Critical Thinking

CRITICAL THINKING
Kevin Possin
Philosophy Department
Winona State University
The Critical Thinking Lab
2002 Kevin Possin For purposes other than making the purchaser's
back-up, no portion of this etext greater than a paragraph or so may be
reproduced in any form without the written permission of the author.
Library of Congress Control Number: NA
ISBN: 0-9712355-1-1
Published by

24847 County 17
Winona, MN 55987
Thanks to Fred Dretske, for making me the roving teaching

assistant for his critical thinking course.
And thanks to Ann.
Many people would sooner die than think. In fact, they do.
Bertrand Russell
Much that is taught in college classes grows soon out of date,

but the skills of correct reasoning never become obsolete.
Irving Copi & Carl Cohen
You can fool too many of the people too much of the time.
James Thurber
. . . all our dignity lies in thought. Let us strive, then, to think well.
Blaise Pascal
Our best guide to truth is free and rational inquiry; we should

therefore not be bound by the dictates of arbitrary authority,
comfortable superstition, stifling tradition, or suffocating orthodoxy.
Statement of Purpose, Free Inquiry Magazine
. . . arguments, like men, are often pretenders.

Plato
TABLE OF CONTENTS
1. Introduction
2. Argument Identification
3. Anatomy of an Argument
26
4. What Makes an Argument Good? -
32
5. The Use and Abuse of Language
51
6. Acceptability
88
7. Relevance
118
8. Groundedness I: Categorical Logic
152
9. Groundedness II: Propositional Logic
222
269
Index
Introduction
1. INTRODUCTION
Why critical thinking?
Everyone agrees that one ought to write well, speak effectively, and know at least
enough math to avoid bouncing checks. Thats why these subjects are the socalled basic skills courses that everyone is required to pass in school. But all
of these skills have a more fundamental skill underlying themthinking.
Composition and speech courses teach you how to express your thoughts more
effectively, but they dont teach you how to decide which thoughts ought to be
expressed. And thats our project hereto learn how to think more effectively
and critically.
Is this project relevant to you and your future career? Well, unless you and your
future are totally devoid of thought, yes! In fact, if youre thinking that you
dont have much need for thinking, youve already proven yourself wrong.
Times have changedluckily!

This package consists of the Critical Thinking etext, the Critical Thinking
software, and Self-Defense: A Student Guide to Writing Position Papers. It
offers you a unique opportunity to develop your critical thinking abilities. It
enables you to work to competency. This can be a self-mastery project
(which used to be called do it yourself), or it can be done in a more traditional
classroom setting.
Let me explain all this by first telling you about my experience taking a critical
thinking course.
I didnt catch on very quickly at all! In fact, I flunked the midterm. As we were
going over the midterm in class, however, all the pieces fell into placethings
suddenly made senseI finally understood the stuff. All I needed was a bit more
practice, just one more homework assignment. But there was no way my
overworked prof could have graded all the homework I needed to finally get my
head around things. And where did that leave me? Well, even though I aced the
exams after that, I was doomed to settle for a C for the coursea grade that
didnt accurately represent my final competencies.
Those days are over! You will never have to worry about whether you really
know the stuff (and find out too late that you dont). Just read through this etext,
which introduces the basic elements and skills of critical thinking, and then
review what youve learned and hone your skills using the CT software.
Introduction
The CT software has thousands of exercises. So youll be demonstrating your

competencies long before you make more than a ripple in its pool of exercises.
You get immediate feedback as you do practice exercises, so you learn from your
mistakes. And you can take and retake exams, to assess and demonstrate what
youve learned. As you take exams, the exercises that youve missed are noted
for review, so you can pinpoint exactly where you need to focus your learning
efforts.
So if you can click a mouse and want to learn how to think critically, the
opportunity has never been better. [And, if you order now, Ill throw in this
lovely set of steak knives absolutely free!]
The goal of critical thinking

The goal of critical thinking is to learn about the world and avoid being fooled in
the processto find truth and avoid error.
Something to think about: Why have both of these goals?

Why not try to get by with just oneeither believing what is true
or not believing what is false?
Discussion: Lets see what would happen if we tried to get by
with having only one of the two goals of critical thinking.
Lets first try believing what is true, without caring a bit about
avoiding error. Well, the easiest way to believe all the truths,
would be to believe everything. Of course, this shotgun
approach means you believe as many falsehoods as you believe
truths. Youve met your single goal, right? Yes, but youve
become the most gullible dolt in the process. Thats not what
critical thinking is!
Now lets try it the other way; lets try avoiding error, without
caring a bit about discovering truths. Well, the easiest way to
avoid erroneous beliefs is to not form any beliefs at all. But
thats simply impossible; as soon as you open your eyes, you
form beliefs! The only way to avoid forming beliefs is to be
comatose or dead. And thats not what critical thinking is!
If believing all and only whats true is our goal in critical thinking, how do we
reach it? What method should we use to form our beliefs so as to maximize this
goal? This is an especially good question in light of the fact that there is an
Introduction
infinite number of methods for deciding what to believe. For example: Believe
the person with the most money, believe the person who talks the loudest, believe
what the Magic 8 Ball says, or believe whatever I tell you [especially about
where to send those cashiers checks for the rest of your life]. But these methods
of belief formation are going to do a pretty lousy job of achieving our goal of
critical thinking, viz., of getting us the truth and nothing but the truth.
So, whats the best method of belief formation? Its having reasons for what
one believes. And not just any old reasons! They have to be good reasons! So,
believe whatever you have good reasons to believe. And believe false whatever
you have good reasons to believe is false. And, if you have no good reason to
think something is true and you have no good reason to think its false, then
simply dont believe anything about it at alljust withhold belief. Honest
ignorance beats false belief any day!
If critical thinking is the practice of forming beliefs on the basis of only good
reasons, what are these reasons and what makes them good reasons?
The best way I know of to come to grips with this fundamental question is by
rewording it a bit. To have good reasons for ones beliefs is to have good
arguments for them. So the practice of critical thinking is the practice of
identifying, having, and giving good arguments for what one believes. By
looking at the structure and quality of arguments, then, we will get a clear picture
of the structure and quality of critical thinking.
The elements of critical thinking

Now that weve clarified a bit what the skill of critical thinking is, lets bust it up
into more bite-sized competencies for further study. These are the skills you will
be developing throughout this project. So by reading my introductory blurbs in
each chapter of this etext, practicing and assessing what youve learned using the
CT software, and studying the Self-Defense manual, you will develop the
following critical thinking competencies:
Identify Arguments: Identify arguments by their function.

Distinguish arguments from, for example, assertions,
descriptions, explanations, and other speech acts.
Dissect Arguments: Identify and distinguish the parts of

argumentspremises, conclusions, and subconclusions.
Assess Arguments: Identify and assess the goodness, or

cogency, of arguments in terms of 1) the truth or
Introduction
acceptability of their premises and 2) the relevance of their

premises to the truth of their conclusions.
Identify Fallacies: Identify types of fallacious arguments

that are commonly used to trick people into adopting beliefs,
values, and actions unjustifiably.
Taxonomize Arguments: Identify and distinguish between

deductive and inductive arguments and know the different
standards of cogency for both. Be able to 1) assess the
validity of deductive arguments and 2) assess the strength of
inductive arguments, namely, the probable truth of their
conclusions given the evidence cited in their premises.
Critically Review Definitions: Analyze terms and concepts

to determine, when possible, the necessary and sufficient
conditions, i.e., the essential properties, of what is being
defined.
Critically Defend a Position: Critically select and defend a

position and critically review relevant alternative positions.
Not everyone is going to agree with this list of priorities. Thats because not
everyone agrees on what critical thinking is. In fact, recently peoples views on
what critical thinking is have sort of slipped the leash to the point where just
about any thought process passes for critical thinking, by virtue of being creative
or contemplative or self-reflective or by virtue of just being thinking. This is
most unfortunate, because this has mucked things up more than clarified
themtotally contrary to what critical thinking should be accomplishing. A
drunken stupor might well be creative, contemplative, self-reflective, or
whatever, but its hardly an instance of critical thinking!
Calling something critical thinking doesnt make it so. We always need to
remember: Calling a dogs tail a leg doesnt make it a leg, and the poor dog
that thinks it does lands right on its ass.
This isnt going to be a logic course, is it?

Not really, no. I wouldnt like that either. Its not going to be the study of how
to jerk symbols around just to see if you can get from one set of symbols to
another set of symbols by means of a certain set of rules. We will be learning
how to use some symbols, but only as a means to helping us think critically.
You know how sometimes you need to abreviate stuff on scratch paper, because
Introduction
you just cant keep it all in your head? Well, thats the extent to which well be
studying how to use symbols.
This isnt going to turn me into a machine, is it?

Probably the most famous example of a critical thinker is Mr. Spock, from the
original Star Trek. He was always so logical and never seemed to pass up the
opportunity to point out how illogical others were. Lets face it, Spock was a
geek.
But a lot of Spocks problems were the result of being misunderstood [and those
pointy ears didnt help him either]. He was constantly accused of being
unfeeling; in fact, even he thought emotions were antithetical to his goal of being
logical. But he was all wrong about this one: Every time he fought so hard
for what he had every reason to believe was right, he betrayed how critical
thinking and the emotions work together. If Spock were purely logical, hed
be completely rational in his beliefs about what ought to be done, but he
wouldnt care enough to do anything about it. Critical thinking without
emotion is inert; emotion without critical thinking is dangerous. One
shouldnt be sacrificed for the other; rather one should assist the other.
This isnt going to affect my social life, is it?

Well, it could. Remember how a bit of Spock sometimes went a long way?
Sometimes unloading your critical thinking on others will similarly not win you
any points at the party. Sometimes you might find that critically thinking about
your own beliefs and values causes you to change them in ways that leave you
with less in common with your friends. Sometimes you might find that the
position you discover as having the most reasons in its favor is one that is quite
out of favor with your peers or your friends or your partner. Popularity can be
the price you pay in your pursuit of the truth. And thats why knowing when to
ease up in your application of critical thinking skills to the views of others is
itself a critical thinking skill, since reaching your goal of truth can be rather
hollow when theres no one around to share it with.
So with introductions out of the way, lets begin working on our first criticalthinking skill: Argument identification.
Argument Identification
2. ARGUMENT IDENTIFICATION
If our goal is to be able to identify and give good arguments for our beliefs and
values, we need to be able to identify arguments. And in order to do that, we
first need to know what an argument is. [Just think about ithow could you
identify birds if you didnt know what a bird is?!]
So, whats an argument?

Well, heres a quite famous example of an argument, from John Stuart Mills
essay On Liberty.
The peculiar evil of silencing the expression of an opinion is that
it is robbing the human race; posterity as well as the existing
generation; those who dissent from the opinion still more than
those who hold it. If the opinion is right, they are deprived of the
opportunity of exchanging error for truth. If wrong, they lose
what is almost as great a benefit, the clearer perception and
livelier impression of truth, produced by its collision with error.
To help us understand what exactly makes this such a good example of an
argument, here are four accounts of what an argument is. All four accounts are
driving at the same idea; they just try to get there from different angles. I hope at
least one of these gives you a good feel for what makes an argument an
argument.
An argument is:
A set of claims or statements {P}, called premises, that are

intended by someone to make another claim or statement C,
the conclusion, acceptable or justified or rational to believe.
A set of statements {P} that is intended to be a reason to

believe C.
A set of statements {P} that is intended to be why one ought

to believe C.
A set of statements {P} the truth of which is claimed to

increase the probability of the truth of C to the point to
which C is probably true, given that {P} is true.
Something to think about: What do the intentions of the

author have to do with making a set of statements into an
argument?
Hint: Everything! Without intentions being a part of the
definition of an argument, what type of argument would be
impossible, which we know all too well is quite possible?
Discussion: If we redefine what an argument is, leaving out the
part about the authors intention, we would get the following, for
example.
A set of claims or statements {P}, called premises, that

actually makes another claim or statement C, the
conclusion, acceptable or justified or rational to believe.
A set of statements {P} that actually is a reason to

believe C.
By this time, you probably see what all-too-popular kind of

argument would become impossible: A bad argument. A bad
argument fails to make its conclusion rational to believe. So,
bad arguments couldnt be arguments anymore; theyd have to
be schmarguments, or something. But thats just silly. Of
course there are bad argumentstoo many of themand so our
definition better make plenty of room for them.
According to our analysis, then, an argument is an argument by virtue of the fact

that someone is using it as an argument; just as a screwdriver is a screwdriver
because someone is using it as one. When someone is using it to pry open paint
cans, its a can lid opener; and when someone is using it to hold down some
paper, its a paperweight.
Something to think about: Heres a popular definition of what

an argument is [from the discipline of rhetoric]:
A linguistic means of persuading others into adopting
ones beliefs or values.
See if you can find whats wrong with this definition.
Hints:
Can you think of A linguistic means of persuading

others into adopting ones beliefs or values that is not
an argument?
Can you think of an argument that is not A linguistic

means of persuading others into adopting ones beliefs
or values?
Discussion: As examples of A linguistic means of persuading

others into adopting ones beliefs or values that are not
arguments, one could cite brainwashing, subliminal advertising,
and hypnosis. In these cases, language is used to influence what
a person believes or values, but those beliefs or values are not
adopted or influenced by way of argumentation. The language is
acting much like a neurological probe, merely causing the beliefs
and values, rather than giving the person reasons for thinking
that those beliefs or values are correct, the person thereby
adopting those beliefs and values by virtue of those reasons.
As an example of an argument that is not A linguistic means of
persuading others into adopting ones beliefs or values, one
could cite one of many popular non-linguistic means of
argumentation, e.g., the gas gaugeits needles being on E is a
great argument for the conclusion that the tank is empty.
One could use the case of an argument that is so bad that no one
is persuaded by it. For example: The world is going to come to
an end on September 17, 2050 because you are reading this
paragraph today. This argument sucks so badly that no one is
persuaded by it; but yet someone could still offer it up as an
argument [at least until theyre back on their meds!].
As another example of an argument that is not A linguistic
means of persuading others into adopting ones beliefs or
values, you could use an argument with a totally obvious
conclusion. For instance, if youre reading this page on-screen
right now, you surely know youre looking at a computer screen.
But I could still give you an argument for that conclusion; e.g.,
your perceptual experience is such that it sure looks as if youre
seeing a computer screen, and your lighting conditions are fine,
and you havent taken hallucinatory drugs [recently!], etc., etc. I
just handed you a fine argumentan incredibly boring and
unnecessary one, thats all. But it persuaded no one, because its

audience [viz., you] already believed its conclusion.
So, it looks as if our proposed account of what makes an
argument an argument is pretty good, especially compared
with the competition.
Argument indicators
The two essential parts of an argument, the premises and the conclusion, are
often connected by certain terms or phrases. These terms indicate whether
statements are being used as premises or as conclusions. As a result, these
indicators are good indicators of arguments. Heres a list of some of the most
frequently used premise indicators and conclusion indicators. [This list is not
exhaustive, but youll easily be able to extrapolate to new cases after becoming
familiar with these.]
Premise Indicators:
since
because
is implied by
is entailed by
is inferred from
for
given that
follows from
as shown by
as is demonstrated by
for the reason that
on the basis of
on the grounds that
is deduced from
is derived by
Conclusion Indicators:
therefore
hence
thus
so
then
it follows that
implies that
entails that
can be inferred that
for this reason
accordingly
consequently
proves that
shows that
demonstrates that
one can conclude that
Something to think about: Whats wrong with the following

arguments:
He has only 45 credits. This infers that he cannot
register for the course until Tuesday, at 9:00 am.
The median stock holdings for those under 35 is
$11,900; for those age 64 or older, $62,500. That infers
abysmal yearly growth of about 5% and no additional
savings over decades. [Danial Kadlec, from his Your
Money column, Time, November 1,1999]
Discussion: Premises can imply conclusions. They can entail
conclusions too. And conclusions can be inferred from
premises. But premises cannot infer conclusions. Think of it
this way: People inferpremises imply. An inference is a
thought process in which one belief causes another by virtue of
their contents. Ones belief of the premises causes ones belief
of the conclusionthats what it is to infer the conclusion from
the premises. Premises, on their own, are not thoughts or
psychological subjects [thinkers], so they cant infer anything on
the basis of anything.
BTW: What do you think? Can a dog or a cat infer?
Mere assertions
Now that we are clear on what an argument is, we can begin to look at some of
the other important things people do with languageother speech acts, as they
are called.
People often do not bother to give an argument for what they believe or value;
they rest content simply stating what they believe or value. If their statement is
not obviously true, if instead it is quite controversial, their statement is a mere
assertion.
10
Consider, for example, Abortion is perfectly moral, or Abortion is murder.

