h different from arguments in the broad sense, arguments in the narrow
ery
sense are intimately connected to arguments in the broad sense. At the core of ev
an
argument in the broad sense is an argument in the narrow sense. Thus there is
important relation between the two. If the narrow (or core) argument is defective, so
Even
too is the more broad (or social) argument. However, the converse need not hold.
ader
if a core argument is conclusive, there may still be respects in which the bro
argument fails. For example, it may simply miss the point at hand, or it may in sorne
other way fail to take account of various social factors.
Logic studies both the structure and the adequacy of argum ents. Corresp0n�S
. . between arguments in the broad and narrow sense is a d'15tincoon
to t he d.1stmct1on
· · d es 1og1c.
t hat d 1v1 · th e b road sense is that par t of logi c that i· nvestigates
· Log,c· m
 CHAPTER 1: The Quarrel 3

�r�ents in the broad sense, and logic in the narrow sense is that part of logic that
mvestigates arguments in the narrow sense. It should be clear by now that logic in the
broad sense is a more complex and nuanced enquiry than logic in the narrow sense,
but it is also true that logic in the narrow sense aUows for more precision and exac­
titude than does its more complex counterpart.
Logic in both the broad and the narrow sense is an element of critical thinking.
The chief value of logic in the broad sense is that it takes account of arguments as they
occur in real life between real-life participants. But since every argument in the broad
sense contains at its core an argument in the narrow sense, it is also possible to be very
precise about certain features of arguments in the broad sense. Not all of critical
th.inking is argumentative. One can read a newspaper description of an event critically
without there being the slightest element of argument in the reporter's account. Even
so, the argumentative component of critical thinking is very large, and some of the
skills that we develop in assessing arguments critically can also be used to assess non­
argumentative discourse. For example, if we have a technique for checking an oppo­
nent's premisses for consistency, we can use this same technique to check for
consistency in other contexts as well.
In order to make the distinction between premisses and conclusion dear, we
sometimes place an argument in standard form. To do this, we explicitly list the pre­
misses and conclusion, eliminating any unnecessary or redundant words and clari­
fying any ambiguities. We then separate the conclusion from the rest of the argument
by means of a horizontal line.
For example, consider the argument
Archimedes must be either a hero or a martyr. After all, anyone who dies in
battle is one or the other and, as we know, Archimedes perished during the
capture of Syracuse.

We can place this argument in standard form as foUows:

Anyone who dies in battle is either a hero or a martyr
Archimedes perished during the capture of Syracuse

Therefore, Archimedes must be either a hero or a martyr.

Once we have identified the premisses and the conclusion in this way, we can consider
a third and essential aspect of the argument, namely the evidential link that relates the
premisses to the conclusion. This link, usually referred to as the argument's consequence
relation, determines the logical strength of the argument. To say that this relation is
strong is to say that the premisses, if true, provide good evidence in favour of the con­
clusion. To say that this relation is weak is to say that the premisses, if true, provide
poor evidence in favour of the conclusion. A consequence relation that is neither
strong nor weak is said to be moderate.
This is a good place to introduce another important distinction. It is common
practice among logicians to draw a contrast between deductive and inductive argu­
ments. As generally understood, a deductive argument is one put forward by the
arguer with the intent and expectation that it will be judged according to deductive
4 PART I Arguments

standards. Similarly, an inductive argument is one that is advanced with the inte
nt and
expectation that it will be judged by inductive standards. What are deduc ti·
ve and

in�;
inductive standards? De tailed answers can be found in Part II of this book b
rough idea of what is meant can be given here . We will say that an argument' a
the standard of deductive ndeq11ncy just when it is impossible for its conclusion
t:
be false, assumi11g that its premisses are all true. When an argument meets this st
an-
dard, its conclusion can be said to be a deductive consequence of its premisses. In
other words, an argument is deductively adequate when its consequence relation is as
strong as a consequence relation can possibly get. On the other hand, when an argu.
ment meets the standard of ind11ctive adequacy, two conditions must hold. The first
is that it must not be ded11ctively adequate. The other is that its premisses, if true,
must significantly increase the likelihood of its conclusion also being true. In other
words, an inductively adequate argument is one whose consequence relation is not as
strong as in the case of deductive arguments, but it is nevertheless stron g enough.
Inductive adequacy is a less strict standard than deductive adequacy; but this is not
to say that inductively strong arguments are "second-class citizens" in the republic of
logic. Both science and everyday life are awash in powerful and important arguments
that happen to be only inductively strong.
We should also note in passing that, strictly speaking, the distinction between
deductive and inductive arguments applies only to arguments in the broad sense.
Deductive arguments are those intended by their participants to be judged deductively;
inductive arguments are meant to be judged inductively. But arguments in the nar­
row sense are defn i able without any regard at all to participants; hence there is no ques­

tion of what may or may not have been intended. Even so, although narrow arguments
are abstractions from what actual arguers say and intend, sometimes it is possible
to infer the appropriate standard for appraisal from the context in which we find a nar·
row argument. Here is an obvious example. Suppose you are working on a homework
exercise that contains two groups of arguments in the narrow sense. Your instructions
are to evaluate the first group deductively and the second group inductively. These
instructions provide a context that makes it possible for you to apply the required stan·
<lard of adequacy to each group of arguments.
As already mentioned, in Part II we will study the deductive-inductive contraSt
t
in a good deal more detail. Until then, however, we shall make do with our pre sen
sketchy exposition of it.
Learning to reason critically about various issues means being able to create and
ey
evaluate arguments. It means being able to use propositions in such a way that tb
en
provide good evidence in support of a conclusion, and being able to evaluate w�
i.e .,
this goal has been achieved. In essence , the overall quality of an ar gume nt (
whether it is good or bad) depends primarily upon two factors: ( l) wh ethe� the
el
premisses are (likely to be ) true, and (2) whether the consequence relation is (lik Y
,
to be) strong. Bot11 are necessary. Unless they are true, or at least known to be PIausible
s
premisses offered in support of a conclusion will be of very little use. At the �;
elaoo
time, unless they are relevant and significant (that is, unless the conse q uence r . a
is stron g), even true premisses will fail to provide a suitable basis for accep tJ S
Jl
 CHAPTER l: The Quarrel S

Arguments in the Narrow Se

nse
At the core of every argum en
· the broad sense .Is a set of propositions composed
t 1n
o f the argument's prem tsse .
· s and cone1us,on. We call this an argument in the narrow
sense. Arguments in the narrow sense
are abstractions from arguments in the broad
sense.
T o ma�e clear t e structure of an argument in
� the narrow sense, and to distinguish
b etw een
its premisses and conclusion, we often place the argum
ent in standard
form. To do this, we identify its premisses and conclusion and
separate the conclusion
from the rest of the argument by means of a horizontal
fine, as follows:

1st premiss
2nd premiss

nth premiss

Conclusion.