Neither of these statements is obviously true; in fact, they are competing
statements in one of the fiercest debates in ethics. And you are given no reason
to believe either of these statements. And thats why they are mere assertions.
Heres another example: In a Reply to a system-wide email detailing some
problems with a new proposal, a recipient writes, I think the message below is
alarmist and obstructionist. This is a mere assertionthe writer gives no
reasons for thinking that the problems discussed in the original email were
alarmist and obstructionist.
So what should you do when someone hands you a mere assertion?
Remember, theyve given you no reason to believe their claima claim thats
not obviously true to begin with. To do anything, including adopting a belief
or value, for no reason is irrational. So the only rational thing to do when
someone hands you a mere assertion isnothing. Dont believe the assertion;
dont disbelieve it either; just withhold belief. Thats the only rational option to
exercise.
Mere assertions are cheap! Anyone can have an opinion and mouth it off to the
world. But you neednt give them the time of day until they give you a reason to
believe what they are claiming is trueuntil they give you an argument as to
why you too ought to believe it.
So thats the difference between an argument and an assertion. Basically, an
argument stripped of its premises becomes a mere assertion; an assertion with an
accompanying reason to believe it becomes an argument.
But isnt everything just a mere assertion?

You hear this view stated many times and in many different ways:
Everything is just a matter of opinion.
Its all just a matter of taste.
Things are only true for an individual or a culture.
Everything is relative.
All opinions are equal.
There are no facts.
There are no truths.
Nothing is absolute.
Nothing is objective.
Everything is just subjective.
11
This view is called relativism. It has become very popular, advocated by many
authors and educators. But its a view that is self-defeating and utterly false.
Heres why.
When someone advocates relativism, they intend their position or statement to
be true. And not just true for them, but simply truetrue for everyone. But
then, if relativism is true for everyone, relativism must be false, since there
would then be at least one statement that is true for everyone, namely the
statement Relativism is true. But thats exactly what relativism says cant
happen!
So if relativism is true, then it would imply that its false. [And, of course, if its
false, then its just plain false.] So, relativism is just plain false!
The only way for the relativist to avoid this is to intend relativism merely as a
matter of their own opinion, applicable only to their own personal opinions. But
then the relativists statement Relativism is true is no more justified than
its denial, and has no relevance to anyone else with the opposite opinion.
Lets explode a couple more renditions of relativism. How about All opinions
are equal? If it were true that all opinions are equal, then the opinion that All
opinions are not equal would be equally true. But if All opinions are not
equal is true, then All opinions are equal must be false. So All opinions are
equal must be false.
How about Everything is just subjective; nothing is objective? Well, if its
true that Everything is just subjective; nothing is objective, then there must be
at least one state of affairs that is objective and not just subjective, namely, that
everything is just subjective and not objective. But then its false that
Everything is just subjective; nothing is objective. The truth of Everything is
just subjective; nothing is objective, would entail its own falsity. So
Everything is just subjective; nothing is objective must be false. [Oooooo, that
blew up real well!]
Ill leave it to you to demolish the remaining versions of relativism on our list.
Have fun showing how these wise-sounding expressions self-refute the instant
theyre proposed:
Everything is just a matter of opinion.
Its all just a matter of taste.
Things are only true for an individual or a culture.
There are no facts.
There are no truths.
Nothing is absolute.
12
So some claims are not just subjective; they are not mere matters of opinion.
Some claims are objectiveeither objectively true or objectively false. The
problem is to figure out which claims are merely subjective and which claims are
objective.
Indeed, some statements are merely subjective and are thereby merely true for
the individuals expressing them; for example, if I were to say Broccoli tastes
pretty good, and you say Broccoli tastes horrid! Here, there is no objective
fact of the matter regarding broccolis good-tastingness for us to be right or
wrong about. I am basically saying I like the way broccoli tastes, while you
are saying I dont like the way broccoli tastes. In fact, we are not even
disagreeing, since there is nothing for us to disagree about.
On the other hand, if I were to claim Broccoli has mass, while you claimed
Broccoli does not have mass, we would have a disagreement, because there is
something for us to disagree about, namely, it is an objective matter as to
whether or not broccoli has massit either does or it doesnt, and in fact, it does.
Note that whether a statement is merely a subjective matter of taste or an
objective matter is itself an objective matter. Its not just a matter of taste
whether or not something is a mere matter of taste. So, the fact that a claim is
merely subjective (or not) is itself objective. And in light of that, the claim that
an issue is merely subjective, unless its obviously true, will need an argument
to support it. The failure to provide such an argument will mean that the claim
remains a mere assertion, leaving its audience with no reason to think its true.
How do you argue that a matter is subjectivea mere matter of taste? It all
hinges on the following question: Does it make sense to think there is an expert
on the matter? Take our case of two people arguing about the good-tastingness
of broccoli; would it make any sense to think that an expert on the goodtastingness of food could be brought in to shed some light on the issue and help
decide whos right? No! Its just silly to think someones going to get their
Ph.D. in the good-tastingness of food. Theres no such thing as an expert on this,
and its silly to think there could be one. Either you like the taste of something or
you dontend of matterall tastes or opinions on this are equal. Whereas, if
these two were arguing about whether broccoli has mass, it would make
perfectly good sense to seek an experts opinionthe chemists or the physicists
opinion. Or, if the disagreement were about the goodness of broccoli in terms of
its nutritional value or its freshness, one could consult a nutritionist in the first
case and a gardener or grocer in the second.
So now we have a sort of litmus test to use to argue that a matter is either
subjective or objective. Once we have reason to believe that were facing an
objective matter, we then have the task of figuring out what we should believe
and what we should not believe about it. And thats the function of arguments
13
to help us believe the true claims and refrain from believing the false ones,
since now we surely know that not all assertions are equal.
Something to think about: But isnt it true that everyone has a

right to their own opinion?
Discussion: Well, yes. Repeat, no!
In what sense is it true that everyone has the right to their own
opinion? In the sense that people have the right that others not
interfere with their opinions by means of assault or coercion.
But, in what sense is it false that everyone has a right to their
own opinion? In the sense that everyones opinion is equally
correct or equally justified.
One may have the right to believe whatever one pleases, but this
does not mean that one ought to believe whatever one pleases.
Weve got the right to believe some pretty stupid things; but
thats not to say we ought to.
Something else to think about: If relativism is as lame as

weve demonstrated, why do so many people think its true?
Discussion: There is a bundle of reasons why people have been
attracted to relativism. Here are a few that strike me as primary.
One popular reason for thinking relativism is true, is the
mistaken view that the only alternative to it is intolerant
dogmatism, in which someone simply insists that their beliefs
and values are objectively true, but for no good reasons. In fact,
the dogmatist would maintain their beliefs and values no matter
what reasons they were provided for thinking they are incorrect.
This manner of maintaining ones views is contrary to critical
thinking, but no more so than relativism is. Critical thinking
maintains that, with objective matters, some opinions are better
than othersnamely, the ones with better arguments in their
favor, no matter who presents those arguments.
Another popular reason for thinking relativism is true is the
mistaken belief that relativism is a means to tolerance of others
opinions. Indeed, the dogmatist is intolerant of all those who
happen to disagree with them. But then the relativist can be
14
just as intolerant, if its their personal opinion that they need

not be tolerant of others. Remember: If all opinions are equal,
then my opinion that I need not tolerate the opinions of others is
as good as the opinion that one ought to tolerate the opinions of
others.
Another reason relativism is so popular is that people often
mistake the diversity of opinion for the equality of opinion.
They think that if there is no way to tell which of the
multiplicity of opinions is true or more reasonable to believe,
then there is no objective truth to have a correct opinion about.
But that cant be right! Just because some truths are hard if not
impossible to discover, that doesnt mean they dont exist. Its
an objective matter as to how many acorns there are in the U.S.
at the moment you read this sentence, despite its being
impossible to know which opinion is right regarding that
number. Its impossible to know, but its not subjective. Thats
whats so ironic about the relativists claim that its pompous to
think there is a real world beyond us to discover. The realist
view is humility itself, compared to the relativists claim that the
world is just our opinion.
A lot of people talk a good relativist line, but when push comes
to shove, they really dont seriously adopt it. Relativists usually
trot out the all opinions are equal slogans when their own
beliefs and values are faring poorly under critical review, viz.,
when they are losing the argument. When this happens,
relativism is more of a dodge to change the subject than a serious
response to criticism. The real test of sincerity would be, for
example, when the relativist feels the dreaded lump while
showering and actually maintains their view that all opinions
are equalwhether they come from an oncologist or an
astrologist. Ive yet to see a relativist pass this test!
A few other speech actsdescriptions and explanations

We just studied what an argument is and how it differs from a mere assertion.
There are, however, a lot of other noteworthy things one can do with words. For
example, one can question, command, warn, or express ones emotion:
Is the door shut?
Shut the door.
Look out for the door!
That @#%*ing door!
15
Something to notice about these cases is that not a single one of them is a
statement. A statement is a declaration, a claim that is either true or false. And
none of the listed sentences is either true or false. Try them. For example, the
question Is the door shut? gets Yes or No, but not True or False, as a
sensible reply.
Pop quiz: Think about the phrase I just used Try them. Is
it a statement?
Arguments and assertions are made of statements, as are two other important
speech acts you should be able to distinguishdescriptions and explanations.
A description is simply a statement or claim that is obviously true in the context
of its use. Heres an example.
Watson and Crick were helped greatly by actually trying to
construct a physical model of DNA. This was a model in the
ordinary sense in which model airplanes and dollhouses are
models. They are all scale models. The big difference between
Watson and Cricks model and more familiar scale models is the
extreme nature of the scale, which, in the case of the DNA
model, was roughly a billion to one. That is, an inch in the
model represented roughly one one-billionth of an inch in an
actual DNA molecule. [Ronald Giere, Understanding Scientific
Reasoning]
When someone is using a statement as a description, no derivative point is
made by means of the statement, i.e., no conclusion is being drawn from it, and
no reason is given to believe the statement, since none is neededwe are simply
willing to take the authors word for it. We are not the least bit inclined to ask,
Now why ought I to believe that?
Pop quiz: If the author were to make a derivative point by

means of their statement or give a reason to back their statement,
the description would become a what? If the authors statement
were not obviously trueif it were controversial it would be a
what?
16
An explanation is a set of conditions, events, or states of affairs that (one claims)

makes it such that another set of conditions, events, or states of affairs is the case.
Heres an example.
With the advent of the Information Age, technological advances
have changed how we process information. With the World
Wide Web becoming an integral part of our lives, we expect to
have substantial amounts of information readily available at
convenient times. As a consequence, one of the major external
changes for colleges and universities is the arrival of waves of
students who are not only computer literate, but actually
techophilic. [Cini & Vilic, Online Teaching: Moving from Risk
to Challenge, in Syllabus]
Note how one event, the change in students computer skills, is claimed to be a
consequence of another set of events, namely, technological advances such as the
World Wide Web.
You probably noticed that our account of an explanation includes a little part in
parentheses. If we were to remove those two little words, we would get what
could be called the ontological or real explanation. The real explanation is the
set of conditions, events, or states of affairs that actually makes it such that
something else is the case.
When we claim that a certain condition, event, or state of affairs is the
explanation, we have proposed what could be called a theoretical explanation.
This is sometimes called a linguistic explanation, because it is made in the form
of a statementa statement describing the condition, event, or state of affairs
that one believes makes it such that something else is the case.
The fact that we are all-too-often wrong in our beliefs or theories or claims about
explanations, illustrates the need for this distinction between the explanations we
propose and the real explanations. Our theoretical explanations are true when
they accurately represent the real ones.
Quite often there are multiple or partial explanations. The universe is way
complicatedoften there are many factors making up the explanation of a single
event, and often a single factor partially affects many events. This is what makes
discovering explanations so hard!
Just as arguments have two basic parts, premises and conclusions, so too
explanations have two basic partsthe explanandum and the explanans. Just
think of it this way: What is in need of explaining is the explanandumits the
dumb part. And then what does the explaining must be the explanans. So, if I
17
tell you that my shoulder hurts because I have a bone spur, the pain in my
shoulder is the explanandum and the bone spur is the explanans.
The distinction between explanations and descriptions gets a bit blurry
sometimes. One reason for this is that when one gives an explanation, one is, in
a sense, describing the conditions, events, or states of affairs that bring about
another set of conditions, events, or states of affairs. But in such a context, ones
statement is functioning ultimately as an explanation and not as a mere
description.
Another source of confusion is that we often speak of people explaining how
things are donefor example, when someone explains how to get to the nearest
gas station or how to tie a shoe. But these are better characterized as descriptions
descriptions of how to get to a certain location or state of affairs. The same
goes for explaining how things lookthese are just descriptions. [Our language
is pretty loose with the use of the word explanation, unfortunately.]
Different types of explanations

Explanations state why something iswhat makes it the case or what brought
it about or what caused it. But not all explanations are causal, stating what
event made something occur. Some explanations state what makes something
what it is. These are essentialist explanations. The difference between causal
explanations and essentialist explanations can be clarified by looking at the
following examples:
1. The liquid is water, because its molecular make-up is
H2Otwo hydrogen atoms and one oxygen atom,
held together by hydrogen bonds.
2. The water in your tailpipe was produced during the
process of oxidation, by the combustion of gasoline.
The first example states what conditions make the water what it is, and the
second example states what event brought together an instance of those conditions. If you havent already noticed, definitions and essentialist explanations are
the same. With this in mind, we could give both an essentialist and a causal
explanation for why someone is a bachelorhes an unmarried male, and hes a
jerk that no one can stomach.
Another kind of explanation is a functional explanation. A functional explanation is a things job or roleits functionthe doing of which explains that
things continued presence or action. For example, the function of the heart is to
circulate bloodthat explains presence in the body and its pumping action.
18
Failing to get that result means the end of the heart and its action. [Note: The
hearts pumping action causally explains a thumping sound too, but that is not its
function, since the heart was not selected for because it produces such a sound,
and its failing to produce such a sound (e.g., for the deaf) would not mean the
end of its pumping action.]
Pop quiz: Answer: Our appendix and disco.

Question: Name two things that have lost their functional
explanations, but linger in withered forms to our occasional
detriment.
Distinguishing between explanations and arguments

Its a bit of a job to distinguish between explanations and arguments. Maybe this
will help:
Explanations state why something occurswhat makes it the
case or what brought it about or caused it.
Arguments give reasons for believing that something is the
casethey state why one ought to believe a statement.
Usually, when someone is giving an explanation, what is explained is assumed
true; and one is just stating what made it so. Usually, when someone is giving
an argument, on the other hand, what is argued for is not assumed true; rather
one is supplying reasons to try to get the audience to believe its true. This is
illustrated in the following examples.
1) Ive got a sore shoulder and am unable to serve hard anymore
because I have a bone spur.
2) Ive got a bone spur on one of my shoulder bones because my
shoulder is sore right here, and Im unable to serve hard at all
anymore, and I went to a sports specialist who looked at an xray and even showed me an extra little growth on one of the
bones thats digging into my muscle.
In 1), its taken for granted that my shoulder is sore and that I cant serve hard;
Im simply explaining whythe bone spur is the cause. In 2), Im arguing for,
not assuming, the claim that its a bone spur, as opposed to arthritis, gout, or
cosmic smite, that is causing my bum shoulder and wimpy serve.
19
Arguments and explanations are especially hard to distinguish, because they

share many of the same indicator terms; for example, so, since, because,
thus, therefore, and hence, to name only a few.
Moreover, many statements can function equally well in either role; an author
can use what theyre saying as an argument, or as an explanation, or as both.
Lets look at some examples.
Some explanations that also function as arguments

Kay is not pregnant, because Kay is taking birth control pills that
are extremely reliable.
The rock hit the ground at 25.3 ft./sec., because I dropped it from
10 ft. high, and it accelerated at 32 ft./sec.2.
In the first example, we are told of the set of events that is preventing Kays
pregnancy, and in the second example, we are told of the set of events causing
the rock to hit the ground at a certain velocity. But these statements of conditions
can function equally well as reasons to believe, in the first case, that Kay is not
pregnant and, in the second case, that the rock hit the ground at the stated
velocity.
But some good explanations could never function as arguments

I flipped 10 heads in a row, because the coin is fair and every
now and then a fair coin produces a lopsided distribution.
Indeed, this is the only explanation of (ever) getting the improbable result of
flipping 10 heads in a rowor 10 tails, for that matter. But this would be a lousy
reason to believe that one will get 10 heads in a row. Remember, an argument is
supposed to make its conclusion at least probable. But this explanation actually
makes it improbable that one would get 10 heads in a row. Only an idiot would
use this statement as an argument; and, since we are not dealing with idiots here,
their use of a passage isnt relevant.
And some arguments could never function as explanations

The pole is X ft. high, because the cable is Z ft. long and touches
the ground Y ft. from the base of the pole.
20
Z
X
This would be a great argument for the poles being X ft. tall. It would just be a
basic application of the Pythagorean Theorem. But weird things happen when
you try to warp this into service as an explanation. You mean to tell me that the
cables being Z ft. long and touching the ground Y ft. from the base of the pole
made the pole X ft. tall?! Say Hallelujah, praise the cable!!
No, the explanation of the poles being X ft. tall is going to be something else
entirely: maybe its because Groundskeeper Willie cut it off at that length, or
that its an Acme flagpole, and Acme flagpoles only come in that size.
Whatever, the explanation is going to be something like this.
And take another look at this example:
K is not pregnant, because K is taking birth control pills that are
extremely reliable.
This was used as both an argument and an explanation earlier. But now lets say
the sentence is about meKevin. How well could it work as an explanation
now? It really couldnt. Even assuming that Im consistently taking my
extremely reliable birth control pills every morning, those pills just could not
work as a prevention of my pregnancy. My lack of the proper wetware, so to
speak, is a better candidate for that.
However, could the fact that Im consistently taking my birth control pills every
morning function as an argument for believing that Im not pregnant? Yes, it
couldjust as well as it could function as a reason for believing that Kay isnt
pregnant. If Kay and I are taking the same birth control pills and they are 99.9%
reliable, then they make it 99.9% probable that each of us isnt pregnant. There
21
is, of course, a much better reason to believe that Im not pregnant, but that
would be a different argument.
I hope these casescases in which an explanation could not work as an
argument and an argument could not work as an explanationhelp you to keep
these two different speech acts separate. Arguments and explanations function
differently, despite the fact that they look so much alike and despite the fact that
the same passage can often function as either.
Heres one last example, to help you distinguish between an argument and an
explanationthe gas gauge. The position of the gauges needle, e.g., its being
on E, works well as a reason or argument for thinking that the tank is empty.
The needles being on E does not, however, explain the tanks being empty.
Just the reversethe tanks being empty explains why the needles on E.
Its time to practice what youve learned

You now know what an argument is and how it differs from other speech acts,
such as an explanation, a mere description, and a mere assertion. Its time to put
this knowledge to work, and practice identifying and distinguishing these speech
acts in the field.
The analogy to biology is helpful here. The biologist first studies the taxonomy
of the speciese.g., what makes a robin a robin. Only after the biologist is
familiar with this taxonomy would it be possible to do any field biologyif the
biologist didnt know what a robin is they could hardly identify them in the field;
they wouldnt know a robin from a rabbit. But just knowing the taxonomy of the
species doesnt automatically make one a good field biologist. Bird
identification takes practice. And so does argument identification. Knowing
what an argument is is the first step; being able to accurately spot them in the
field of daily discourse is quite another knack.
You can best practice your skills at argument identification by using the Critical
Thinking software. The module titled Argument Identification is dedicated to
exactly this.
This critical reading skill of argument identification is foremost a matter of being
an active readerof asking questions and looking for indicators as you read.
The first and most important question to ask is, What is the ultimate point or
claim in the passage? Once youve located that, you then begin asking more
specific questions and looking for more specific telltale features. The key is to
keep prodding the passage until it reveals what it is. Heres what I mean:
22
After youve located the ultimate point in the passage:
Mere AssertionAsk for reasons to believe the statement.

If they are needed, but none are given, then its a mere
assertion.
ArgumentIf reasons are given for believing the

statement, then its an argument.
Mere DescriptionIf no derivative point is made by

means of the statement, and no reasons are needed to
believe the statement, then its a mere description.
ExplanationIf the statement discusses the conditions,

events, or states of affairs that made it such that another set
of conditions, events, or states of affairs is the case, then its
an explanation.
As you work on this module, remember to be an active reader, constantly

applying this list of considerations.
Important reminder: Indicator terms are so helpful that they

can actually enable you to identify the speech act without even
knowing what is being discussed. Identify the following:
1. Since blah blah blah, blah blah blah. And so it
follows that blah blah blah.
2. What made it so blah blah blah was blah blah blah.
3. No matter what people think, blah blah blah is clearly
the wrong procedure to adopt.
But isnt it possible to use a statement every which way?

Were unique. We are the only creatures we know of that are real language
users. We use symbols, for example, linguistic squiggles and sounds, to mean
certain things as opposed to othersto express one thought as opposed to
another. All other creatures so far examined merely respond in quite limited
ways to a quite limited set of stimulus conditions. Sorry, but those chimps are
still not signing robustly enough to qualify as using sign language. You and I, on
23
the other hand [no pun intended], can use the sentence Oh, thats real good! at
one time to compliment someone and at another time to level a sarcastic insult.
We can even use it to mean I have the State secrets on me, lets meet at the
usual rendezvous for their delivery. We can use the phrase to mean an infinite
number of different things.
Does this mean that words mean whatever we take them to mean? Well, Yes,
repeat No! We indeed are the ones in charge of assigning meanings to words
and sentences; we can use a sentence to mean one thing as opposed to another.
But, if we ever want to use language successfully as a form of communication,
we cant be glib and anarchistic about it. If we wish to communicate our
thoughts by means of linguistic symbols, we need our audience to be familiar
with the symbols we use and to know that we mean one thing as opposed to
another by those symbols, or else our audience simply will not understand what
were expressing.
So, its possible to use a sentence in an infinite number of ways to mean an
infinite number of things; just dont expect to get your point across as you
exercise this possibility.
You can use a screwdriver as a screwdriver or as a can opener or as a
paperweight, but dont try to make it function as a wedding ring and expect to be
taken seriously.
So, as you read and listen to others and try to figure out what they are saying and
which speech acts they are performing, dont so much wonder what they could
mean; rather, ask what in all probability do they mean, given that they
arent insane and wish to communicate with the average audience? This
is exactly what to keep in mind as you practice Argument Identification, either
in the CT software or in daily discourse.
Oh, and another thing [or two]

Even though some days youd swear that most people cant do two things at
once, the truth is they do, especially with respect to speech acts. People can pack
a lot into a paragraph. And this makes it all the harder to identify which speech
acts theyre performing. So, be on the lookout for multiple answers as you
practice Argument Identification.
When trying to identify the speech act, always identify its function within its
context. An example will illustrate what I mean.
Any diet poses some problems. Heres why. If the diet doesnt
work, well, thats a problem. If the diet does work, then the
24
dieters metabolism is altered. An altered metabolism as a

result of dieting means a person will need less food. Needing
less food, the person will gain weight more easily. Therefore,
after successful dieting, a person will gain weight more easily.
[Adapted from Trudy Goviers A Practical Study of Argument]
This is an argument, for the conclusion that any diet poses some problems.
Outside of this context, the two statements in boldface, concerning states of
altered metabolism and needing less food, would be explanations. But within the
context of this passage, they are not explanations, but rather premises of an
argument.
A final suggestion, as you work on Argument Identification [and any of the
other modules too, for that matter]: If your answer is mistaken, dont go on to
the next problem until you know exactly why your answer is wrong and why the
correct answer is correct. You dont learn anything just by clicking the mouse.
If that were the case, all those who smoked cigarettes would be brilliant by now,
for all the finger flicking they do!
25
Anatomy of an Argument
3. ANATOMY OF AN ARGUMENT
Ive been milking the analogy between critical thinking and field biology pretty
hard: We first learn the features that distinguish the species of speech acts and
then learn to identify them in the field. Well, Im going to milk this analogy a
bit more.
To fully understand a particular organism, what makes it tick and all, you have to
dissect it. Youve got to look inside, to see its parts and how theyre connected.
Likewise with arguments. To fully understand how they are functioning so as to
provide reasons for believing their conclusions, you have to dissect them into
their component premises and conclusions. This is also the only way to tell the
healthy, well-functioning members of the species from the sickly ones. And that,
remember, is our projectto study what makes a good argument a good
argument.
Things to keep in mind when identifying the anatomy of an argument:

1. Make sure its an argument!
Use the skills you honed in Argument Identification to do this. Its a
bit tough to identify the parts and structure of an argument, in a
description, for example. Sort of like looking for lungs in a fish.
2. Identify its premises and conclusions
Arguments often have multiple premises and even multiple conclusions.
The best way to locate the conclusion is to ask the passage, Whats
your overall, ultimate point? Find the passages answer to that question
and youve found its conclusion. Now ask the passage, And why
should I think thats true? Its answer to that question will be its reason
for thinking the conclusion is truethats a premise.
3. Identify any subarguments
A conclusion can also be used as a premise in support of another
conclusion. Or, an alternative way of describing this is to say that a
premise can be a conclusion of an argument preceding it. Heres how to
tell when this is happening: As just discussed above, locate the main
conclusion by asking, Whats your overall or ultimate point? And then
locate the premises in support of that conclusion by asking, And why
should I think thats true? But now, ask that question again, this time
of the premise you just found: And why should I think thats true?
26
The answer to this question is a premise given as a reason to believe

the other premise. See? And then, you might even ask, And why
should I think THATS true? If the passage answers that question too,
youve found a premise in support of a premise thats in support of a
premise thats given in support of the conclusion. Wow! Youve got an
argument in an argument in an argument. This is not unusual.
Arguments are often like onions in this waywith layers that support
other layers.
4. Ignore fluff
For the purposes of detailing all and only the parts of an argument, you
must ignore all the extraneous side comments. For example, ignore
background information and hedging. An example will make this
clearer. Whats the anatomy of the argument in this passage?
The issue of abortion has perplexed mankind for
hundreds of years, and still remains an issue of debate
for all who take moral problems seriously. Many people
have differing opinions on the morality of abortion, but I
think that it is indeed morally permissible in early stages
of the pregnancy, because at that stage, the fetus lacks
even sentience, a necessary condition for having any
moral status whatsoever.
After all the fluff has been removed, the structure of the argument is
merely:
P1. The fetus lacks sentience in the early stages of pregnancy.
P2. Sentience is a necessary condition for having any moral status.
C. Abortion is morally permissible in the early stages of pregnancy.
Distilling an argument in this way is called putting the argument in
standard formit doesnt make for stimulating prose, but its the
clearest way of presenting an argument.
5. Note any implicit [unstated] premises or conclusions
Most arguments in daily discourse have implicit premises or conclusions.
This is usually out of courtesy to the audience, because if all premises
and conclusions were explicitly stated, arguments would become so
long-winded, by virtue of including so many obvious statements, that the
audience would simply die of boredom or leave in a huff for being talked
down to like a child. So, to speed things along and to avoid insulting the
27
intelligence of the audience, the already-granted premises and

subconclusions are left unstated.
Lets look at the previous example to illustrate this. Heres the
same argument in standard form, including all its implicit
parts, indicated with stars and square brackets:
P1. The fetus lacks sentience in the early stages of pregnancy.
P2. Sentience is a necessary condition for having any moral status.
[C*. The fetus in the early stages of pregnancy lacks moral status.]
[P*. If something lacks moral status, it is morally permissible to kill it.]
[P**. Abortion is a form of killing.]
C. Abortion is morally permissible in the early stages of pregnancy.
See? Half the argument was behind the scenes.
6. Arguments come in any order
Few arguments have a perfect premise, premise; therefore, conclusion
form. Many arguments have a conclusion, since premise and premise
form. Some even have the format of premise; therefore, conclusion,
since premise. Its not the particular order of the statements that
determines their status as premises or conclusion, rather its how they are
functioning in their context.
Pop quiz: Whats the format of Mills argument, repeated here?

The peculiar evil of silencing the expression of an
opinion is that it is robbing the human race; posterity as
well as the existing generation; those who dissent from
the opinion still more than those who hold it. If the
opinion is right, they are deprived of the opportunity of
exchanging error for truth. If wrong, they lose, what is
almost as great a benefit, the clearer perception and
livelier impression of truth, produced by its collision
with error.
Answer: Mill states his conclusion first, and then gives us two
premises in support of it.
Here is a way of representing the anatomy of another argumentin this case, a

rather involved argument. Note how subconclusion C1 becomes a premise [P3]
for another conclusion [C2].
28
P1*
P1
P2*
P2
C1/P3
P4
C2/P5
C3
Pop quiz: What will be the diagrams answers to the following?

Whats your overall point?
Why should I believe thats true?
Why should I believe thats true?
Why should I believe THATS true?
For those of you who are more imagistically inclined, with respect to how you
think about things, heres another way to map out the taxonomy of an argument.
I happen to like this way of representing arguments, because I tend to think better
using images. Others find it rather unhelpfuljust a bunch of goofy circles and
29
arrows. The bottom line is this: Use whichever means works best for you to
understand the anatomy and dynamics of arguments.
When are two premises and a conclusion not an argument?

When they are two arguments. Sometimes the premises of an argument work
together to form a single reason to believe the conclusion, and sometimes they
work as independent reasons to believe the conclusion. This difference in the
dynamics of arguments is captured as follows by our diagramming.
Pop quiz: Which diagram better represents the anatomy of

Mills argument?

Youre all ready to practice dissecting arguments and identifying their parts,
using the CT software. In the module called Anatomy of an Argument, youre
given passages that have their sentences or clauses numbered. The project is to
30
indicate how each clause is functioningas a premise, a conclusion, or both,

viz., as a [sub]conclusion used also as a premise to support another conclusion.
Use scratch paper!

My advice concerning almost every module in the CT software is to use scratch
paper. In fact, my policy is that you should have a piece of scratch paper and a
pencil no matter what youre doingI dont care if youre having sex; I still
think youre allowed scratch paper and a pencil. They are standard equipment
for critical thinking, and heres why.
We evolved to be pretty smart, but we didnt evolve to have particularly good
memory capacities, especially working memory capacity. We are smart
enough, however, to use tools to overcome this shortfall. I consider the pencil
and paper to be a natural extension of our brains; and so I always carry a pen and
a scrap of paper [maybe its the program handed out at the presentation, for
example, or a credit card receipt from my wallet or whatever]. And thats why,
for me anyway, trying to practice CTs critical thinking skills without the aid of
scratch paper would be like trying to get by with only half a brain [or only a
quarter of one, for us guysif the latest research is to be believed!].
31
What Makes an Argument Good?
4. WHAT MAKES AN ARGUMENT GOOD?

We know what an argument is and how to identify it and all of its parts. Its time
to move to the qualitative issues and study what makes an argument good and
what makes an argument bad.
Lets introduce a slightly technical term at this point, because its not always
clear what someone is saying when they call an argument good or bad
good for what and bad for what? We will be interested only in the quality of
an argument in terms of its cogency.
To say that an argument is cogent is to say that the premises of the argument
give rational support to the conclusionthe premises make the conclusion
rational to believe.
And what does that mean?
Well, think of an argument as a kind of input-output device: You feed it
premises as its input and it gives you conclusions as its output. If the device is
working and you hand it premises that are rational to believe, it will hand you
conclusions that are rational to believe. But, as the old saying goes, Garbage in;
garbage out. Even if the argument is working and you hand it unbelievable
premises, theres no telling what it will poop out for a conclusion.
So essentially, cogency is a matter of two conditions:
The premises must be acceptablereasonable to believe.
The premises must be properly connected to the

conclusion in such a way that the conclusion is probably
true, given that the premises are true.
An argument that meets both of these requirements becomes a conduit of

acceptability. It has acceptable premises being poured into its hopper, and it is
constructed tightly enough so that it conveys this acceptability to its conclusions
well enough to make them probably truebelievable premises in; believable
conclusions out.
You can have one without the other

To get a feel for these two cogency conditions and for how they differ, lets look
at a couple examples in which one condition is met but the other is flunked.
32
All cats are mammals.