The premisses form the basis of the argument. They constitute the evidence that is
being offered in support of the conclusion.

conclusion. It follows that there are two main ways to criticize an argument: we can
show the premisses to be unacceptable, or we can show the consequence relation to
be weak. Other factors may aJso be important, but these two are absolutely central.
Since at the core of every argument lies a series of propositions, it is worth taking
a moment to understand how logicians use this term. In classical logic, a proposition
is something that is either true or false. For any given proposition we may or may

Reasoning Critically
Reasoning critically about various issues requires that we learn to evaluate arguments.
The quality of an argument depends primarily upon (1) the likelihood of the truth
of the argument's premisses, together with (2) the strength of the consequence
n. It follows that
relation that holds between the argument's premisses and conclusio
the premisses to
there are two main ways to criticize an argument: (1) we can show
nce relation to be weak. Other
be unacceptable, or (2) we can show the conseque
factors may also be important, but these two are absolutely central.
To say that a consequence relation is strong is to say that the premisses, if true, pro­
that this relation is weak is to
vide good evidence in favour of the conclusion. To say
say that the premisses, if true, provide poor evidence in favour of the conclusion. A
consequence relation that is neither strong nor weak is said to be moderate.
It is a primary goal of logic to help determine the quality of arguments. Logic in
the narrow sense does this by studying the nature of the consequence relation that
holds between the premisses and conclusions of arguments in the narrow sense. In
contrast, logic in the broad sense studies arguments in the broad sense. It studies
those features, in addition to the consequence relation, that are relevant to an
argument's success or failure, including an argument's purpose and social context.
6 PART I. Arguments

not know which it is but, if it is a proposition, it must be one or the other. Questions
and commands, for example, are not propositions, even though they typically contain
or make use of propositions. A proposition is that which is asserted by a declarative
sentence. Although "Snow is white" and "La neige est blanche" are two distinct
sentences, they (norma!Jy) assert exactly the same thing. Thus we say that they express
one and the same proposition.

Identifying Premisses and Conclusions

In an argument, the premisses and conclusion are of ten indicated by the use of key
words or phrases. For example, words and phrases such as "thus," "therefore,"
"consequently," "so," "it follows that," and "we may conclude that" typically indicate
which proposition is functioning as the conclusion. Similarly, key words or phrases such
as "since," "because," "given that," and "on the grounds that" regularly indicate
which propositions are functioning as premisses.

It is not always clear which proposition in an argument is meant to be the con­

dusion and which propositions are meant to be the premisses. Fallacies can occur when
conclusions and premisses are confused with one another. A fallacy is any argument
that deceives us because it appears to be a reliable argu­
Fallacies ment but, in fact, is not. The unwary, who may not be
A fallacy is a bad argument or a piece of
bad reasoning that has a tendency to
acquainted with how the fallacies work, are especially
appear good. It is a common or vulnerable.
seductive error in reasoning. The notion of fallacy was first recognized in the
West almost two and a half thousand years ago by
Aristotle. Seen the Greek way, fallacies are misuses of reason or rationality (or what
the Greeks called logos, the methods of rational inquiry that lead to good reason­
ing). The first fruits of logos were developments in physics and astrophysics in the
sixth century BCE, but before long these same methods were being applied to other
areas as weil-for example, to issues in ethics and politics, and to arguments about
the human good. Thus we can see in these early inquiries the stirrings of what would
eventually become the humanities, the natural and social sciences, and the main sub­
disciplines of Western philosophy: metaphysics (the theory of being), epistemology
(the theory of knowledge),value theory (the study of ethics, politics, and aesthetics),
and logic. Of these subdisciplines, it was logic that set itself the task of identifying and
analyzing the fallacies and of distinguishing good arguments from bad.
At the heart of Aristotle's logic was his theory of the syllogism. For Aristotle, a
at
syllo gism was a special kind of cogent argument2 and a faUacy was an argument th
sy
appears to be a sy!Jogism but is not. In present-day usage it is more common to �
f its
that a fallacy is any argument whose conclusion appears to be a consequ ence o
premisses, but is not.
Identifying fallacies re quires that we become clear about the distinction between
inter·
inference and argument. As we have said, arguments in the broad sense are
personal, social phe nomena-a kind of public even t. Inferences, in con trast , are
tt
essentially private affairs. They involve the adjusting of an individual's mental s a es,
 CHAPTER 1; The Quarrel 7

Flash back on... /

Aristotle (384-322 BCE)
ARISTOTLE is generally regarded as the founder he left well endowed by Alexander; others say
of formal logic, or of what he called analytics. that he was lucky to escape with his life. In any
For more than two thousand years he was logic's event, his influence could not have been too
most influential writer. great since Alexander soon ordered Aristotle's
Aristotle was born in Stagira, in northern nephew, Ca/listhenes, hanged for ref using to
Greece. His father had been court physician to bow before the new king.
the Macedonian king, Amyntas II, but both of When Aristotle finally returned to Athens in
Aristotle's parents died when he was still a boy. 335 BCE, he op ened his own school, the
In 367 BCE, Aristotle was sent to Athens to Lyceum. In 323 BCE, following the death of
study under the famous philosopher Plato. He Alexander, anti-Macedonian feeling in Athens
remained at Plato's school, the Academy, first increased and, because of Aristotle's ties to
as a student and later as an instructor, for almost Alexander's court, an accusation of impiety was
twenty years. Initially, the students at the raised. Recalling the fate of Socrates, who had
Academy made fun of this new foreigner who been put to death after being found guilty of
spoke with a lisp, but Plato himself was similar charges, Aristotle returned to Chalcis,
impressed with Aristotle and nicknamed him his mother's hometown, saying that he did not
"the intelligence" of his school. Upon Plato's want Athens to "sin twice against philosophy."
death, Aristotle left Athens, perhaps for politi­ He died the following year.
cal reasons, or perhaps because he had not been Aristotle's logical writings, referred to as the
appointed Plato's successor and was dissatisfied Organon, consist of six books: the Categories,
with how the Academy was now being directed. De lnterpretatione (On Interpretation), the Prior
In 342 BCE Aristotle was appointed by Philip Analytics, the Posterior Analytics, the Topics,
II of Macedon to tutor his son, the future and De Sophisticis Elenchis (On Sophistical
Alexander the Great. From Aristotle's point of Refutations).
view, the appointment could not have been a Transmitted in large part by Arab scholars,
pleasant one. He was expected to provide guid­ Aristotle's writings shaped the intellectual devel­
ance to the young Alexander in a palace noted opment of medieval Europe. His logical trea­
for its savagery and debaucheries, and at a time tises occupied a central place within the
when both the father and son were intent upon medieval curriculum, at which time Aristotle
assassination and conquest. When Alexander came to be known as the source of all knowl­
finally claimed his father's throne in 336 BCE, edge.
Aristotle left the court. Some reports say that

or the transition from one belief state to another. Thus, they are something the indi­
viduaJ handles solo. Even so, there are similarities. We may say that each public act of
argument serves as an invitation to make an inference. As a result, both arguments and
inferences may be modelled by what we have called "the core" of an argument, by
arguments in the narrow sense of the term.
8 PART I Arguments

When two people-say, Sue and Bill-argue, Sue is bidding Bill to update or
revise what he believes in light of certain considerations. In other words, through
argument she is trying to get Dill to make an inference. Guiding this transaction
between Sue and Bill is the following general principle: If Sue's argument is good,
Bill's corresponding inference, if he draws it, will also be good. By the same token, if
Sue's argument is good, then the inference it bids Bill to draw is one that Bill ought to
draw. It will be the rational thing for Dill to do.
It is for this reason that arguments and inferences may be said to share a common
core. This core is what remains when we abstract an argument or inference from its
particular context, suppressing the fact that it was Sue rather than Sandra who
presented the argument, and that the argument was directed to Bill rather than to Bob.
Also suppressed are those institutional, social, cultural, and other background
conditions that give shape and emphasis to real-life social exchanges. In cases in
which we believe that these factors are important, we elect to study arguments in
the broad sense; in cases in which we believe that they are unimportant, we elect to
study arguments in the narrow sense.

Exercise 1.1

Note: A "*" indicates that the solution to this exercise appears at the back of the book.

1 Determine whether each of the following 2. Place each of the following arguments in
sentences expresses a proposition: standard form by first writing out the
*(a) Athens is the capital of Greece. premisses and then separating them from
the conclusion by a horizontal line:
*(b) Athens was the capital of Greece.
*(a) Selling illicit drugs is not only illegal, it
*(c) Athens will be the capital of Greece. is also immoral. No one who brings that
degree of misery to any one can be said to
*(d) What is the capital of Greece?
be acting in a morally neutral fashio n.
*(e) I believe that Athens is the capital of
*(b) Although Socrates was thought by
Greece.
many to be a Sophist, in fact he was not.
(f) Either Athens is the capital of Greece or He did not accept money for his teaching.
it isn't. he did not travel with the Sophists. and he
(g) I wish I knew whether Athens is the did not share their goals or beliefs.
capital of Greece! *(c) Pleasure is the only ultimate good. It is
(h) Tell me whether Athens is the capital of the only thing we seek for its own sake.
Greece. After all, we try to obtain money, health.
and education only to help us achieve
(i) It is possible that Athens is the capital of happiness . All these things are instru mental
Greece. in a way that pleasure or happiness is not.
al of Greece, then
(j) If Athens is the capit *(d} It is often claimed that pleasure 15. th
e
Athens must be in Greece. only ultimate good; that pleasure is the
 CHAPTER 1: The Quarrel 9

only thing we seek for its own sak

e. But if (h) Any government that caves in to
this is right, why do some people
give up terrorists must be suffering from a lack of
pleasure in return for achieving som
e other willpower. We know that our government
end? Why would we ever work hard
for has already implemented a policy of never
our children? Why would we ever put
giving in to terrorists. It follows that our
ourselves in danger for another human
government does not suffer from a lack of
being? willpower.
*(e) The only things that we can ever (i) Assume for the sake of argument that
perceive-even in principle-are a = b. Jf we multiply both sides by a, we
sensations, ideas, memories, and get a2 = ab. Then, subtracting b2 from both
impressions. In other words, we can never sides we get a 2 - b2 = ab - b2 . Then,
perceive-even in principle-anything factoring both sides, we get
outside our own minds. So it follows that (a + b)(a - b) = b(a - b). But this equation
we can never have any evidence for the has a common term: (a - b). It follows that
existence of anything but the mental. by cancelling this term we get (a + b) = b.
Recalling our original assumption, we see
(f) We cannot learn what we do not know.
by substitution that (b + b) = b. So, in other
In order to learn, we need to be able to
words, 2b = 1 b, or 2 = 1 !
distinguish truth from falsehood and, if we
do not already know something to be the 0) The freedom of all people to speak
case, we will not be able to recognize it as openly, without fear of government
being true. It follows that all learning is reprisal. is what allows us to raise issues
impossible and any knowledge that we will that we as citizens think are important. So
have in the future must already be known free speech, together with freedom of
to us in the past. association and peaceful assembly and the
freedom to exercise our franchise, is what
(g) Even though we say we live in a allows us, as citizens, to exercise our
democracy, a universal franchise is not in sovereignty over government. In other
effect. Prisoners cannot vote. The mentally words, free speech is one of the
ill cannot vote. Non-resident citizens and cornerstones of contemporary democracy.
citizens travelling overseas often cannot
vote. Until these things change, the
franchise is a long way from becoming

--------·-------------
universal.

2. The Quarrel
One common type of argument (in the broad sense) is the q11arrcl. We begin by
discussing two specific examples. Here is the first:

The "Yes, You Did:' "No, I Didn't,, Quarrel

Sue insists that Bill told her he would spend last Friday in the library. Bill, who was
spotted at the beach, contends that he told Sue he planned to relax there all day. The
following conversation results:
10 PART I: Arguments

Sue: 1 thought you told me that you were going to be at the library last Friday.
Bill: No. Don't you remember? I said I would be at the beach.
Sue: No. You said you would be at the library. Otherwise I would have wanted
to go with you.

Bill: You're not remembering things correctly-that was the day I said I would
be at the beach I
Here, the cLisagreement is about wlint the facts are. As anyone who has ever had such
a disagreement knows, they can be abusive and inconclusive. Sometimes such argu­
ments are great fun; more often they are a nuisance, and an impediment to reasoning.

The "You Don't Love Me Anymore" Quarrel

Our second example of a quarrel is quite different. Bill is racked by doubts about
Sue's behaviour. When they were first in love, Sue seemed to hang on Bill's every
word. She laughed at his jokes, even those that made his other friends groan. She
reserved all her free time for him. Recently, however, she has been less interested in
what he has to say, and last week she even fell asleep while Bill was regaling their
friends with a tale about his adventures as a moose hunter. In addition, she remarked
that his joke about the farmer's daughter and the astrophysicist was "mindless and sex­
ist." When Bill discovers that Sue has gone to a concert with John, conversations such
as the following become typical:
Bill: You don't love me.
Sue: But I do! How could I call you a "mindless bore" without caring for you
enough to risk giving offence and perhaps even losing you?

Bill: Sure, sure. How do I know that this isn't just what you have in mind­
losing me, as you call it, or breaking up, as I would say?

Sue: Bill, you just don't understand. Be reasonable.

As we can see, in this case Sue and Bill are arguing not so muc h about what the
facts are, but about what the facts mean and how they are to be interpreted. Here, the
disagreement is about what the facts imply.
Using the terminology of the previous section, we may say that the part of an
argument dealing with what the facts are constitutes the premisses of the argument;
the part dealing with what the facts imply constitutes the conclusion of the argu·
ment. In other words, in the first argument Sue and Bill disagree about their pre­
misses; in the second, they disagree about their conclusion. In the first argument we
find premissory instability, that is, we find that Bill and Sue agree about very
few, if aJJY•
of the relevant facts of the case. They have few or no premisses in common. In �e sec·
ond argument we find conclusional instability;' that is > we find that although Bill and
Sue agree about (most of) the salient facts they cann ot agree a bout what these factS
,
imply; thus they cannot agree about what should be concluded from the r o °11
i c tn.01
pre mis ses.
 CHAPTER 1: The Quarrel 11