All dogs are mammals too.
Washington D.C. is the U.S. capital
Huh?! Yes, thats exactly the right reaction to have to this argument. Which
condition of cogency did this argument violate?
Thats right, the conduit condition. If your huh?meter goes into the red, thats
a moderately good indication that the conduit condition has been flunked and you
have a noncogent argument in front of you.
The acceptability condition, on the other hand, was met quite nicely in this case:
Its certainly reasonable to believe that all cats and all dogs are mammals
youve known that since your grade school biology days. But those biological
facts are totally irrelevant to whether or not D.C. is our capital. If someone were
unsure what the U.S. capital was, maybe because they thought the largest city
might be the capital and so New York is, telling them that dogs and cats are
mammals would not give them any reason to change their mind.
Furthermore, lets change the conclusion to read: Winona, Minnesota is the U.S.
capital. This conclusion is patently false, but the premises about all cats and dogs
being mammals are as much a reason to believe this claim about Winona as they
are to believe the claim about D.C.
So flunking this condition means that the argument is not a conduit of
acceptabilityacceptability in, but quite possibly garbage out.
Now take a look at another little argument.
All cats are lovable.
All lovable creatures are furry.
All cats are furry.
Thats right; the acceptability condition is violated this time. Both of these
premises are flamingly false. Garfield immediately comes to mind to falsify the
first premise; and your mother hopefully came to mind to falsify the second.
And yet note that the argument is constructed in such a way that if it were
believable that all cats were lovable and if it were also believable that all
lovable things were furry, that would make it equally believable that all cats
are furry. The connection between the premises and the conclusion of this
argument is so tight that if there were any believability to these premises at all, it
would all be transferred to the conclusion. So this argument passes the conduit
condition fine; it just falls flat on the acceptability condition.
33
Lets begin a closer study of these two cogency conditions by looking at one very
important way of meeting the conduit condition.
Logical entailment
The strongest way an argument can meet the conduit requirement for cogency is
called logical entailment. There are many names for this relation, so let me just
list the most popular ones:
The premises logically entail the conclusion.
The premises logically imply the conclusion.
The premises deductively imply the conclusion.
The premises deductively entail the conclusion.
The conclusion follows logically from the premises.
The conclusion follows deductively from the premises.
The conclusion follows validly from the premises.
There are two ways of describing the tight connection between premises and
conclusion in logical entailment:
Given the truth of the premises, the truth of the conclusion is

guaranteed.
Its impossible for the conclusion to be false while the

premises are true.
For now, lets focus on this second way of capturing the strong relation of logical
entailment.
Heavy-duty impossibility
When premises logically imply their conclusion, its logically impossible for the
conclusion to be false while the premises are true. Its not just technically
impossible for the conclusion to be false while the premises are true. And its not
just physically impossible for the conclusion to be false while the premises are
true. Its logically impossible for the conclusion to be false while the premises
are true.
Heres what I mean. You probably thought that when something is impossible,
that just means it cant happen. Well, not quite; because impossibility comes in
different strengths. For example, its impossible for a laptop computer to process
at 5 gigaflops [5 billion floating-point operations per second]. But this is only a
technical impossibility. The desktop Mac G4 is there already, and, as soon as a
34
few more technical limitations are surmounted, the feat will be accomplished by
a laptop tooand probably not long after Im done keying this sentence in.
Technical impossibility is the weakest; its only a function of a lack of enabling
conditions.
Its quite a different matter, however, that its impossible for my laptop computer
to hover six feet off the ground unassisted. This is not just due to a limitation of
conditions; its not just due to technical shortcomings. No, if a laptop were to do
this, it would be breaking a law of nature. A law of physics called gravitation
makes it physically impossible for laptops to do this. If the laws of physics were
differentfor example, if the law of gravitation were no longer the caseit
would be possible; but not until then.
But there is even a s t r o n g e r form of impossibility; and thats logical
impossibility. Its the impossibility of, e.g., ever having a square circle or a
married bachelor. These two things could just never happen, even if you had
complete run of the laws of nature to form them however you wanted to make
things. For example, we could make a much brighter universe if we played
God and changed the laws of optics from inverse-squared laws to simple
inverse-proportionality laws. [Or we could do a little mood-lighting and
change the laws of optics to an inverse-cubed relation of brightness to distance.]
But no matter how we might tinker with the laws of physics in this manner [and
of course we cant!], we could never, ever produce a squared circle or a married
bachelor. They are logical impossibilities.
A square circle and a married bachelor are contradictions. A bachelor is an
unmarried male. And so a married bachelor would have to be married and
unmarried at the same time. This cant be, no matter what tinkering with the
laws of nature might permit. God couldnt even create a married bachelor, and
Gods supposed to be all powerful. My point here is that logical impossibility is
the strongest of all impossibilitiesits so impossible that even God couldnt do
it. Wow; thats pretty impossible!
Logical entailment and validity

Now we can really appreciate how tight the connection is between the premises
and the conclusion when those premises logically or deductively imply their
conclusions.
When it is logically impossible for an arguments premises to be true while
its conclusion is false, the argument is said to be deductively valid.
The word valid has many uses and meanings. You hear people say, Your
point is a valid one, by which they mean that the statement made is true or
35
reasonable. You hear psychologists call a test valid, by which they mean that the
test actually measures the desired phenomenon.
But we will be using valid in the more technical sense stated in bold type
above; this is what validity is from the point of view of logic. This will cut down
on potential confusions as we talk about the cogency requirements of arguments,
I promise.
Thats not to say it is true!

When an argument is valid, its logically impossible for the premises to be true
and the conclusion false; given that the premises are true, the conclusion must be
true, on pains of resulting in a contradiction. But this does not mean that the
premise must in fact be true. The truth or acceptability of the premises is one
consideration; the validity of the argument is quite another. Validity is neutral
on whether or not the premises are in fact true. Validity is a relation between
the set of premises and the conclusion; a relation such that if the premises are all
true, then the conclusion must be true too. But thats a big if.
But is the argument sound?

Here, once again, are the two conditions of cogency:
Cogency:
1. The premises must be acceptablereasonable to
believe.
2. The premises must be properly connected to the
conclusion in such a way that the conclusion is
probably true, given that the premises are true.
There is another, more traditional, way of discussing the quality of arguments,
and you should be at least aware of this method of assessing arguments. Its
called soundness. The requirements for a sound argument are twofold and they
sort of correspond to our conditions for cogency:
Soundness:
1. The premises must be true.
2. The argument must be valid.
36
While the conditions of cogency make an argument a probable conduit of

acceptability, the conditions of soundness make an argument a perfect conduit
of truth. [See why?]
Something to think about: Can there be a cogent argument

thats not sound?
Discussion: Sure; in fact this can happen in two different ways:
An argument can have premises that we have every reason to
believe are true, but yet those premises can be in fact false.
Take, for example, fourteenth-century conclusions drawn from
the popular and reasonable premise that the Earth was flat. If
this premise had been true, it would have logically implied
conclusions about the Earth having an edge. So, as cogent as
those arguments were at that time, they were not sound.
An argument can also have premises that are true and yet do not
logically imply their conclusion. In fact, many arguments are
cogent and have true premises, but are not valid. We will
discuss this at length in a bit; for now, let me just say that
arguments in the sciences fit this description. The most
scientific arguments can obtain is cogency; they in principle
cannot obtain validity; so they in principle can never be sound.
Ones premises in science can make ones conclusions at most
very probably true, but they can never guarantee their truth.
Something else to think about: Can you have a sound

argument thats not cogent?
Discussion: Sure. You could have a valid argument with true
premises and yet have no reason whatsoever for believing that
those premises are true; you just made a lucky guess. So these
premises are true, but totally unacceptableyou are totally
unjustified in believing them. Hence, a sound but noncogent
argument.
The ARG conditions for cogency

To facilitate our investigation into what makes an argument cogent, we should
breakdown the conduit condition into two related but separable conditionsthe
relevance condition and the groundedness condition. Here, then, is the final
37
list of cogency conditions we will be studying and implementing for the rest of
this project.
The ARG conditions for cogency

Acceptability of premises:
One has more reason to believe

the premises than to not believe
them.
Relevance of premises:
The premises provide some

reason to believe the conclusion.
Premises Ground the conclusion:
Premises provide adequate

evidence or reason to believe
the conclusion, making the
conclusion probably true, given
the truth of the premises.
I first ran across this way of unpacking the cogency conditions in Trudy Goviers
A Practical Study of Argument. Its a wonderful text (the second best I know of
;-), with a helpful way of remembering the ingredients for a cogent argument:
ARG is the acronym for our three conditionsacceptability, relevance, and
groundednessand its also the first three letters of the word argument. [Admittedly, groundedness is a bit clumsy, so I considered calling it the support
condition, until it came time to do the acronym.]
We will soon take a close look at the acceptability condition, in Chapter 6,
studying what having acceptable premises is all about. We will also examine
some of the most popular ways of meeting the acceptability condition and some
of the most popular ways of failing to meet it.
In a similar fashion, the new relevance condition will be studied in depth later,
in Chapter 7. But for now, a basic understanding of the relevance of premises
can be gained from the following example.
Say that I conclude that chickadees love sunflower seeds, on the basis of my
seeing a single chickadee eat a single sunflower seed. Is my reason relevant to
my universal conclusion? Well, yes; its at least some evidence. Its at least
more evidence than I had before I witnessed any chickadee at all eat a sunflower
seed. Admittedly, my reason would be a very weak one; but its at least relevant.
In order to meet the groundedness condition for cogency, however, I would
need enough evidence to adequately support my universal conclusion about
38
chickadees loving sunflower seeds. I would need evidence that would make it at
least probable that all chickadees love sunflower seeds.
Lets begin taking a more detailed view of our ARG conditions by zooming in on
this groundedness condition a bit.
Ways for premises to ground conclusions

There are fundamentally two ways for premises to ground their conclusions,
corresponding to the two basic types of arguments:
Deductive arguments
Inductive arguments
We have already briefly discussed the first way for premises to ground their
conclusions, viz., by deductively entailing those conclusions. In Chapters 8-9,
we will study deductive arguments in detail and learn a number of ways to assess
their validity.
In future chapters, Ill discuss how to assess various forms of inductive argumentsso please be on the lookout for the upgrade of this etext.
At this time, however, lets just get clear on what the difference is between the
two types of arguments, since it has been a major source of confusion for many,
many people.
Deductive vs. inductive arguments

An argument is an argument by virtue of someones using it as one. So too with
what kind of argument it is. An argument is a deductive argument by virtue of
someones using it as one. This is what allows for the possibility of unsuccessful deductive arguments [of which there are many!].
So, when someone claims that their argument is a deductive argument, they must
be claiming that its valid, viz., that it is logically impossible for the premises of
their argument to be true and its conclusion false. They are claiming that it
would be a contradiction for the premises to be true while the conclusion false.
They are claiming that if their premises are true, their conclusion must be true
too.
The following two-premise argument is a paradigm example of a deductive
argument:
39
If it rains, I will have my umbrella.

Its raining.
I have my umbrella.
Note the perfect truth and acceptability preservation: If the premises are true
and reasonable to believe, the conclusion would have to be true also and equally
reasonable to believe. The conduit of truth created by validity has no leaks! This
is even the case for the following rather strange argument.
The Moon is square.
All squares are rectangular.
The Moon is rectangular.
This argument is valid. Dont let the false first premise get you thinking
otherwise! Just keep firmly in mind what validity is and youll soon see that if
the Moon were indeed square then the Moon would have to be rectangular, given
that all squares are rectangular. Again, perfect truth preservation.
But this virtue of deductive arguments has a rather high price. Truth preservation
is obtained at the expense of the informativeness of the argument: The
conclusion of a deductive argument merely makes explicit what is already
implicit in the premises.
A rendition of one of our examples will illustrate this point.
At all times when it rains, I will have my umbrella.
So, this time, when its raining, I have my umbrella.
This particular time, referred to in the conclusion, is certainly included in all
such times referred to in the premise. So, nothing is claimed in the conclusion of
a deductive argument that isnt already contained in the premises. Deductive
arguments are, in this sense, uninformative.
This is definitely not the case with inductive arguments.
Heres a paradigm example of an inductive argument:
Animals receiving high doses of radiation
disproportionately develop cancer.
Humans unfortunate enough to receive high
doses of radiation also disproportionately
develop cancer.
High doses of radiation are carcinogenic.
40
Two features to note about this inductive argument:
The claim in the conclusion goes beyond the claims in the

premises. The premises are about past instances, and the
conclusion is about present and future cases.
The argument is not deductively valid. Its logically

possible for high doses of radiation to have caused cancer in
the past, but now (or sometime in the future) the laws of
nature alter such that they no longer cause cancer. That
would be extremely unlikely, even physically impossible,
but not logically impossible. So, the truth of the premises
does not guarantee the truth of the conclusion; even though it
does make the conclusion probable.
So inductive arguments are informative, but have only imperfect truth preservation.
As with all things in life, it seems, there is a trade-off. This time it is between
informativeness and truth preservation. Inductive arguments leak when true
premises are thrown into their hopper. At most they make their conclusions
probably true, but never true with logical certainty.
This is why science can use only inductive arguments as it tells us about the
world on the basis of experiential evidence. As it informs us about how the
world is or predicts how the world will be, it does so on the basis of recorded
observations. But the degree to which it can so inform us is the degree to which
it cannot be certain in its conclusions. If certainty is our goal, we must lower our
expectations and settle for only explicitly repeating claims we already implicitly
granted in our premises.
But thats not what I was told before!

Youve probably been told for years and years that the difference between
deductive arguments and inductive arguments is that deductive arguments go
from general premises to specific conclusions, while inductive arguments go
from specific premises to general conclusions. Even the most recent editions of
the MLA Handbook for Writers and the Prentice Hall Handbook for Writers say
that this is what distinguished the two kinds of arguments. Likewise for almost
every rhetoric and speech text on the market. But theyre wrong; its just that
simple.
41
Rather, the distinction is that deductive arguments claim validity, while

inductive arguments dontand inductive arguments are informative, while
deductive arguments arent. Its just that simple.
Im not saying that, for example, your beloved Freshman Comp instructor lied to
you; Im just saying that they made a very popular error.
But now that Ive made this rather radical charge, I owe you an explanation as to
why so many educated people are so mistaken about such a fundamental truth
about arguments. By looking at the following arguments, I think we can find the
source of the confusion.
All 4 Runners are Toyotas.
This vehicle is a 4 Runner.
This vehicle is a Toyota.
We know full well that this is a deductive argument. And note its format: It
argues from a general premise to a specific conclusion about a particular vehicle.
Many swans were examined and found to be white.
All swans are white.
We know full well that this is an inductive argument. And note too its format of
arguing from more specific premises about particular swans to a general
conclusion about all swans.
This is where the mistaken account of the difference between deductive and
inductive arguments originates, I thinkwith a sampling error. Many deductive
arguments do draw specific conclusions from universal principles stated in their
premises; and many inductive arguments do state generalizations in their
conclusions based on specific cases cited in their premises. But this is just an
accident.
Just because many plants are green, this doesnt mean that being green is what
makes something a plant. And just because many animals are furry, this doesnt
mean that being furry is what makes something an animal. Anyone who thinks it
does should not call themselves a biologist, especially when they are pointing at
green frogs and calling them plants and pointing at fuzzy African Violets and
calling them animals!
If there is anything that a biologist should get right, its the most fundamental
distinction within their disciplinethe distinction between plants and animals!
Well, likewise for critical thinkers. If theres one thing they should get right, its
the most fundamental distinction between argumentsthe distinction between
deductive and inductive arguments.
42
The old mistaken distinction gets things accidentally correct once and a while,
but it often gets things just plain wrong. Here are some examples to illustrate
this:
I own a vehicle with Minnesota license number XXX123.
The vehicle with Minnesota license number XXX123 is a Toyota.
I own a Toyota.
We know full well that this is a deductive argument. But if we were to classify it
on the basis of the popular account, what would it be? It argues from specific
premises about a particular vehicle and license number, to a specific conclusion
about that same vehicle. So it couldnt be either deductive or inductiveits
schminductive?!
The popular account cant handle this very common type of argument. But our
analysis had no problem figuring out what type of argument it is. Try this one:
Many overhead projectors that Ive dropped have broken.
The next overhead projector I drop will break.
This is obviously an inductive argument. But what would the popular account do
with it? It argues from a more general premise to a specific conclusion; so, the
popular account would mistakenly classify this as a deductive argument.
Pop quiz: Correctly classify the following arguments. How

would the popular account handle them?
An overhead projector I once dropped broke.
The next one I drop will break too.
All overhead projectors are physical objects.
All physical objects have mass.
All overhead projectors have mass.
Check this out: Even one of the greatest critical thinkers of all
time, Sherlock Holmes, was totally screwed up about the
deductiveinductive distinction. When he thought he was
deducing conclusions on the basis of his observations, he was
in fact making inductive inferences. Have a lookgo to the
website www.concordance.com and do a search on Sherlock
Holmes, and then do a couple of searches on deduce and
43
deduction, to access the relevant portions of the Complete

Works.
Validity revisited
Ive discussed how validity is a very technical notion for us as we study critical
thinking. From our point of view of logic, remember, validity is simply a matter
of it being logically impossible for the premises to be true while the conclusion is
false.
To really highlight how technical and fine-grained our working concept of
validity is, lets critically examine the following claim from Trudy Govier, in A
Practical Study of Argument:
Deductive entailment is so tidy and complete a connection that
when we have it there is no need to consider separately the
relevance of premises and the issue of whether they provide
good grounds for the conclusion.
Goviers point is that when an argument meets the groundedness condition for
cogency, we can automatically check off the relevance condition as also having
been met. Well, that might normally be true, but its not always the case. And
by taking a look at some of those exceptions, we will get an even more robust
understanding of what validity is and an even better idea of what the relevance
condition is all about.
Bill Clinton was president.
Bill Clinton was not president.
This is a critical thinking etext.
Is this argument valid? Its squirrelly; thats for sure. But is it valid?
The question to ask, of course, is whether its logically impossible for the premises to be true and the conclusion false at the same time. And the answer is
Yes. Its impossible for the premises to be true while the conclusion is false,
because its simply impossible for the premises to be trueits impossible for
Clinton both to have been president and not to have been president. That would
be a contradiction, and contradictions cant be true. Look at it this way: If its
simply impossible for me to be in Chicago right now, then its surely also
impossible for me to be in Chicago right now having a good time. See?
So the argument is valid. And, by being valid, it meets the groundedness condition [if only by getting by on a technicality!]. Does it meet the relevance
condition? Do the premises provide one with even the weakest reason to believe
44
the conclusion? The answer is No. Neither the Clinton presidency nor its
termination provided any reason whatsoever to believe that this is a critical
thinking etext.
So we see already that meeting the groundedness condition for cogency does not
guarantee that one has met the relevance condition.
Pop quiz: Is the Clinton argument above sound [why/why not]?