It is not hard to see why prem1ss · ory an d coneI us,on · ·

. . · al rnstab1l1ty · lead to drffi-
cult1es m the evaluation of an · ·
• arg umen t . If there rs prem1ssory or conclusr·onal insta- ·
bility, then in fact we have two or more se arate argu
p ments that do not interact with
·
each other at all · lf we try to I oca t e the premiss · ·
es and the conclusion-th e logrcal
core of the argument-we cannot find them. Bill has his
premisses and conclusion,
Sue has her premisses and conclusion, and we can say nothing about
the single argu­
ment as a whole. Where there is no common ground, there is no single argum
ent.
Instead, we have a plurality of unconnected arguments. SmaJI wonder, then, that the
quarrel tends to be neither productive nor enlightening from a logical point of view.
It is now easy to see why quarrels are typicaUy such
noisy, inconclusive affairs. If there is premissory or con­ The Quarrel
clusional instability we cannot even get started on the An argument is a quarrel if (1) it suffers
road to agreement. Frustration, accusation, and hurt from premisso,y or conclusional
instability (or both), (2) the disputants
feelings are bound to follow. References will tend to
lack a shared method for conflict
become personal and disagreeable. Sue might eventually resolution, and (3) they ref use to agree
complain that if Bill cannot recall what he said last to disagree.
Friday, then he is a simpleton; Bill might retort that Sue
is a shrew and, before we know jt, things will have taken a nasty turn. Bill might
contend that Sue shows herself to be a "typical woman" in having no capacity to
reason. Sue may even threaten to slap Bill's face (and perhaps would be forgiven for
doing so).

What's So Special about Logic?

Logic is a principal part of the study of argument. It is often distinguished from
rhetoric, the study of persuasion. The difference is that, while rhetoric is interested in
discovering the class of arguments that are in fact persuasive, logic is interested in
discovering the class of arguments that ought to be persuasive, or that would be
persuasive to an ideally rational reasoner. Thus, in contrast to rhetoric, logic is often
said to be the study of reasoned (or rational) persuasion.
Jn this context rationality (or what the ancient Greeks called logos) is said to be the
capacity (or ability) to effect the systematic manipulation of one's beliefs in such a way
that they become proportional to the available evidence. In other words, we are
rational whenever we are able to change our beliefs on the basis of evidence. If we
are unable (or unwilling) to alter our beliefs, even though the evidence demands
otherwise, we are not acting rationally. Certain "propositional attitudes" (i.e., attitudes
that we have towards propositions, including fear, affection, apprehension, belief,
acceptance, denial, etc.) are more difficult to deal with rationally than others. For
example, a person who know s that a roller-coaster ride is safe, but who still fears
for his or her safety while riding it, may be said to suffer from an irrational (or perhaps
a non-rational) fear.
Why do we place such high importance on rationality and logic? There are several
reasons, among which are the following:
• First, beliefs based upon reason may be tested and improved in a way that other
kinds of beliefs cannot. This is because reasoned arguments are capable of being
►
12 PART t Arguments

discussed and evaluated openly according to well-defined sta ndards. Unlike "feel­
ings" or "intu1t1om" -which also motivate beliefs. but which are essentially pri­
vate-reasoned arguments provide us with a well-defined, public mechanism for
the resolution ol disagreement
• Second, reasoned arguments provide us w,th a safe, non-threatening way of
resolving disputes Unlike physical means of conflict resolution-which range
from personal int1midat1on to warfare-logic gives us a non-harmful mechanism
for the resolution of disagr eement.
• Third. careful, logical reasoning from acceptable premiss es to an acceptable con­
clusion 1s the method most likely to lead to accurate beliefs and. hence. to sensi­
ble decisions In turn, these beliefs and deos1ons are more likely than others to help
us live well and improve our lives

The logical lessons to be learned from the quarrel are thus mainly negative:
■ Because a quarrel has little or no chance of starting or concluding
satisfactorily, there is little chance that it wiU advance the course of
knowledge. Consequently, a quarrel is a systematically inefficient and
inconclusive kind of mg11ment.

■ Premissory and conclusional instability often lead to frustration, and

,frustration often leads to aggression. Aggression is a well-known trigger of
fallacious reasoning. Therefore, not only is n quarrel likely to be inconclusive; it
also risks giving rise to fallacious reasoning.

■ Because premissory and conclusional instability in an argument turn the

argument in Lo a quarrel, it is not reasonable to proceed with such arguments
until tl,eir instability is elimi11(1ted, red11ced, or q11ara11tined. 3 If one's goal is to
reason well, and if an argument's instability shows no promise of being
reduced, one should stop talking, agree to disagree, or simply change the
subject.

The log ical moral is that, whenever possible, quarrels are to be avoided.

Exercise 1.2

1 For each of the following arguments.

*(b} The secretary at the mayor's office is
identify the premisses and the conclusion not answering her phone. I guess she has
and place the argument in standard form: left for the day.
•(a) No compute r will ever think. To think,
*(c) None of Europe's current leaders a e
r

you need to be able to understand the real revolutionaries. After al l, every one of
meanings of words, and this is something them was born after the end of the sec0nd
no compute r, whatever its speed or World War.
sophistication, will ever be able to do.
 CHAPTER 1:TheQuarrel 13

*(d) Every right places a corresponding

bright, but until we have safely initiated a
obl'.gation upon someone. If I have a right
human colony on Mars. they don't deserve
to lite, then you have an obligation not to
the salaries they receive.
kill me. If you have a right to free passage,
then I have an obligation not to hinder (h) No woman who has been raped can
your progress. If we both have a right to be ever fully recover. After all, her memories
treated equally before the law, then society will stay with her forever.
has an obligation to see that this is so. (i) No soldier who has seen action can ever
*(e) International law is bound to be fully recover. After all. his memories will
ineffective. The reason is that no law can stay with him forever.
be effective without appropriate Q) All languages involve both syntax and
enforcement. But international law. despite semantics and this is something that is
its noble origins. can never be strictly characteristic of many complex systems. So
enforced. all languages are complex.
(f) If we knew what the good life was, we *2. For each of the arguments listed in
would naturally act in such a way as to question 1, determine whether the
achieve it. Thus. any unhappiness we have consequence relation is strong. moderate.
is a result of our ignorance. or weak.

(g) Software engineers are all overpaid.

They may work hard. and they may be

3. The Ad Baculum
In the remainder of th.is chapter we will discuss two important fallacies often associated
with the quarrel: the ad baculum and the ad horninem. The first of these is the fallacy
of inappropriately appealing to force or intimidation; the second is the fallacy of
inappropriately appealing to facts (or alleged facts) about one's opponent, rather
than addressing his or her argument directly. We emphasize that quarrels are only one
i
k ind of context for the commission of these fallac es. As a little reflection will show,
xts as well. We pick the quarrel
these are fallacies that crop up in other argument conte
simply becau se a quarrel is a very
natural context for the ad baculum The Ad Baculum Fallacy
An ad baculum fallacy occurs whenever
and ad hominem. a conclusion is drawn. or invited to be
In the scenario described a few drawn. on the basis of an irrelevant
paragraphs ago, we imagined Sue and appeal to intimidation or force. (When
Bill in the heat of battle, with Sue such an appeal is relevant, the argument
threatening to slap Bill's face. This is said to be a non-fallacious ad baculum
may not be a very impressive exam ple argument.)
of an ad baculum fallacy because it is
likely too obvious to fool anyone. However, even the threat of a slap on the face may
give Bill pause, making him have doubts about the wisdom of his position. At the
14 PART I Arguments

ations in which disputants resort to more subtle

same timt', it is easy 10 imagine: situ
s npleasant consequences that are not re
forms of innuendo and Lo �uggt·stion of u l­
evant to the bsuc al h.111d.
e because they release strong
lnapµropri,,te ::ippc,1b to force can often be effectiv
ls often work, not so much a5
emotions al the t·xpcn!>e of reason. Such appea
,irgumcnts, but as distractions from argumcnl. Aristotle shrewdly points out that
the strategy of stimulating anger is often a highly effective, even if undesirable, tac­
tic in argurncnt.'1 If your opponent is winning the argument because of the logic of
his position, and you arc .11 a loss to find sufficient evidence in supp o rt of your own
view, an all-100-effectivc !'llratcgy is lo derail the argument by unleashin g powerful
emotions. Aristotle obser ves that arguers arc often less able to defend themselves
when agitated. One.• wa)' lo produce anger is lo make it clear to your opponent that you
w ant to use unf.,ir 111<.'ans in the argument; to indicate that you might even resort to
violence or forct'. This is an ex.11nplc o( the atl bnrn/11111. Since baw{um is the Latin
word for "stick," Lhl.' phrase il!.cl( cc1lls to mind former U.S. president Teddy Roosevelt's
nd bnrnl,1111 about American foreign policy: "speak softly and carry a big srick."
Another not-so-subtle example of an ad /Jarn/11111 is furnished by the historian
Richard Grunberger. Gru nberger reports that, in pre-war Germany, the Nazis used to
send this reminder to those who had let their subscription to the Nazi newspaper
lapse: "Our paper certainly deserves the support of every G e rman. We shall continue
to forward copies of it to you, and hope that you will not want to expose yourself to
unfortunate consequences in the case of cancellation."5
This example is worth a little reflection . Did the notice give former readers a
reason to renew their subscriptions to the Nazi newspaper? In one sense, the answer
would seem to be that it did. The party that publishe d the paper had a wholly con·
vincing track record of coercion and mistreatment. Thus, at one level, the argument
is a prudetttinl argu ment, an argument abou t what it is sensible, reasonable, or pru·
dent to do. On this analysis, the argument sketches a cost-b
enefit analysis for the
lap s ed reader-and none too subtly. "You may n o t like
subscribing to this paper,
which involves a cost," the au thors imply, "but we shall
se e to it that not subscribing
will exact even higher costs; f?r example, harm to
y our property or possible arrest. So,
on balance, the prudent thing is to avoid the highe
r cost."
It is clearly immoral to make threats of this
nature, but d o we want to say that
th ose who advanced this argu ment were also 1·ncomp et ent reasoners.? 1m ilarly, do '"e
say that thos
' s·
want lo e readers who were moved b
. . y t h"1s argume nt and who resub·
s cr ibed were t ri ked by faulty logic? Probably not.
_ �
Even so, ll 1s true that faulty lo11ic lurks ne
. o a 1·by, rea dy to .mvad e "f I w..., Jet it ·
It
would be a mistake lo reason that the premisses of th· . t
. . ts argument imp ly or JU · stify' no
the prude ntinl l
c aim that readers would do we · ll to resub sen"be, but th era th er differ·
t . the news , .
en c laim that pa p er s Journalistic merits ma ke . sub-
. 1t a paper wor thy of .
scri ber sup port . In th" 1s latter cas e, w e may indeed say t I1at tllde
. we have a fa II acy. -r.l•0 oi,-
us in these matters, here 1s a rule of thumb·· Whenever your oppo nt m akes threats,
ne .
the most t hat you shou ld. concede to him is the pru dence of your do ing what his
roV1ded that the co
threats bid you to do, •p .
sts of non ·co mpuanc 1:
e are too h"1gh. D •
r.iwin8
st o g r c ncl u si n wi -1 I re lt 1
su 11 an instance of the ad bacu/11m fallacy.
any r n e o o
 CHAPTER l. The Quarrel 1S

Exercise 1.3
1 Deter ine whether each of
� the following
choice of whether to believe or not. But if
contains an ad baculum argument
: you don't believe, that will be the result."
*(a) Sue says to Bill, "You had bet
ter stop (h) In a session of collective bargaining, the
smoking, since smoking can kill you ...
union spokesman comments, "If that's the
*(b) Bill says to Sue, ''If you do me this
one
best that management has to offer, we are
favour, I promise to help you with your left with no choice except to take industrial
political science assignment." action."

*(c) Bill says to his employer, "I really do <O A political leader, while defending his
deserve this promotion. After all, an party's dire warnings that racial minorities
unhappy worker could really cause will one day become majorities and then
subject whites to harsh treatment, makes
P,roblems for the firm, couldn't he 7"
the following claim: "What's wrong with
*(d) A gunman stops Bill in the dead of appealing to fear? Lots of people are
1-..night
, tl:
in Stanley Park and. aher afraid, and justifiably so. The ref ore it's a
to K117 him, demands his wallet. legitimate political factor to respond to it.
There's nothing immoral about it."
*(e) Sue says to her neighbour, "If you pay
me the going rate, I'll mow the lawn every (j) Seeing that Socrates is making a
week for the rest of the summer." number of enemies 1n Athens, a friend
gives him the follow,ng adv,ce: "Socrates, I
(f) The latest heavyweight boxing
think that you are too ready to speak ev,I of
champion was arrested and found guilty of
men: and, 1f you will take my advice, I
fraud. So he's obviously not someone who
would recommend you to be careful.
1s especially good with numbers!
Perhaps there is no city in which 1t 1s not
(g) An interviewer, wishing to play the role easier to do men harm than to do them
of the skeptic, asks a fundamentalist good, and this 1s certainly the case at
preacher what reasons he can give fo
r
Athens, as I believe that you know. "6
believing in an afterlife. The p reacher "'2. For each of the arguments listed in
replies: "Do you want to go to hell? Ther
e
r question 1 that contains an ad bacul um,
are two possibilities-eternal sa vation o
l
determine whether 11 invo lves a fallacy.
eternal suffering. If you accept the Lord,
the
you will avoid going to hell. You have

4. The Ad Hominem
r
A nother fallacy often associated with the qua rel is the nd ho111i11c111 fallacy. The term
r
is Latin for "to the man," suggesting arguments that a e directed primarily at one's
opponent, not at one's opponent's argumellfs. The most easily identified example of
the ad hominem fallacy is the abusive ad hominem
.
In our scenarios involving Bill and Sue, there have been at least r.-vo occasion
when the contenders have resorted to personal abu e. Bill stands accused of being a
16 PART I: Arguments