Is it cogent [why/why not]?
Something to think about: Take yet another look at our rather

strange argument above about Clinton. What else would validly
follow as a conclusion? How about if we inserted for our
conclusion, This is a cookbook? Would this resultant
argument be valid? How about if we concluded, It takes a
mighty big dog to weigh a ton? Valid? See the trend? So,
what can be deduced from a contradiction?
Answer: Anything and everything! And thats why you never
want to use contradictory premises; because if you use those
premises to draw your conclusion, what can your opponent
do? And you could hardly objectwhy?
Heres another rather strange argument:

If Bill Clinton was president, then 2 + 2 = 4.
2 + 2 = 4.
Is this argument valid?
The answer, again, is Yes. Its logically impossible for the premises to be true
while the conclusion is false, simply because its logically impossible for the
conclusion to be false. [To deny that 2 + 2 = 4 would lead to a contradiction;
e.g., if the way that 2 + 2 4 is for 2 + 2 = 5, then that would contradict the fact
that 2 + 2 2 + 3, since 2 + 3 = 5.]
Are the premises relevant?
And again, the answer is No. The Clinton presidency certainly was evidence
for many things, some good and some not so good, but it certainly provided no
reason whatsoever for thinking that 2 + 2 = 4. So, once again we see that one
45
cannot validly infer that the relevance condition is met simply on the basis of an
arguments being valid.
Pop quiz: Is the argument sound?

Answer: Its a bit unfair to ask you right now, but the answer is
Yes. Both of the premises are true, although its probably a bit
tough for you to understand how the first premise could be true,
since Clintons past presidency certainly isnt whats making
2 + 2 = 4. We will thoroughly study statements such as this one
latertheyre called conditional statementsand well see
exactly why, as squirrelly as that first premise is, its indeed true.
Another pop quiz: Ill leave the answers and their explanations
for you to discover:
Is this argument valid?
Are the premises relevant?
Is the argument sound?
Whats the big deal about distinguishing these ARG conditions?

Acceptability of premises:
One has more reason to believe

the premises than to not believe
them.
Relevance of premises:
The premises provide some

reason to believe the conclusion.
Premises Ground the conclusion:
Premises provide adequate

evidence or reason to believe
the conclusion, making the
conclusion probably true, given
the truth of the premises.
46
By this time it must be striking you that Im on a crusade about keeping the three
ARG conditions distinct. Before you begin chalking me up as an obsessivecompulsive, let me lay out my reasons.
Say your car is running a bit rough and you take it into the shop; what would you
think if all the mechanic could say about your car is, It dont run so good? You
knew that! This clown calls themself a mechanic, and thats all they can offer by
way of a diagnosis?!
You want a mechanic who knows why its not running. You want someone who
can distinguish the various systems of the car and knows what makes them wellfunctioning and what causes their malfunctioning; so he or she can fix the broken
parts and leave the good parts alone.
And so it is with arguments. We want to be able to say more than just, That
dont sound so good, when were confronted with a poor argument. We want to
be able to spot exactly where the argument is going right and where its going
wrong; so that when arguments go haywire, we can fix what needs fixing and
only what needs fixing.
And thats why we are making such a big deal about distinguishing the ARG
conditions. And thats why our whole project here, in our goal of becoming good
critical thinkers, is to become proficient at applying the ARG conditions to
argumentswhether they be inductive or deductive arguments; whether they be
others or our own.
Applying the ARG conditions

Even with only our brief introduction to the ARG conditions, we should already
begin cultivating the habit of making arguments jump through their hoops,
thereby demonstrating their cogency.
Use the ARG conditions to construct your own arguments:

Ask, Have I met all of the ARG conditions?
Use the ARG conditions to critically review the arguments

of others: Ask, Have they met all of the ARG conditions?
Dont be lulled into thinking that an argument is cogent simply because you
happen to agree with its conclusion. Remember: There are plenty of terrible
arguments for conclusions that happen to be accidentally true. So when someone
offers you a noncogent argument for your conclusion, you would do well to say,
Thanks, but no thanks. Many people cant understand how you could accept
their conclusion but object to their argument; but thats just their problemtheir
47
inability to distinguish between the truth of a statement and the cogency of an

argument.
Dont be lulled into thinking that an argument is noncogent simply because
you disagree with the conclusion. Someone has just handed you an argument;
someone has just offered you a reason to believe their conclusion. The ball is
now in your court; the burden of proof is on you. You have three options under
these circumstances:
1. Admit that their argument is cogent and that they have just
given you a reason for changing your mind about their
conclusion.
2. Admit that their argument is cogent and continue to disagree
with their conclusion just as strongly as ever. This is the
option of intellectual dishonesty; totally antithetical to our
goals as critical thinkers.
3. Continue to disagree with the conclusion, but only after
finding out what is wrong with their argumentviz., finding
out exactly which ARG condition their argument violates.
Failing this third option, one is left with the first option as the only rational thing
to do. For us, intellectual dishonesty is not an option.
The limits of criticism

The most you can produce, by pointing out someones failure to meet all the
ARG conditions, is an argument that their argument is not cogent, i.e., an
argument that their argument has failed to provide adequate reason to believe
their conclusion.
Your criticism, then, has the following schematic form against their argument, A:
1. As premises are unacceptable. OR
1*. As premises are irrelevant. OR
1**. As premises fail to make the conclusion reasonable to believe.
2. A fails the ARG conditions.
3. A is cogent only if it satisfies the ARG conditions.
4. A is not a cogent argument for its conclusion, C.
At most, this is a successful criticism of an argument for conclusion C.
48
But this just removes one reason for believing C is true. This is not the same as
providing a reason for believing C is false. This would require giving an
argument for not C.
Lacking a reason to think that C is true does not mean that one has a reason for
thinking C is false.
Furthermore, providing such a reason for thinking C is false, viz., giving a n
argument for not C, is not the same as providing a reason for believing some
other alternative conclusion C* is true. This would require giving a different
argument for C*.
We must always keep the following distinct:
A criticism of an argument for C
An argument against C
An argument for C*
We must always remember that:
A criticism of an argument for C is not an argument against C.
An argument against C is not an argument for C*.
Something to think about: How often have you thought your

view must be right simply on the basis of thinking that your
opponents arguments were not very persuasive?
A complete critic provides both forms of criticism:

A criticism of the opponents argument. and
A criticism of the opponents conclusion.
This is the one-two punch of the successful critic:
Remove the opponents reason for believing C is true. and
Provide them with a reason for believing C is false.
Sometimes the critic cannot be completely successful and must settle for only
onea criticism of their opponents argument or a criticism of their opponents
position.
49
The project of providing an argument for an alternative position C* would mark

the end of ones role as a critic. At that point, one has shifted from being a critic
to being an advocate of a position.
Whats next?
This ends our first look at the cogency conditions for arguments. In the next
chapter, we will look at the most fundamental material of argumentation and
critical reasoninglanguage. Coming up: The use and abuse of language in
critical thinking.
50
The Use and Abuse of Language
5. THE USE AND ABUSE OF LANGUAGE

Twas brillig, and the slithy toves
Did gyre and gimble in the wabe:
All mimsy were the borogoves,
And the mome raths outgrabe.
Huh?! Exactly! This is the first stanza of Lewis Carrolls poem entitled
Jabberwocky, found in Through the Looking Glass. If you read it aloud, it
sounds sort of OK, but then it hits you: What does it mean? What is it to be
brillig? What are toves? How do you gyre and gimble, especially in the wabe?
[And isnt that illegal in most states?]
In fact, these words have no meaning at all, as much as they might sound as if
they do. The Jabberwocky is just great-sounding nonsense.
Seldom is our problem one of talking the likes of the Jabberwocky poem.
[Although I have to wonder about some folks and their psychobabble of selfhelp and management stylesyou know what I mean?] Usually the problem is
not lack of meaning, but rather lack of clarity of meaning.
How often have you witnessed, or even participated in, the following type of
exchange?
Jean: Genetic design is safe, because its been going on since
the first cell mutated into a cell that did better than its
predecessors. There is only a slight chance that selective DNA
manipulation will produce an organism that could cause any
widespread environmental damage. Its not like with cases of
high doses of radiation, where there is significant risk of damage
to the organism or its genetic material. We can be no more
justified in restricting genetic research as we are in restricting the
natural evolution of the species.
Parry: Thats totally wrongheaded! Genetic research cant be
safe. Manipulation of DNA in the laboratory setting is
unnatural. There is always some risk of producing an organism
that could severely damage the environmental or usperhaps a
virus against which we can form no antigens to ward off.
Genetic design must be prohibited for the sake of our safety and
the safety of our environment.
Jean and Parry are doomed, because theyre simply talking past each other. In
a sense, they think they disagree about the safety of genetic engineering, but in
51
another sense they are not disagreeing at all. In a sense, they are just talking two
different languages and, as a result, might as well be on different ends of the
planet. The only solution to their problem is to call a linguistic time-out from
their debate and come to some kind of agreement on what constitutes a safe
research practice.
Both Jean and Parry agree that its physically possible for genetic research to
result in an organism that causes widespread environmental damage. For Parry,
however, this means that the research is unsafe. Parry is using a different
definition or meaning of what it is to be safe than Jean is using. For Parry, safe
means the physical impossibility of damage. That is not what Jean means by the
word safe. For Jean, safe means being within normal and tolerated risks of
damage. So for Jean, an activity would have to significantly increase the chances
of harm over the normal base rate in order to qualify as being unsafe.
It appears that Jean is more in tune with what we standardly mean by a safe
activity, and Parry is quite out of step with our ordinary understanding. Look at
it this way: If Parry were right about what being unsafe is all about, then
everything we do is unsafe, and we would have to call ourselves a constant
danger to others. As you meet a friend on the street you wave to them; but in so
doing, there is a chance that your watch band could break, flinging your watch
into their face, and taking their eye out! Its physically possible! Does that make
you a walking hazard to others? Does that mean we should prohibit you from
waving to your friends, at least when you have your watch on? Of course not!
No, by simply sharing the planet with others, we are forced to accept some risks
in exchange for the benefits of doing social actions. Its called having a life.
Driving a car while being stone sober and attentive still poses the risk of mowing
a pedestrian downhe was right in the window post and he dashed out on the
road and you didnt see him in time. Thats a tragic accident; its not reckless
driving. But if you do two six-packs and get one eye looken at the other and
then go cruising around, thats unsafe driving. Driving while drunk significantly
increases the probabilities of injuring others over the base rate of sober, attentive
driving.
So before Jean and Parry started their exchange of arguments, they should have
called a little linguistic convention, to come to some agreement about what they
both mean by the words safe and unsafe.
And, if they wanted their debate to be understood and used by others, they
should also agree to use those words in the same way the general public does, or
they will still be talking past everyone else.
52
Definitions
The episode with Jean and Parry illustrates the importance of clearly and
explicitly defining ones terms. Later, we will see other benefits of doing this.
Right now, however, lets address the fundamental question of what a definition
is.
What is the definition of definition?

Bill Gates asked this very question while being deposed in the Justice
Departments antitrust suit against Microsoft. Fortunately, for us anyway, theres
a clear answer:
A definition is the meaning of a word [or phrase] as it is
commonly intended or used by ones linguistic community.
The best way to get a full understanding of what a definition is, is to think of it as
a recipea fundamental recipe for making the thing being defined. Lets
play with an example to illustrate this. Lets define apple pie. Lets list the
fundamental, most basic ingredients for being an apple pie.
Whats first on our list? How about apples? Does an apple pie have to have
apples? Youd think so, but what about that recipe on the back of Ritz Crackers
boxes years ago, in which Ritz Crackers were substituted for apples in the pie?
Yes, well, even the Ritz folks didnt have the nerve to call this apple pie. They
called it Mock Apple Pie. So indeed, one needs apples in order to have an
apple pie; you just cant have an apple pie without them.
So weve found our first fundamental ingredient, but certainly thats not enough.
What else must we add in order to make an apple pie? How about crust? Sounds
right, but how many? Usually, an apple pie has two crustsa top and a bottom.
But even though paradigm cases of apple pies have two crusts, there is a kind of
apple pie that only has one. Its Dutch Apple Pie; its the kind with that crumbly
junk on top instead of a dough crust. If an apple pie needed two crusts in order to
be an apple pie, wed have to say that Dutch Apple Pie isnt apple pie at allits
something like Dutch Schmapple Pie. But we know full well that Dutch Apple
Pies are still apple pies.
So we should adjust our definition, so that it requires only at least one crust,
because indeed at least one crust is needed! Without a crust, wed have apple
sauce, apple crisp, apple cobbler, but not apple pie. [We could go on here to
define what a crust is, but lets focus on finishing our definition of apple pie.]
53
So far we have:
apples
crustat least one
Is that enough? Not really. This would mean that if someone rolled out a crust
and placed an apple or two on top of it, it would be an apple pie. And it just
isnt; its just a crust with a couple of apples laying on it. An apple pie must be
put together in the right way. The apples must be sliced up a bit. To capture this
in our definition, lets just say Some assembly required. [This too would need
further elaboration, but I think we have a pretty good idea of what were referring
to here.]
Lets round up our ingredients and see if theyre enough:
apples
crustat least one
proper assembly
This still falls short, because you could roll out a crust, slice some apples on it,
fold one half over the other, pinch the edges, and all youd get is an apple
turnover, not an apple pie. Shape is important here. Its got to be, well, pieshaped. But whats that? Well, its that the crust is shaped or formed in what
we call a pie plate or baking tin. Lets call this ingredient pieshaped and add it
to our list. [Well see in a bit, whats wrong with this, but lets move on for
now.]
apples
crustat least one
proper assembly
pieshaped
Will this do the job? If we get all these ingredients together, do we automatically
have an apple pie, or do we need some more ingredients?
Do we need to bake it? Well, let me tell you a little story. When I was a kid, we
had an apple tree. My job was to pick the apples and clean, peel, and slice them.
In the mean time, my mom made crusts, rolling them out, laying them in tins, and
passing them to me. I would then fill them with apple slices, a little sugar, and
more cinnamon than she liked, and pass them back to her. Shed then lay the top
crusts on and slide them back to me. Id press the crusts edges together with a
fork, trimming off the excess, and wrap them in tin foil and put them in the
freezer by the dozens for winter. What was I putting in the freezer? If we say
that, by definition, apple pies need to be baked, then I cant have been putting
54
apple pies in the freezerthey must have been schmapple pies. But we know
full well that they were apple pies; so baking should not be added to our list.
How about sugar? Can we get by without it? Well, yes. Youd have an apple
pie so tart it would grab your throat like a pit bull, but it would still be very sour
apple pie. How about cinnamon? Do we need it? Again, we can get by without
it. Our definition guarantees only an apple pie when were done, not necessarily
a tasty one.
So it looks as if weve finished our definition.
Weve discovered all the ingredients that an apple pie must have in order to be
an apple pie. These are called the necessary conditions.
Weve also discovered all the necessary conditions which when rounded up
automatically guarantee us an apple pie. These are called the jointly sufficient
conditions.
So, the primary goal of a definition is to provide the necessary and jointly
sufficient conditions for the proper use of a term.
Conceptual analysis
The goofy little example we just did, clarifying the definition of apple pie, was
a good example of whats called conceptual analysis.
Use this method of conceptual analysis to discover and refine any definition:
Begin with some paradigm examplessome obvious cases

[such as we did with plain old double-crust apple pies].
Among those examples, discover those features that are

necessary in order to be the thing being defined. These features
are necessary conditions [as with our apples].
Eventually gather all the necessary features that will make for
the thing being defined. These features are jointly sufficient
conditions.
Use counterexamples to argue against erroneous candidates for

necessary or jointly sufficient conditions [counterexamples such
as my story about making (frozen) apples pies with my mom,
indicating that baking is not a necessary condition].
55
The Goldilocks method

Its pretty easy to fall short of your goal of providing necessary and sufficient
conditions, while youre attempting to provide a definition.
When you mistakenly add a condition that is in fact not necessary, your
resultant definition is said to be too narrow. For example, when the first draft of
our definition of apple pie included the requirement of having two crusts. This
definition was then too narrow, too restrictive, since it prohibited our citing
Dutch Apple Pies as apple pies.
When you mistakenly omit a condition that in fact is necessary, your resultant
definition is said to be too broad. For example, when our draft omitted the
requirement of being pie-shaped. This definition was then too liberal with the
way it would label things apple pie and would classify apple turnovers as apple
pies.
Definitions need to correctly pick out all and only the things being defined. You
dont want a definition thats too narrow, and you dont want a definition thats
too broad. You want a definition thats juuuuuust right!
And thats why we use conceptual analysisto locate all and only the necessary
conditions for the thing defined, by testing all our candidates for necessary
conditions by using paradigm cases and counterexamples. Were working
towards a definition thats not too narrow, not too broad, but instead just
right.
Something to think about: Think about your current [or future]

status, occupation, or profession. Detail its definition. For
example, if youre a student, what is the definition of being a
student attending your institution? Your universitys catalog is
your guide on this. Note any sets of requirements that might
seem plausible at first but that ultimately result in definitions that
are either too narrow or too broad.
Something else to think about: A definition can actually fail in

both respectsit can be both too narrow and too broad. Its not
so strange or uncommon, when you stop to think about it. See if
you can come up with a definition of apple pie that falls victim
to this.
56
Psst, do you realize what youre doing?