"simpleton," and uc of being ,1 "shrew." H:rc the abuse is obvious, and so ob:iously
.As
misconCl·i vcd or irrelevant that it stands little chance of success as a deception
we have observed, though, if pa!>sions run high enough, even a gross and stupid
remark can be cnrnging :1nd can subvert the objectives of good argument.
ln a second, more harmful, form, the ad ho111i11cm is not abusive but, as logi­
cians say, cirrnms1 1111 1iol. A speaker commits a cirw111sta11tial ad f10minem fallacy
when he refers not to his rival's argument, but rather
The Ad Hominem Fallacy 10 some ( non-abusive) fact or allege d fact about his
An ad hominem fallacy occurs whenever
opponent or his circumstances.
a conclusion is drawn, or invited to be
drawn, on the basis of an irrelevant The cleverest circumstantial ad hominem is one in
appeal to some fact or alleged fact about which the reference to one's opponent is perfectly true,
one's opponent. (When such an appeal is and known to be true, and yet not noticed to be irrele­
relevant. the argument is said to be a \l(Jllt. I Jere is why the strategy often works: If the remark
non-fallacious ad hominem argument.) is true, then how could such a reference be contrary to
In the modern tradition. there are two
main types of ad hominem: the abusive the purposes of responsible reasoning? For is t ruth not
ad hominem (in which an insulting or the natural ally of reason?
unwelcome fact or alleged fact about The following example is instructive: Bill and Sue
one's opponent is advanced) and the are again battling. This time the point of contention is
circumstantial ad hominem (in which whether abortion on demand is morally justifiable. Sue
some non-abusive fact or alleged fact
says, in utter exasperation, "Blast you, Bill, you're a man..
about one's opponent or his circum­
stances is advanced). You just can't see it!"
True, Bill is a man. Even so, the potential for dam­
age Lies not in what Sue is saying, true as it is, but rather in what she leaves unsaid and
only implied. Her implication is that a man will oppose abortion on demand because.
as a man, he cannot ever experience an unwanted pregnancy. Accordingly, he cannot
help himself in opposing abortion. Because Sue's accu ations are not explicit y st ated,
l

they may have an 1111detected effect on the argument. This example crops up repe t­
a

edly in contemporary discussions about abortion and related issues. Many pe ple
o

commit the "you are a man (so you can't help being opposed)" fallacy. Ironically,
they seem not to be aware that women are open to a similar attack. Women, pr ­
e

sumably, are in a position to experience an unwanted pregnancy; thus, by parallel

reasoning, they will be unable to keep themselves from favo11r it1g abortion on dema d
n

you
Yet if neither you nor your opponent is free to choose which side of the argument
n
are to defend on the basis of evidence (rather than as a result of your own perso �
5
circu mstances), then what is the use of continuing the argument? The moral her� 1
pamt
that even when the truth is told, if the claim being advanced is irrelevant to the
of an argument, then what we have is an ad homi11em fall acy.
A circumstantial ad ho111ine111 may also arise whenever there is an apparent
inconsistency between the thesis that some participant in an argument is putting
n
forward and some circumstance or circumstances pertaining to the participant hir ·
i le
self. 7 For example, a person might recommend some policy as a worthwhile princ p
even though circumstances indicate that he has not followed--0r is not following-­
that policy himself. Any person caught in such an apparent confl ict can be highly
vulnerable if his opponent attacks ad hominem. "You don't even practise what �u
preach!" his opponent may imply. The policy's advocate then appears to be caught Ul
 CHAPTER 1: The Quarrel 17

significant kind of inconsistency. He

tells us we O ugI1t to dO som
ething, but he does
:.i

not even do it himself.

It is important not lo ovei·est·imate
the force of ti115
. fiorm of argument. A person's
inability or fuilure to follow th e advice . embodied . .
.� 11'1 his own thes1s . 1s
. intere
. sting, sig-
n1f 1cant, and sometimes even conitc . aJ but, des .
pite this • he moy st1·11 be able to support
his thesis by cogent argument · A spea ker c,rom Alc
oholics A nonymous may give us
good advice on how to avoid .
� beeommg an alcoholic ' even though hc imse
h. If may
remain an alcoholic. Another exam 1 o curs .tn the following
dialogue. A parent
has just finished citing evidenc; of mek s � etween smokin
i- _
g and its negative health
consequences , such as I ung cancer:

Parent·. So you see, smo k'mg is ·

a bad business to start. Once you start it is
. '
d1ff1cult to stop.
Child: Yes, but you smoke. How can you tell me not to smoke if you do it
yourself?

Here th� child may have a worthwhile point. If the parent is not following the
_
course of act.Ion being recommended, this couJd be evidence of insincerity. Even so,
the parent's argument that smoking is bad for one's health may also be based on
good evidence, and the fuct that smoking is addictive may weU figure into both the par­
ent's continuing to smoke and the advice being given to the child. Thus the child
should not reject the parent's argument solely on the basis of a circumstantial incon­
sistency. In such circumstances, two questions need to be addressed: First, is the par­
ent's thesis substantiated by the evidence being cited? Second, does the parent's
is incon­
recommendation conform to his or her own circumstances and, if not, this
sistency relevant or irrelevant to the issue at hand?
their core two propositions that
Circumstantial ad hominem arguments have at
proposition is a thesis recom­
are thought to be inconsistent with one another. One
on that describes some circumstantial
mended by the arguer, the other is a propositi
ositions are the premisses of the ad
fact about the arguer himself. These two prop
the inconsistency between these two
hominem. Nor mally, a fallacy occurs when
issue at hand, but sometimes a fallacy
premisses tur ns out to be irrelevant to the
n these very pre mis ses are incorrec tly
oc curs in an ad hominem argument whe
umstances, what can happen is that a third
described or confused. Under such circ
the original pair, may be illicitly substituted for
proposition, which resembles one of
ument. When this happens, there may appear
one of the original p remisses in the arg
ly exists. The result is another subtle form of
to be an inconsistency where none real
the ad hominem fallacy.
i us unter's Argument repor ted
An interesting ex.ample of this point is the notor o H
8
-centu ry Irish ar chb ishop Richard Wl1ately. The argument goes
by the nineteenth
roughly as follows:
Critic How can you derive pleasure fro m gunning down a helpless animal?
t for amusement is barbarous.
Surely the killing of deer or trou
f ?
Sportsman: If you're so concerned, why do you feed on the flesh o animals
Aren't you being inconsistent?
18 PART I: Arguments

What the sportsm.in is suggesting is that the critic's actions are inconsistent with
his moral pronouncements. The sportsman replies Lo the critic by arguing ad hominem.
According Lo the sportsman, the critic cats meat while at the same time assert­
ing that hunting is morally repugnant. ls this position inconsistent? The answer is
(likely) not. It would be inconsistent to hunt while asserting that hunting is morally
repugnant, but there is no inconsistency in condemning, as the critic does, the killing
of animals for nI1111se111c111 while eating a steak dinner. If it can be shown that hunt­
ing is (at least in some circumstances) the same as killing for amusement, there is
enough of a difference between hunting and meat-ec1ting to invalidate the sports•
man's argument.
Related to this example, the nineteenth-century logician Augustus De Morgan
describes the ad l10111i11e111 foUacy as an iUicit comparison producing what only appears
to be an inconsistency: "It is not absolutely the same argument which is turned
against the proposer," says De Morgan, "but one which is asserted to be like it, or
parallel to it. But parallel cases are dangerous things, liable to be parallel in immate­
rial points, and divergent in material ones." 9 We can see this type of illicit compari­
son more dearly if we contrast the genuinely inconsistent pair
(1) X asserts that hunting for amusement is wrong
(2) X hunts for amusement

with the parallel pair

(1 ') X asserts that hunting for amusement is wrong
(2') X eats meat.

The sportsman, it seems, has not made a sufficiently strong case against his critic.
To make an accusation ad hominem is to make a charge of inconsistency. We
will learn more about inconsistency as this book proceeds. We will see that a logical
inconsistency occurs whenever a proposition is advocated and denied at the same
time. To assert that "Socrates was born in Athens and Socrates was not born in
Athens" is to assert a proposition that is logically inconsistent. In examining the ad
hominem, we have not been discussing logical inconsistency, strictly speaking; instead,
we have been discussing an inconsistency between what actually happens and what
one says should happen. We call this kind of inconsistency a circumstantial or pragmatic
inconsistency. In short, a person's action is circumstat1tially (or pragmatically)
inconsistent with what he proposes or recommends if and only if, were the proposal
or recommendation to be an order, the person's act ion would be an act of se lf•
disobedience.
A speaker can be involved in a circumstance, or even create one himself, that
not only makes what he says inconsistent with that circumstance, but also actu lly
a
refutes what he says. A very simple example arises if Bill claims, "Since I never speak
English, what I am now saying is not in English." So it is worth noticing that the cor­
rect attributio_n of an ad hominem can be conclusion•defeating; not all uses of the ad
hominem are incorrect or fallacious.
lf a pair of propositions is logically inconsistent, one of them must be false.
However, it is also worth noting that further argument may remove or explain awaY
 CHAPTER l; The Quarrel 19

Flashback on...
the Ad Hominem
ARISTOTLE was the first to identify ad hominem It is natural to wonder what a genuine refu­
arguments. Centuries later the Englishman John tation (in Aristotle's sense) establishes. What it
Locke took up the topic in his 1690 publication does not show is that the respondent'.s thesis, T,
An Essay concerning Human Understanding. is false. In the absence of further information, all
_
There he introduced a variety of arguments with a refutation shows is that the respondent has
the words: "it may be worth our while a little to given an inconsistent defence of his thesis. This
reflect on four sorts of arguments. that men, leads Aristotle to conclude that. "About such
in their reasonings with others do ordinarily matters, there is no proof in the full sense,
make use of. to prevail on their assent; or at though there is proof ad hominem." 12
least so t o awe them, as to silence." 1° What Aristotle means is that the argument
The third of these four sorts of argument he from A 1 and A2 to the denial of T does not by
describes as follows: "a third way is to press a itself prove T to be false, so it is not a "proof
man with consequences drawn from his own in the full sense." However, it does prove that
principles or concessions. This is already known the respondent cannot consistently continue to
und er the name of argumentum ad affirm T while also accepting A 1 and A2. In
hominem." 11 Perhaps the most important thing other words, because the respondent has given
to say about Locke's ad hominem is that. unlike an inconsistent defence of T, the refutation is
its modern namesake, it is not a fallacy. To see a proof ad hominem.
Locke shares this understanding of the dif­
r_hy this is so, it is instructive to consider a kind
of ad hominem argument that Aristotle called ference between a proof in the fu/1 sense and a
"refutation." An Aristotelian refutation is a proof ad hominem. An ad hominem refutation
deducuon (i.e .. a condusive argument) produced of the respondent's thesis shows. Locke says,
in the following way: There are two parties to a that the respondent is "in an error" and "in the
discussion, a questioner and a respondent. The wrong," but it does not show exactly where the
respondent gets things started by putting for­ error lies. It could be T, but for all we know it
ward a t hesis , T (e.g .• "All v irtue can be could instead be A 1 or A2.
taught"). The questioner's role is then to attempt Thus, as understood by both Aristotle and
to refut e T. He does this by putting questions Locke, ad hominem arguments need not be fal­
to the respondent that can be answered either lacious. Even so, whenever we make the addi­
yes or no. Out of the respondent's answers to tional assumption that just because there is a
these questions (say. A 1 and A2). the questioner refutation of T (in Aristotle's sense) T must be
then tries to select propositions that logically false. we will have committed a fallacy.
imply the opposite of the respondent's thesis. It is interesting that the modern conception
In this way, T's refutation comes from the of the ad hominem has evolved into something
respondent's own words, since the questioner quite different from the older concept. It has
has presented the respondent with a conse­ been quite a transformation: from non-fallacy
quence drawn from the respondent's own con­ to fallacyl In discussing the ad hominem, care
cessions. Such a refutation is distinguished from shoul ? therefore be taken not to attribute mod­
a sophistical refutation (or fallacy), an argument ern views to ancient writers, or ancient views
on but to modern writers.
that appears to be a (successful) refutati
that in fact is not.
20 PART I Arguments

a merely app,1rc11t i11co11sisle11cy. In somc circumst,,nces the nd hominem can be

defuse d. Shortly af1cr the Second World W.,r, the British philosopher and social
activist l3crtrand Russell .1<lvoc.11�·d ,1l1;11.:king 1hc Suvicl Union as a way of guaranteeing
the non-prolifcratio11 of nuclear wc.ipons. Liter, in the 1950s, he advocated a "better
Red than de.id" argument, preferring appc,1scmc11t lo confrontation. Are these two
views logically i11co11.,istc11t? l\:rhaps. But Rw,�cll bclicvtd th.ii he was able to explain
why he had changed his mi11d. Ucforc thc Soviet Union had acquired nuclear weapons,
Russell thought that aggression was a rntional policy; l:llcr he came to hold the view
th:11 such a policy would be irr,1tio11al in de.ding with a country that now had a
strong, effcctiw 11uclc.1r c,1p.1bility. In short, rightly or wrongly llu5..�cll believed he could
give :i satisfactory account of why ,1 ch,tngc in his pmition was war ranted. Whether
one agree with hi:. ultim.1\e conclusion or not. llu,sdl should not be faulted for
simple inconsi tcncr in his vkwpoint. A,,, rc,ult, there i'I a reasonable basis for
defending him .,g,1in. t this type of ad lio111111c,,1.

Exercise 1.4

1. Determine whether each of the following government should be spending more

contains an ad hominem argument money on education." says one, "you have
*(a) At a political rally, Sue says to 81ll, five children!"
"Since Dr. Jones is a tw1ce-conv1cted !f} After fund,ng a questionable youth­
embezzler, he's unfit to serve as city grant pro1ect, the government is accused
treasurer .
H
by the opposition of misusing funds. A
*(b) While attending their biology class. member of the government replies that the
61II says to Sue. "S,nce Dr. Jones ,s a twice- oppos,t1on should be careful in pressing
convicted embezzler, we shouldn't pay th,s sort of cnt1mm, given its own
much attention to his views about notorious record of squandering money on
evolutionary biology." youth grants when it was in power.