By now youve gotten comfortable working with definitions and the method of
conceptual analysisthe Goldilocks methodused in discovering and refining
definitions. You are now ready for the shocking truth about this method or
process of inquiry. Its also called philosophical analysis [imagine the
soundtrack from the shower scene in Psycho playing right now].
As a definition provides necessary and sufficient conditions, it details an analysis
of that which is defined. The definition is a complete set of the essential
features of the thing being defined. This is one of the primary projects of
philosophyto discover what something is. For example, Platos Republic is
dedicated solely to the question What is justice? Whenever philosophy asks
one of those heady What is ? questions, its inquiring into the nature of
something; its trying to give an analysis or an essentialist explanation of what it
is to be that thing.
So, when we were trying to get clear on our definition of apple pie, we were
really doing some pretty heady philosophical analysis, believe it or not. And it
didnt hurt a bit, did it?
The courage to define

When you detail a definition, youre offering an analysis of the nature of that
thing being defined. This is one of the gutsiest things you can do, intellectually;
because this means that the definition you are offering covers all logically
possible cases.
Take, for example, the definition of bachelor, viz., an unmarried male. Are we
claiming that bachelors are unmarried males only in our state, or country, or only
on Earth? No, were claiming that bachelors are unmarried males in every nook
and cranny of the universe! Just in the past thousand years and in the next
thousand? No, for all time so far and for all time to come! Were saying that no
matter how the universe was, is, will be, and could be, a bachelor is nothing more
nor less than an unmarried male.
Because a definition is claimed to account for all logically possible cases, it is
fair game to use merely logically possible cases as counterexamples to
someones proposed definition.
A famous example of this occurs in ethics, when defining what it is to be a
person. A person is the kind of being with moral status to such a degree that it
warrants moral respect of the highest order, viz., it has rightsoften called
human rights. The most obvious definition of a person is being a humanviz.,
57
being a Homo sapienleaving it to biology to determine the set of necessary and

sufficient conditions that make and mark out our species. The recently recognized problem with this definition is that it is too narrow: It completely misses
perfectly good instances of persons such as Spock, on the original Star Trek
series. Spock was Vulcan on his fathers side and Homo sapien on his mothers
side. He was not human. By the biologically-based definition of a person, then,
Spock had no human rights and would not command the same moral respect that
we do. At most he would warrant the moral respect of other sentient creatures
such as dogs and cats. But that cant be right! Spock was a better person, from a
moral point of view, than most of us! And any definition of personhood is
wrongheaded for not recognizing Spock as being as much a person as you or I.
Spock is a hypothetical character, and yet any definition of what it is to be a
person and a rightsholder better apply to him as well of any of us. Being Homo
sapien is not what makes a person a person; its not the essence of personhood;
so its not what explains our status as persons and rightsholders. So we need
to find a broader analysis of personhood, to explain what makes us personsan
analysis that would explain what would make Spock a person and a rightsholder
too. This is true even if in actuality, we are alone in the universe and only Homo
sapiens happen to be persons.
Hey, its not just us goofy philosophers!

Before you are inclined to think that only philosophers are interested in these
kinds of weird hypothetical cases, let me just ask you this: What would a
biologist say about the following definition?
A Homo sapien is a mammal with a normal body temperature of
98.6 degrees Fahrenheit.
A biologist would say that this definition of our species is so wrong that its just
plain silly. And yet that same biologist probably knows that it is a biological fact
at this point in the history of the species that all and only Homo sapiens have a
normal body temperature of 98.6 degrees Fahrenheit.
If this is in fact true, then how would the biologist object to the definition? By
pointing out that this condition only accidentally picks out our species; its not
what makes our species what it is; its not essential. And how is the biologist
going to argue for that claim? Well, not by pointing out the occasional case of
someone with a body temperature of 98.7, because those cases are not normal;
they are cases of infection or of abnormally high rates of metabolism.
Biologists will make their case by pointing out that we could have evolved with a
normal body temperature of 98.7 or 98.5, or that we could still evolve to have
such normal body temperature, and still be Homo sapiens (as opposed to Homo
58
schmapiens or something). But notice that, in so arguing against this definition,

the biologist is appealing to what kind of counterexamples? Totally
hypothetical cases! So its not just philosophers that dive into the pool of purely
hypothetical cases to find counterexamples of proposed definitions. Everyone
who is trying to find the nature and explanation of things dives into that pool.
Something to think about: Critically review the following

definition of human being: mammal with 46 chromosomes.
Something else to think about: So how would you define what

it is to be a person and a rightsholder, so as to accommodate
Spocks obvious personhood?
Even more to think about: If there are persons who are not
human, are there humans that are not persons? And do they have
rights?
The dictionary
Youve been told since childhood that the best [or only] source of definitions is
the dictionary. Well, after analyzing what a definition is, we can now see that
thats not necessarily true.
The dictionary should have the first word on what the definition of a word
is, but not necessarily the last word.
So far weve discussed two goals of a definition:
1) To detail the meaning of a word as it is commonly
intended or used by ones linguistic community.
2) To detail the necessary and sufficient conditions for being
the thing defined.
Quite often, these goals do not coincide. Quite often, the general linguistic
communitys understanding of the complete set of necessary and sufficient
conditions for the appropriate application of a term is rather impoverished or in
error. So, since the primary function of the dictionary is to accurately reflect how
the linguistic community uses terms, its not surprising that it will frequently
stray from accurately detailing the actual analyses of things.
59
If there werent this divergence, standard dictionaries would be as complete as

science texts; in fact they would incorporate all the definitions of the sciences.
But, of course, standard dictionaries dont do this. They let the sciences detail
the physical natures or essential properties of things and, instead, focus more on
representing ordinary usage.
So, come time to tackle some tough definitionssuch as defining what justice is,
what a human right is, what a person is, what a life form is, what a virus isthe
dictionary is a start, but it must be supplemented with some thorough conceptual
analysis, analysis that often requires a great deal of background knowledge in, for
example, the sciences or even philosophy.
Other virtues of a definition

A good definition tries its best to detail the necessary and sufficient conditions
for what is being defined, but it should also increase our understanding. The
definition should give one a clearer idea as to the nature of the thing defined.
This is not achieved if the word being defined appears in the definition in some
form. Remember how irritating it was when the dictionary would hand you
definitions like: bachelorperson in a state of bachelorhood? Or:
gamekeeperperson charged with keeping game. If you knew what bachelorhood is, you wouldnt be looking up bachelor; if you knew what keeper of
game was, you wouldnt be looking up gamekeeper.
Dictionaries arent always this bad. I use these drastic examples only to drive
home the moral that you should never use the word being defined in its own
definition. If you stick your nose up in the air, this rule would be stated as
follows: The definiens must never repeat the definiendum. [A good way to
remember which is which, is that word that needs defining is the dumb one.]
A definition that falls victim to this charge is called a circular definition or a
question-begging definition.
Pop quiz: Heres a rather famous instance of a circular

definition that is used in the context of the abortion debate, about
whether or not the fetus is a human being like you and I. Find
and explain the circular definition.
Once conceived, the being was recognized as man
because he had mans potential. The criterion for
humanity, thus, was simple and all-embracing: if you
are conceived by human parents, you are human. [John
Noonan, An Almost Absolute Value in History]
60
Another pop quiz: Where was our definition of apple pie

circular? How would you remedy it?
Another way in which a definition can fail to increase ones understanding of the
definiendum is when the definiens uses more esoteric terms or concepts than the
definiendum was in the first place. Remember how irritating it was to look up a
word only to have to look up another word that was used in the definition?
The moral of this story : Use more primitive and understandable terms or
concepts in the definition.
Of course, sometimes this virtue of a good definition is incompatible with
another virtue. Sometimes, for example, increasing the accuracy of a definition
entails using less primitive concepts, with the resultant definition being less
intelligible to the layperson than was the original term. You have to make a
judgment call at that point, between losing a bit of your accuracy or losing a
portion of your audience.
The use-mention distinction

Perhaps youve already noticed the fact that Ive been putting single quotation
marks around words now and then. For example: bachelor means unmarried
male.
Well, heres what Im up to. Im using single quotation marks to indicate that
Im merely mentioning the word, instead of using it. When I claim that
bachelor has eight letters, Im obviously talking about the word, the type of
squiggles on your page or screen, and not any unmarried males that the word
stands for.
This has come to be called the use-mention distinction. Other forms of
punctuation have been used to indicate that one is merely mentioning a term, e.g.,
italics, regular double quotation marks, and underlining; but these have other
linguistic functions as well, while single quotation marks are rather specifically
used to indicate mere mention.
The use-mention distinction is not just important for giving definitions or
discussing pieces of the language. The failure to recognize the use-mention
distinction can cause a great deal of misunderstanding and false accusation. Let
me demonstrate.
61
If I tell you that Ann said that there are no good restaurants in Winona, am I
claiming that there are no good restaurants in Winona? No, and anyone ascribing
me with that claim would be misrepresenting me and what I said. Not much
hangs in the balance in this instance; but in other cases, it might be different.
If I tell you that Bob said youre an idiot, am I calling you an idiot? No, but if
you ignore the use-mention distinction, you might mistakenly think that I just
insulted you by calling you an idiot, when I didnt at all. And this is how
feelings can get hurt and accusations begin to fly.
This especially happened to a friend of mine who was teaching a writing seminar
at a major corporation. The topic was the use of informal language in business
correspondence, and one participant asked if there were ever an occasion in
which the word nigger could be used. After the shock of the question wore off,
my friend said that it would never, never, ever be appropriate or permissible to
use that expression. She was later accused by the corporations Human
Resources Department of using racist language in her seminar. This was a most
unfortunate misrepresentation of herher attempt to battle against racism was
treated as an expression of it.
Indeed, here is a case in which a word has such a history that it may be
psychologically impossible for us to maintain the use-mention distinction. This
is why its often best to only indirectly mention those offensive cases when we
must, e.g., by means of abbreviations, as with the N word.
Pop quiz: Which rock star relied heavily on the use-mention

distinction?
Answer: Prince, whose name was Prince and then The artist
formerally known as Prince, and then Prince once againas
opposed to The artist formerally known as The artist formerally
known as Prince. [That would have been worse than that
goofy symbol was!]
Another pop quiz: Puzzler, from Cartalk, show #9827: A

customer walks into a hardware store and asks How much is
1? The clerk says, $60. How about 12? $120. Ill take
200. Thatll be $180. Whats the person buying?
Answer: House numbersthe customer is merely mentioning
the numbers. A lot of brain teasers and jokes are a function of
playing with the use-mention distinction.
62
Conclusions by definition
We have been discussing definitions and the importance of sharing the same
definitions in order to communicate with others and understand their claims and
their arguments. Lets now examine a more specific use of definitions.
We often say that a statement is true by definition. For example, if Bob is a
bachelor, then we could say that Bob is unmarried, by definition. Arguments in
which the conclusion follows by definition are all valid deductive arguments.
Their conclusions will be uninformative, in the same sense that all deductively
valid arguments are, and often rather obviously so, as with the case about Bob.
The conclusion is just unpacking one of the necessary conditions making up the
definition.
And this is why, in this kind of argument, all the real work is being done by the
definition given in the premise. But, garbage in might well get you garbage
out, as we see in the following argument.
The American Heritage Dictionary defines murder as The
unlawful killing of one human being by another, esp. with malice
aforethought. And State executions are neither unlawful nor are
they done with malice aforethought. Therefore, the death
penalty is not murder.
This illustrates the danger of blithely appealing to the dictionary as a source of
argumentation. Remember, the dictionary may have the first word, but never the
last; here the dictionary definition is too narrow, by requiring that the murder be
against the law. Certainly, a legalized case of murder is not a contradictionthe
Holocaust illustrates this.
There is no substitute for thorough conceptual analysis. And then, once a
clear and well-tested analysis or definition is determined, conclusions can be
more confidently drawn from it.
This is exactly the project in crucial ethical debates about the moral
permissibility of abortion or of euthanasia. Lives literally hang in the balance
depending on determining the correct analysis or definition of a person or of
death. We must first figure out what makes a person a person before we could
judge whether the fetus is one, just as the field biologist must know what makes a
robin a robin before they could go out into the field and identify one.
Likewise with death. Throughout the history of medical science, there has been
an evolving definition of death. With our current analysis of persons, in terms of
their minimal psychological capacities, the ceasing of those capacities has
63
become the defining characteristic of death. So if the patient is all-but-for-thebrainstem-brain-dead, they are dead, despite the presence of a heartbeat,
breathing, and even reflex movements. Obviously, the cognitive analyses of
personhood and death have profound implications in issues of abortion and
euthanasia.
The abuse of definitions