*(c) Sue says to Bill. "Since you concede {g) Two friends are arguing over which
that there are particular c,rcumstances in political party has the better foreign policy.
which abortion ,s morally perm1ss1ble, your One of the two suggests that his friend' s
blank.et opposition to abortion ,s uncalled party has never been a strong advocate of
a h19h-prof1le international presence. wof
for."
course not." ag rees his fnend... all their
•(d) B111 says to Sue. "I don't think that I politlcal support is to be found at home. N
should trust your Judgment about what
(h) We are told by the Scottish biographer
1
software to buy. After all, the last time I
used your computer it was always James Boswell that Dr. Johnson. the
crashin g." famous English lexicographer, once asked
someone who argued for the equality of all
•(e) Two friends are discussing the current mankind-in the presence of a servant­
fund1ng of their local elementary school: whether she would allow her footman to
"It's not surprising that you think the sit down beside her.
 CHAPTER I: The Quarrel 21

(i) Another time, after listening

to a man abolition of bourgeois property, the
who had cleverly argued that the
difference standard of your bourgeois notions of
between virtue and vice is illusory, Dr.
freedom, culture, law, etc. Your very ideas
Johnson is said to have remarked:
"When are but the outgrowth of the conditions of
he leaves our house let us count our
your bourgeois production and bourgeois
spoons."
property.... " 13
U) In their infamous Communist Manifes • 2. For each of the arguments listed ,n
to,
Karl Marx and Friedrich Engels reply lo their
question 1 that contains an ad hominem,
critics as follows: "But don't wrangle with
determine whether it involves a fallacy.
us so long as you apply, to our intended

Summary f
Our goal in this book is to learn what it means to argue well. As a result, we have
begun studying logic i11 the 1111rroiv sense and logic in the broad sense. We have also
begun studying the fallacies-those arguments that often appear to be good but are
not.
An arg11mem is a presentation of reasons or evidence in support of some claim.
11 is an attempt to build a case in favour of a conclusion. Normally this means that
arguments are social exchanges involving a series of speech acts uttered by two or
more parties. Familiar examples include quarrels, legislative debates, diplomatic
wrangles, labour negotiations, mathematical proofs, legal briefs, scientific demon­
strations, and so on. Because they involve a variety of social and psychological, as
well as logical, factors, we say that these are arguments in the broad (or social) sense.
At the core of every argument in the broad sense is a set of propositions composed
of the argument's premisses and conclusion. We call this an argument in the narrow (or
propositional) sense. Arguments in the narrow sense are abstractions from arguments
in the broad sense.
In order to make clear the structure of an argument in the narrow sense, and
to distinguish between its premisses and conclusion, we often place an argument in
standard form. To do this, we identify its premisses and conclusion and separate the
conclusion from the rest of the argument by means of a horizontal line. The pre­
misses form the basis of the argument. They constitute the evidence that is being
offered in support of the conclusion.
Reasoning critically about various issues requires that we learn to evaluate
arguments. The quality of an argument depends primarily upon ( 1) the truth of the
argument's premisses, together with (2) the strength of the co11seq11mct· relatio11 that holds
between its premisses and conclusion. It follows that there are two main ways to
criticize an argument: ( J) we can show the premisses to be unacceptable, or (2) we can
show the consequence relation to be weak.
It is a primary goal of both logic in the narrow sense and logic in the broad
smse to help determine the quality of arguments. logic in the narrow sense does this
by studying the nature of the consequence relation that holds between an argument's
22 PART I: Arguments

premisses and conclusion. lb say that a conseque nce relation is strong is to say that the
premisses, if true, provide goorf cl'irle11cc in favour of the conclusion. To say that this
relation is weak is to say that the premisses, if true, provide poor ev;dence in favour of
the conclusion. A consequence relation that is neither strong nor weak is said to be
moderate. Lo�c" i11 1/ie uroarl sense studies other
factors in addition to an argument's
consequence relation.
One type of argument with which we arc familiar is the quarrel. The quarrel is
an argument that suffers from prc111issory or co11c/11sio11<1I i11stnuility (or both) in ways
that dispose the disputants Lo hostility.
In addition, quarrels often involve two well-known types of fallacy: the ad
bacu/11111 and the nd lzo111i11c111. An <1rl u11rnl11111 ,ugurnent occurs whenever a conclusion
is drawn, or invited to be drnwn, as a result of an appeal to intimidation or force.
The argument is fnl/11cio11s whenever the appeal turns out to be irrelevant to the
conclusion under consideration.
An nd ho111i11e111 argument occurs whenever a conclusion is drawn, or invited
to be drawn, as a result of an appeal to some fact (or alleged fact) about one's
opponent. Once again, the argument is fallacio11s whenever the appeal turns out to be
irrelevant to the conclusion under consideration. There are two common types of
ad hominem fallacy in the modern tradition: the nb11sive ad hominem and the
circumstantial ad hominem. Both differ from the ad homi11em as understood by John
Locke and Aristotle.

Weblinks I
Episteme Links-Logic and Philosophy of Logic
www. epistemelinks.com/Main/Topics.aspx ?Topi( ode=Logi
Philosophy in Cyberspace-Logic
www-personal.m onash.edu.au/-dey/phiVlogic.htm
Episteme Links-Aristotle
www.epistemelinks.com/Main/Philosophers.aspx7Phi1Code=Aris
MacTutor History of Mathern Archive-Aristotle
��
��-dcs.st-and.ac.uw/history/Mathematicians/Aristotle.html

Notes I
1. In this book we adopt the spellings "premiss" and " prem isses ... The American philOSO­
pher C.S. Peirce points out that the spelling appropriate lo logic derive from the medieval
Latin praemissa. In contrast, the spellings "premise" and "premises" (as ins "He was removed
from the premises for trespassing") come original ly from the French premise, as in the
expression Jes choses pr�mises, used in inventories. we came upon this point in TA Goudge,
The Thought of C.5. Pe,rce (Toronto: University of Toronto Press, 1950), 5.
2. We will say more about these historical issues as we proceed,
especially in chapter 8.
3. This third alternative occurs as a result of procedure s d i d bates and r i l ls
su e n e c im na tria .
These procedures are discussed in chapter 2 a d
n chapter 13, respectively.
4. Aristotle, On Sophistical Refutations, 15, , 74a, 20.
 CHAPTER f. The Quarrel 23

5 Richard Grunberger, A Social History of the Third Reich (Lon

don: Weidenfeld and
Nicolson, 1971 ), 398.
6 Plato, Meno, 94e.

7 Circumstantial ad hominems of this type are sometimes also called "tu quoque" argu­
ments (Tu quoque is Latin for "you too.") For a more detailed analysis, the Interested reader
may consul t John Woods, "Dialectical Blindspots," Philosophy and Rhetoric, 26 ( 1993),
251-265.
s Richard Whately, Elements of logic (New York: WIiiiam Jackson, 1836), 197.
g Augustus De Morgan, Formal Logic (London: Taylor and Walton, 1847), 265.
10 John Locke, An Essay concerning Human Understanding (1690), bk. 4, chapter 17 ,
§19
11 Ibid , §21
12 Anstotle, Metaphysics, bk. K ( 11 ). 5, 1062a, 2-3. See also his On Sophistical Refutations.
22, 178b. 17; 33, 183a, 22-24; and Topics, bk. 8, 11, 161a, 21-23.
l 3 Karl Marx and Friedrich Engels, Manifesto of the Communist Party (New Yorlc: International
Publishers. 1948). 26.