They should simply administer a lethal injection to Uncle Bob.
His knees are so arthritic that he can no longer get around. And
someone who cant get around at all is already essentially dead,
because they have lost the quality of life.
If this argument strikes you as a bit strange, it should. It is less a use of the
definition of death and more the abuse of a definition of death. The author came
to the conclusion that Uncle Bob was already deadand so administering the
injection is not even a case of wrongful killing, since one cant kill someone
whos already deadjust by playing with the definition of death.
The author defined death as merely the loss of the quality of life. But what
constitutes the quality of life? Its going to differ from one person to another.
And a moments reflection tells us why: Because what constitutes the quality of
life is purely subjectiveit makes no sense to think there could be an expert on
the matterand so all opinions of what constitutes the quality of life are equal,
including the authors.
However, if I had the squirrelly notion that life is not worth living unless one
owns a fleet of speedboats, I could slit your throat and only be subject to the
charge of desecrating a body, but not killing you, since you were dead already for
lack of any quality of life.
This devious argument stems from my abuse of the definition of death. It stems
from my use of a stipulative definition of death.
A stipulative definition is necessary and appropriate if one is introducing a word
into the language. In a sense, all words began their careers in the language with
stipulative definitions. But the moment a word has been adopted and has a
common or shared use within the linguistic community, introducing a new
meaning of the word results in a very misleading stipulative definition.
This is also called the redefinist fallacy.
One of the most famous advocates of the stipulative definition was Humpty
Dumpty, in Lewis Carrolls Through the Looking Glass. In fact, Humpty
64
Dumpty thought that the only definition was a stipulative one! Lets see where
that would leave us:
Theres glory for you!
I dont know what you mean by glory, Alice said.
Humpty Dumpty smiled contemptuously. Of course
you donttill I tell you. I meant theres a nice
knock-down argument for you!
But glory doesnt mean a nice knock-down
argument, Alice objected.
When I use a word, Humpty Dumpty said, in rather a
scornful tone, it means just what I choose it to
meanneither more nor less.
The question is, said Alice, whether you can make
words mean so many different things.
The question is, said Humpty Dumpty, which is to be
masterthats all.
Alice was too much puzzled to say anything; so after a
minute Humpty Dumpty began again. Theyve a
temper, some of themparticularly verbs: theyre
the proudestadjectives you can do anything with,
but not verbshowever, I can manage the whole lot
of them! Impenetrability! Thats what I say!
Would you tell me, please, said Alice, what that
means?
Now you talk like a reasonable child, said Humpty
Dumpty, looking very much pleased. I meant by
impenetrability that weve had enough of that
subject, and it would be just as well if youd mention
what you mean to do next, as I suppose you dont
mean to stop here all the rest of your life.
Thats a great deal to make one word mean, Alice said
in a thoughtful tone.
When I make a word do a lot of work like that, said
Humpty Dumpty, I always pay it extra.
Oh! said Alice. She was too much puzzled to make
any other remark.
People seldom buy into Humpty Dumptys theory of meaning [thank goodness!],
but they still too often commit the redefinist fallacy: Rather than taking time to
argue for a conclusion, people will often just redefine it in terms of something
already granted. Its tough to argue that dogs have five legs; but it seems a lot
easier if one simply defines a tail as a leg.
65
But remember: If we call a dogs tail a leg, how many legs does a dog have?
Four! Calling a tail a leg doesnt make it one!
For example: Stephen Young, former dean of Hamline

University Law School, filed a lawsuit against two
environmental groups, to block them from trying to reduce
logging on federal forests. In the suit, he argued that the groups
efforts to reduce federal timber sales were equivalent to
imposing the religion of ecology on the government and on the
loggers whose livelihood would be affected by the reduction in
logging. He said the groups were violating the constitutional
separation of church and state, as a result. [Adapted from
Associated Press, Minneapolis Star Tribune, May 28, 2000]
Youngs redefinist fallacy is, of course, his equating ecology
with a form of religion. A judge promptly dismissed the case
and actually fined and sanctioned Young for filing such a
frivolous case.
Learning right by doing wrong: Use the redefinist fallacy to

argue that:
One should be able to teach religion in public

schools, because science is already taught.
Domestic violence is a terrorist act.
Domestic violence is a right to privacy.
We are all political prisoners.
BTW: Heres another stanza from Jabberwocky:

And hast thou slain the Jabberwock?
Come to my arms, my beamish boy!
O frabjous day! Callooh! Callay!
He chortled in his joy.
Lewis Carroll was so good at making meaningless sounds sound
meaningful that he actually created meaningful wordschortle
soon appeared in the Oxford English Dictionary and remains in
66
our language today, meaning a combination of a chuckle and a

snort.
Pop quiz: You seem very clever at explaining words, Sir, said
Alice [to Humpty Dumpty]. Would you kindly tell me the
meaning of the poem called Jabberwocky?
Smart move on the part of Alice?
Persuasive definitions
Persuasive definitions are not so called because they persuade people into
adopting them by virtue of how good they are. They are so called because of
their persuasive powers despite how bad they are.
These are definitions achieved by including valued or disvalued features that
are not necessary conditions. The result is what could be called a loaded
definition.
Why would anyone do this? Well, it makes it easier to elicit value-laden
conclusions from those value-laden definitions, as they appear in the premises of
ones arguments. If one starts by defining abortion as the murder of an
innocent child prior to its birth, its remarkably easy to draw the conclusion that
abortion is always morally wrong no matter what the circumstances. But this is
only because one has included the disvalue of being a murder in the very
definition of abortion; however, murder is not a necessary characteristic of
abortion.
Ridding the definition of its unnecessary values or disvalues is called
neutralizing the definition, and if we neutralize this definition of abortion, one
is left with the rather medical definition concerning the removal of the embryonic
or fetal contents of the uterus prior to birth. And now we are left with the
abortion issue, viz., whether abortion is morally permissible or not; its not a
foregone conclusion by virtue of a stipulative definition loaded with unnecessary
value-laden conditions.
Learning right by doing wrong: Heres another example of a

persuasive definition.
A police officer is a state employee with the legal right
to assault or even kill people when he or she objects to
their behavior.
67
[Wow; its hard to like or respect someone like that!]
Neutralize this definition of what a police officer is.
Now, help the cops out by pumping up the definition

with unnecessary values, so as to create a persuasive
definition to the advantage of the police.
The moral of the story: Dont use persuasive definitions! And neutralize them
whenever they appear in an argument youre assessing for cogency.
Loaded language
This warning against loaded definitions should be extended to descriptions in
general. People use loaded language quite often, as they present their arguments.
They pack their premises with values or disvalues that are not necessary, thereby
making it much easier to draw conclusions with those same values or disvalues.
We discussed how the groundedness condition for cogency can be thought of in
terms of an arguments capacity for truth and acceptability preservationif you
put true or believable premises in the hopper of an argument meeting the
groundedness condition, you probably get true or believable conclusions coming
out the back end of it as a result. Well, the groundedness condition preserves
values and disvalues in the same way.
This is how one can rather deviously draw value-laden conclusions, either good
or bad, about someone, simply by putting the right spin on their descriptions in
the premises. In this way, one can make Mother Theresa a slacker, for never
paying taxes; or make the Unabomber a saint, for having such respect for our
environment. Ive noticed that talk-radio hosts can make a pressing ethical
dilemma out of anything, just by loading the descriptions: The topic today is
whether one ought to wade out to save a drowning infant in two feet of water.
Some say Yes, but do we really have the right to snatch another person,
without their consent, from the soft caress of the sea, at some risk of bodily harm
to them?! All of a sudden, the call-in lines are all ablaze with people who are
adamantly claiming that the only right thing to is to let the child drown! What
right do you have molesting that child and interfering with the natural flow of
events?! Thats the power of loaded language.
68
Pop quiz: Pick the neutral name:

Star Wars vs. Strategic Defense Initiative
Peacekeeper vs. MX Missile
The moral of the story: Dont use loaded language! And neutralize it wherever
it appears in the argument youre assessing for cogency.
Euphemistic language
This is the error of omitting values or disvalues that are necessary for a
complete and accurate description.
Loaded language is used to make it easier to draw normative conclusions
[i.e., conclusions about what ought or ought not to be the case], while
euphemistic language is used to make it more difficult to draw normative
conclusions. For example, its tough to fault someone for working on their
proposal, unless they are more accurately described as throwing it together at
the last minute.
Pop quiz: Neutralize the following euphemisms:

1. Not asked to continue ones employment
2. Revenue enhancement
3. Incomplete success
Pop quiz continued: Courtesy of Three Mile Island:

4. Energetic disassembly
5. Rapid oxidation
6. An event
An incident
An abnormal evolution
A normal aberration
A plant transient
69
The moral of the story: Dont use euphemistic language! And neutralize it
whenever it appears in the argument youre assessing for cogency.
We see once again that Goldilocks was correct, this time with respect to
value-laden languagenot too much, not too little, but juuust right.
Pop quiz answers:

1.
2.
3.
4.
5.
6.
Fired
Tax
Failure
Explosion
Fire
The accident
[I love the last two: Its so reassuring to know that this

was a normal aberration! And I thought a plant
transient was that little bag lady that hung around the
place a lot.]
Ambiguity
All my life, Ive wanted to be somebody. But now I see that I
should have been more specific. [Lily Tomlin, In Search of
Intelligent Life in the Universe]
Two very important phenomena in the language are ambiguity and vagueness.
Lets see what ambiguity is all about first.
Ambiguitywhen there are multiple definitions or meanings
of a word or phrase.
For example, the word bank has many meanings: A financial institution, the
side of a river, the action of bouncing or caroming off of [as in bank shot], to
have confidence in [as in you can bank on it]. And Im sure there are others.
Ironically, ambiguous is misused so often that its almost become ambiguous.
It is also used to say that a word or phrase is vague or unclear. This is most
unfortunate, however, because being ambiguous, being vague, and being unclear
are three quite different things, and we would do well to keep them distinct.
70
If a word or phrase is unclear, its simply that the reader or listener doesnt
understand it, for whatever reason. Its really that simple. So lets focus on what
ambiguity is now and talk about vagueness in a bit.
Many words are ambiguous, many in more than just two ways. So theres no
avoiding using ambiguous language. Ambiguity is quite a blessing, really. If we
couldnt make our words do double-duty like this, wed have to institute a
different word for each different meaning we wished to convey. What a very
different [and wordier] language wed have!
Since theres no getting out of using ambiguous language, you just have to be on
the alert to ensure that the context makes it obvious which way youre using it.
Some pretty silly things happen if youre not careful. For example, heres the
lead sentence from an Illinois newspaper, reporting the death of a man from the
town of Clinton: A 56-year-old Clinton man died as a result of something that
happened yesterday in a machine shed in Texas Township southwest of Clinton,
according to the coroners office. A natural reaction to this is to think, Well, of
course, the guy died of somethingthats no news. What would be news is if
he died of nothingthat would be worthy of an Extra: Uncaused Event
Happens in Clinton, Illinois; All Laws of Physics Proven Bankrupt! What the
newspaper was trying to say was that the Clinton man died of unknown causes;
but that intended meaning was not singled out by the choice of words, and the
result was unfortunateit undercut the credibility of the writer and might well
have offended the survivors of the poor guy from Clinton.
Ambiguity undercuts some of the most serious messages; take, for example, the
church billboard that read, Dont let worries kill you, let the church help.
Recently I saw an ad against domestic violence, consisting of a wedding picture
of a couple, and printed across the bottom of the picture was the caption, Fortytwo percent of all murdered women are killed by the same man. [Then put this
guy behind bars; youve even got his picture!] I also ran across a headline, in a
human resources department newsletter, which read, Sexual Harassment: Why
Arent People Getting It?. [I thought people didnt want it!]
Ambiguity thrives in headlines, because there is no context to disambiguate the
unfortunate phrase.
Another famous case occurred following the death of a tightrope walker, who
was crossing Niagara Falls [back when they permitted that sort of thing]. When
he was part way across, a gust of wind knocked him off the tightrope and sent
him down the falls to his death. The headline was: Man Blown to Death.
71
Here are some other headlines that slipped the leash on their authors. These
examples appear in the famous book, Squad Helps Dog Bite Victim and Other
Flubs from the Nations Press.
Old Miners Enjoy Benefits of Black Lung
Milk Drinkers Turn to Powder
Ban on Soliciting Dead in Trotwood
Fish & Game to Hold Annual Elections
Teen-age Prostitution Problem is Mounting
Man Robs, then Kills Himself
Woman Better After Being Thrown from High-Rise
Less Mishaps Than Expected Mar Holiday
Wives Kill Most Spouses in Chicago
Prostitutes Appeal to Pope
Informal fallacies
The headlines above are all honest mistakes using ambiguous terms and phrases.
But there are some dishonest mistakes too.
People use language to argue for their conclusions, but all too often people abuse
language to try to persuade their audience into believing conclusions for which
theyve been given no good reasons to believe. These rhetorical tricks are called
informal fallacies. If youre not on to them, theyre very persuasive; and thats
why they have such a long history of use. They are so infamous that theyve
even been named. So, the quicker you can identify them by name and understand
why they are fallaciousviz., why they flunk one of our ARG conditionsthe
quicker you wont fall victim to them any longer.
Well be studying about a dozen or so informal fallacies.
one involving ambiguity.
Lets begin with the
Equivocation
This informal fallacy is committed by abusing the ambiguity of a word or phrase.
Equivocationsupporting a conclusion by shifting from one
meaning of an ambiguous term to another.
Throughout our study of informal fallacies, Ill give you a basic account of what
each one is; but these definitions will probably make sense to you only after
looking at an example or two.
72
Here is a plausible-sounding argument that loving parents might well trot out
upon hearing that their daughter is thinking of pursuing a non-traditional career
as a truck driver:
Its abnormal to be a female truck driver.
Its not good to be abnormal.
Its not good to be a female truck driver.
If we assess these premises for acceptability, we find that in one sense it is true
that its abnormal to be a female truck driverviz., its infrequent. And in a
sense it is true that its not good to be abnormalviz., its not good to be
dysfunctional either physically or mentally. [Just think of those poor folks with
abnormal growths or only one oar in the water!] But is there any connection
between an events being infrequent and its being dysfunctional? No! This
argument is talking apples in the first premise and oranges in the second. Its
only by shifting from one meaning of abnormal to the otherfrom the
statistical sense of abnormal to a medical sensedoes this argument look as if
its premises build a bridge to the conclusion. But now that we see that the
premises are totally unrelated and provide no reason to believe the conclusion,
we see that the argument flunks the relevance condition for cogency. The
argument equivocates on the term abnormal.
Sometimes its so obvious that an argument equivocates that its laughable:
Ive got my money in a bank.
The bank is made of mud.
My money is stuck in mud.
But sometimes its not so obvious, and thats why people get tricked by the
fallacy of equivocation.
Pop quiz: Find and explain the equivocations in the following

cases:
1. The appeal to the Separation of Church and State is bunk!
Theyre already teaching religion in our public schools the
minute they talk about the Big Bang Theory. This is a
theory about the creation of the universe. So theyre already
talking about the Creator. All we want them to do is to let us
talk about Him too in the school curriculum.
2. If I claim that a belief of mine is only a theory, I would hope
that no one would put much stock in itI certainly wouldnt
want to pass that belief off on others as if it were fact. And
73
yet thats exactly what is happening in our schools, when the

theory of evolution is being taught as fact in biology classes.
3. Most liberals say that the death penalty does not deter
murderers. But thats nonsense. There is not a single case in
which a killer who was executed has killed again. This
nation is inundated with victims of killers who, via escape,
furlough, or parole, have lived to kill again. [Adapted from
a letter to the editor, The Dallas Morning News]
4. Science is just as subjective as music is. Professional
science is as full of political intrigue and personal rivalry as
professional music is. Take biogenetics, for instance. Music
and biogenetics entrepreneurs are equally obsessed with
money and power and fame. The average biogeneticist is,
like the average musician, pursuing a project largely
dependent on what people happen to be willing to fund.
5. Background: In the summer of 1978, Andrew Young, the
American ambassador to the United Nations, made the claim
that there were many political prisoners in the U.S. In
support of that claim, Jesse Jackson said the following:
Some may debate the diplomacy and timing used by
Ambassador Young, but the truth and accuracy of his
statement is beyond question. Thousands are in jail because
they are too poor to pay bail bond. Thousands are in jail
because of delayed trials. Thousands are in jail because they
were not tried by a jury of their peers. That is political.
[Adapted from Goviers A Practical Study of Argument]
6. From the Wizard of Id: The prisoner who is about to be
hanged says to his lawyer, I thought I was getting life.
The lawyer replies: I got it reduced.
Pop quiz answers:

Re. 1. Equivocation on the word creation, between a mere
initiating event and a result of someones action.
Re. 2. Equivocation on theory, between merely a contemplated
idea and a proposed description of actual events.
Re. 3. Equivocation on the phrase The death penalty deters
murder, between deterring people from becoming killers
74
by threat of execution and keeping killers from killing

again by actually executing them.
Re. 4. Equivocation on science, between the set of practicing
scientists and the objective method of inquiry scientists
ought to use.
Re. 5. Equivocation on political prisoner, between one who
cannot use the legal system to their advantage and one
who is imprisoned because their political beliefs are
different than the governments.
Re. 6. Equivocation on life sentence, between being confined
to prison for the rest of ones natural life and being
confined to prison until one is killed.
Vagueness
Even though people unfortunately confuse or slur together ambiguity and
vagueness, the two are quite different, and quite different still from simply being
unclear with respect to what someones saying.
Vaguenesswhen the definition of a term does not consist of
necessary and sufficient conditions.
Not all definitions are as clean cut as, for example, the definition of bachelor.
Many [most?] of our terms and concepts lack such analyses. Think about a very
famous example for a minute, an example spotted by Ludwig Wittgenstein [I kid
you not], in his Philosophical Investigations: What is a game? Try to come up
with the necessary and sufficient conditions for being a game.
If we begin thinking about paradigmatic games, we might think that an opponent
is necessary; but then solitaire comes to mind. We might think that equipment is
necessary; but then charades comes to mind. We might think that keeping score
is necessary; but then tag or catch or keep away comes to mind. We might think
that reaching some end state is necessary; and we might be right. But then this is
so vacuously true of any activity, that it just doesnt count as a distinguishing
feature of being a game.
There just doesnt seem to be an interesting necessary condition for being a
game. And if there is no substantive necessary condition, then there will be no
necessary and sufficient conditions constituting the definition of game.
75
Some games share certain features with others, and those others share certain
features with others still, but there is no one feature that all games have in
common. As Wittgenstein suggested, games are interwoven, like fibers forming
a rope; but, just like the rope, games contain no common thread.
And, lo and behold, most of our concepts and terms are like this. Try analyzing
some on your own: e.g., chair, computer, fast, or safe.
Pop quiz: Answer the following questions.
Is vague vague?
Is vague ambiguous?
Is ambiguous vague?
Living with vagueness

As a result of the lack of necessary and sufficient conditions for use of a vague
term, the extension of the term is indeterminate. Wow! What does that
mean?!
Well, the extension of a term is all the things to which the term appropriately
applies. For example, the extension of dog is all the dogs in the universepast,
present, and future.
If a term has necessary and sufficient conditions, these conditions could be used
as an exact checklist, whereby if every single thing in the universe could be
passed by on a conveyer belt, a definite Yes or No answer could be given for
each and every thing as to whether the term applies to it or notits just a matter
of whether or not the thing meets those necessary and sufficient conditions. But
for vague terms, this isnt possible. For vague terms, there will be cases that are
indeterminatecases which will be in principle undecidable.
Think about the terms bald and hairy, and the concepts they stand for. If we
lined everyone up on a conveyor belt and classified them as they passed by, wed
all apply bald to Jesse Ventura and Michael Jordan. And as Santa and the
dudes from ZZ Top passed by, wed all apply the term hairy. But as I came
down the belt, Im going to be presenting you with a problem, because Im no
longer hairy [Im sad to say], but Im not a cue ball like Jesse or Michael [at least
not yet!]. Ive got a fairly big pate, but there is still some fuzz on it, and my sides
are still looking pretty good. So, Im not exactly hairy anymore; Im balding, but
76
not yet quite bald. Im what is called a fuzzy case, literally and figuratively: Im
indeterminate.
What would it be like if bald and hairy were not vague terms? There would
be necessary and sufficient conditions for being bald and for being hairy. There
would be a specific number of hairs fewer than which would mean one is bald,
and a specific number of hairs greater than which would mean one is hairy. But
there simply is no such magic number of hairs! There is no magic hair the
removal of which transforms one into a bald person; and there is no magic hair
the implantation of which transforms one into a hairy person. If there were such
a magic hair, bald and hairy would be determinate; but there isnt, so they
arent. This is the defining feature of a vague term or concept.
Vague, but not arbitrary

How often have you heard the likes of the following, on the evening news, The
Supreme Court today ruled that the Such-and-Such Law was unconstitutional due
to its vagueness?
After a moments thought, however, we now realize that vagueness alone cannot
be a legitimate reason to flush a law down the toilet. If it were, all our laws
would go down that toilet, because all our laws contain vague terms. For
example, all our laws pertaining to searches and seizures require that they be
carried out only under reasonable circumstances, given reasonable suspicions of
a crime. All speed limits hold only under normal, reasonable driving conditions
theoretically, under conditions of glare ice, when the roads are as slick as snot
on a glass doorknob, one could properly be ticketed for speeding when creeping
along at only 10 mph.
Lets take a closer look, then, at this popular criticism of vagueness, and see
where exactly it goes wrong:
1. The law is vague because it contains vague terms; and vague
terms are indeterminate.
2. If the extension of a vague term is indeterminate, it is
arbitrary.
3. If the use of a term is arbitrary, then the law containing it will
be unworkablethe people enforcing it wont know how to
apply it correctly, regularly, and consistently.
4. If a law cant be enforced uniformly, it cant measure up to
constitutional standards of due process, whereby all citizens
are treated equally before the law.
5. A vague law is unconstitutional.
77
Now its easy to see where the argument goes wrong. The second premise is just
plain false: just because a term is vague, and thereby indeterminate, doesnt
entail that the use of the terms is arbitrary.
Keep in mind our paradigm cases of vague terms, bald and hairy. Is it
totally arbitrary whom we may appropriately call bald or hairy? No! Clearly,
Jesse Ventura is bald! Clearly, the guys from ZZ Top are hairy. Just because we
cant draw the line between being bald and being hairy, doesnt mean anything
goes. As long as we are talking about clear-cut cases [sorry about the pun], we
have no problem appropriately identifying the extensions of bald and hairy.
Its only the fuzzy cases that stump us. But thats not our faultthose cases are
in principle indeterminate.
The moral of the story: You must use vague termsget used to it. Just be
sure to use them to refer to clear-cut cases. And with the fuzzy cases, be
intellectually honest and admit that such cases are indeterminatefor you and
for everyone.
Arbitrary, but not vague

Think about the definition of fun or repulsive. Are these vague terms and
concepts? Well, not really. Fun is basically when someone is having an enjoyable experience. Repulsive basically means when someone finds an experience
so undesirable and distressful that they are sickened by it or made to feel uneasy.
This seems to pretty much capture the meanings well enough to qualify as
providing necessary and sufficient conditions for somethings being fun or
repulsive. So these terms dont seem to be vague.
But now, what are the extensions of these terms?
And now were stumped! The extensions of these terms are going to be different
for different peopledifferent people find different things fun or repulsive.
Thats because whats fun or whats repulsive is purely a matter of personal
tastepurely a subjective matter.
As we discussed earlier: With subjective matters, all opinions are equal. So
one persons extensions of these terms are just as legitimate as any other
persons. This means that, unlike with vague terms, there are no clear-cut
cases for the legitimate application of the termno clear-cut cases that everyone
must agreed to.
So, the indeterminacy of the extension of subjective terms is greater than just on
the fuzzy cases. What is fun or repulsive is totally a function of ones likes and
78
dislikes; so whatever agreement there is among peoples extensions of subjective

matters is purely accidental.
Vagueness and subjectivity are very different, but they are often slurred together.
I think a classic example in which this has happened is with obscenity laws.
According to the U.S. Supreme Court, obscenity is defined as a work that:
1. Appeals to the prurient interests of the average person,

2. Describes or depicts sexual conduct in a patently offensive
wayviolating the standards of decency of the community, and
3. Lacks serious literary, artistic, political, or scientific value
when taken as a whole.
But all three of the components of this definition contain elements that are purely
subjective. Whether one finds something of prurient interest, viz., sexually
arousing, is purely subjective. Whether one finds something offensive, viz.,
repulsive or revolting, is purely subjective. And whether something has any
artistic value is purely subjective. U.S. Supreme Court Justice Potter Stewart
thought he was pulling in the reins on what counts as obscene material when he
said, I may not know what it is, but I know when I see it. But he only succeeded in confessing that there are no reins at all on whats obsceneits purely
subjectivefor, similarly, he doesnt know what makes for bad-tasting food
either, he only knows it when he eats it.
So its inaccurate to argue against obscenity laws by calling them vague and
thereby unworkable. If obscene were merely vague, its use would be perfectly
workable, just as with bald and hairy and the zillion other vague terms we use
with ease everyday. Obscenity instead is purely subjective, and thats why any
laws concerning it are unworkable when they are intended to apply to all people
or even just a community of people [unless one is talking about a community
of only one person].
The slippery slope fallacy

We just looked at the problem with treating a subjective concept as if it were a
merely vague concept. Now lets look at a different but related problemthe
problem of treating a vague concept as if it were merely subjective.
This very subtle way of abusing vague terms is often called the slippery slope
argument, but we will soon see why we should call it the slippery slope fallacy.
79
The slippery slope fallacyTo argue from the fact that one
cannot determine the exact boundaries of a terms extension,
to the conclusion that a quite liberal use, or even any
extension, of the term is appropriate.
The best way to fully understand and appreciate how the slippery slope works,
and why its so fallacious, is to closely exam a popular instance of one. The best
example I know of comes from the abortion debate [a prime source of arguments,
both good and bad].
Our current legal practices concerning the right to abortion have their foundation
in the famous Supreme Court case of Roe v. Wade. Basically it ruled that women
have the legal right to an abortion 1) before the fetus is viable [viz.,
approximately the end of the second trimester, when the fetus could be removed
and sustained with state-of-the-art artificial aid] and 2) even after viability, but
then only if the pregnancy poses a significant health risk to the pregnant woman.
This moderate position on the morality and legality of abortion is objected to by
both the pro-life advocate, who would say that all abortions, except possibly
those performed to save the life of the pregnant woman, are morally
impermissible, and the pro-choice advocate, who would say that all abortions,
even those in the third trimester, are morally permissible.
Lets begin with a slippery slope argument that is sometimes used by the prolife advocate. Our first premise is the claim that the newborn has significant
moral status. [Certainly it seems morally objectionable if the woman, moments
after giving birth, were to take the newborn tenderly in her arms and proceed to
squash it on the delivery room floor like a cigarette butt.]
Our second premise consists of a claim that stems from the fact that there was a
very gradual and continuous process of development that resulted in the
formation of the newborn from the original conceptus formed approximately nine
months earlier. This process of development is so gradual that it is difficult if not
impossible to distinguish one days development from the next. In light of this, it
would seem impossible to find a developmental difference that makes an ethical
differencea developmental difference that marks the point at which the fetus
acquired moral status. Certainly one day prior to its birth, the newborn had no
less moral status than it has at birth. For what is the difference of birth? Well, on
the day of birth, the newborn is outside the pregnant woman, while the day prior
to that the newborn was inside her. Does a change in relative location make any
moral difference? Would I lose the right to life as I walk around you, first being
on your right and then on your left? Of course not; so it looks as though we can
claim that one days difference does not make a moral difference for the
newborn. And now we can put these two premises together and draw a
conclusion:
80
1. The newborn has moral status.

2. One day prior to that, it has no less moral status.
3. The newborn minus 1 day has moral status.
And this is why such a very late abortion would be on an ethical par with
infanticide, one might argue.
But in light of the fact that fetal development is so very gradual and continuous,
we can justifiably reiterate [i.e., repeat] Premise #2just as justifiably as we
used it in the first instance. Surely a modicum of difference in tissue
development during a single day makes no moral difference. And so our
argument could be extended as follows:
1.
2.
3.
2.
4.
The newborn has moral status.

One day prior to that, it has no less moral status.
The newborn minus 1 day has moral status.
The (newborn - 1 day) - 1 day has moral status.
Premise #2 seems so justified, that it seems perfectly appropriate to reiterate it

yet again:
1.
2.
3.
2.
4.
2.
5.

The [(newborn - 1 day) - 1 day] - 1 day has moral status.
And again. And again and again and again , until we finally reach the desired
conclusion:
1.
2.
3.
2.
4.
2.
5.
2.

*
The {[( newborn - 1 day) - 1 day] - 1 day . . .} has moral status.
The conceptus has moral status.
Abortion is always as morally wrong as infanticide!
81
If one objects to this conclusion, one must find something wrong with the argument supporting itone must find where the argument is noncogent. So where
does it flunk one of the ARG conditions?
Are the premises acceptable? Premise #1 seems true. Likewise Premise #2.
And the rest of the argument just consists of reiterations of Premise #2.
Are the premises relevant, relevant to the point of adequately grounding the
conclusion? It sure seems sothere seems to be a very smooth, logical
progression to the conclusion.
So the argument appears to meet all the ARG conditions for cogency. Are we
forced to concede, then, that the argument gives us good reason to believe the
conclusion? Not at all!
There is a bundle of problems with this slippery slope argument.
Criticism #1
Lets first demonstrate that the argument is completely unreliable in its ability
to come to a true conclusion.
Think of someone you know who has a full head of hair; lets call them Harry. If
we were to pluck one hair off Harrys head, it wouldnt make any difference, at
least no difference to Harrys still being hairy. We can describe the situation in
the following way:
1. Harry is hairy.
2. One hair less doesnt make one non-hairy. [ouch!]
3. Harry is still hairy.
But now what happens when we reiterate Premise #2 again and again and again,
each time drawing our subconclusions? Well, after many, many reiterations
[OUCH!], Harry is not only ticked-off, hes bald as a cue ball. And were still
drawing what conclusion? That Harrys still hairy!
So now we see that the slippery slope argument in support of the moral impermissibility of abortion is likewise completely unreliable, since its exactly the
same type of argument as this one that erroneously led to the view that Harrys
still hairy despite the fact that weve ripped every hair off his head.
82
Criticism #2
Our argument about Harry enables us to see a deeper flaw in the reasoning of
those who use and defend slippery slope arguments. When their conclusion is
challenged, they respond with the following counter-challenge: In order for you
object to my conclusion, you must find something wrong with my
premisesfind the false or unacceptable premise! And of course we cant.
But then do we have to find the false premise, in order to object to the slippery
slope argument?! We certainly do not, and heres why.
We know that the slippery slope argument that concludes that Harry is still hairy
is not cogent. But if someone were to demand that we find where in that
argument the premises went false, it would be a completely asinine demandit
would be the demand that we find the magic hair, the removal of which
transformed Harry from being hairy to being no longer hairy. But, of
course, there is no such magic hair! To make such a request would indicate
that one just doesnt really know what hairiness is, or at least that one fails to
appreciate that hairiness is a vague concept.
And now we see the problem with the demand that we must find the false
premise in order for us to object to the slippery slope argument that all abortions
are as morally impermissible. There simply is no such magic day of fetal
development that transforms the fetus into a being with the moral status of an
infant. To think there is such a magic day simply betrays a poor understanding of
what it is to have moral status, or at least indicates a failure to appreciate that
this too is a vague concept.
Its this fundamental misunderstanding of vague concepts that is the root of
all slippery slope fallacies. In essence, one is claiming Unless you can draw
the exact line between the appropriate use of this vague term and its
inappropriate use, you must accept my very liberal use of it. And now we know
why this is completely in error:
The inability to draw such a line never legitimizes the inappropriate use of a
vague term.
Criticism #3
Take another look at the argument:
83
1.
2.
3.
2.
4.
2.
5.
2.

*
The {[( newborn - 1 day) - 1 day] - 1 day . . .} has moral status.
The conceptus has moral status.
Abortion is always as morally wrong as infanticide!
What are you itching to ask its author? You were on such a roll, reiterating
Premise #2 again and again and again; so whyd you stop?! The answer, of
course, is that the author stopped because they found the conclusion they wanted
to find. But they didnt stop where the argument actually leads.
If we were to reiterate Premise #2 again, we would draw the conclusion that the
sperm and the egg that formed the conceptus had moral status. [After all, what is
the significant difference between a right glove and a left glove and that pair of
gloves?] We would need a very strong argument for treating conception as the
magic day that transforms organic material into a being with moral status.
Without such an argument, the conceptus is no more a being with moral status
than an acorn is an oak tree [J.J. Thomson, A Defense of Abortion].
And, on the basis of the conclusion that the sperm and the egg have the same
moral status as does an infant, one could draw the conclusion that male masturbation is mass murder [considering there are millions and millions of sperm in
the average males ejaculate]. Whoa! Thats crazy!
This demonstrates a third problem with slippery slope arguments: They have no
brakes. They shoot right past the conclusion one wants and equally-well support
patently absurd conclusions. Once again, slippery slope arguments are loose
canonstotally unreliable.
Criticism #4
A fourth problem with slippery slope arguments is that they work both waysif
you can use one to get to your favorite conclusion, your opponent can use it to
slide to the opposite conclusion. Lets see how this happens.
84
We can start with the obvious truth that the sperm and the egg are not moral
entities on a par with infants. And then we add, as our second premise, that one
day more does not make a significant moral difference [again, what is the
significant difference between a right glove and a left glove and that pair of
gloves?].
After reiterating this second premise again and again and again, we get the
following:
1.
2.
3.
2.
4.
2.
5.
The sperm-and-the-egg has no moral status.

One day later, it has no greater moral status.
The sperm-and-the-egg + 1 day has no moral status.
The (sperm-and-the-egg + 1 day) + 1 day has no moral status.
The [(sperm-and-the-egg + 1 day) + 1 day] + 1 day has no moral
status.
2. One day later, it has no greater moral status.
*
The {[( sperm-and-the-egg + 1 day) + 1 day] + 1 day . . .} has no moral status.
The fetus throughout the pregnancy has no moral status.
Abortion is always morally permissible.
One cant really object to having their slippery slope argument reversed on them
like thisits their own argument, after all. Using an argument puts it in the
public domain; one cant say the likes of This argument is good, but only when I
use it.
Pop quiz: Critically review this slippery slope argument above:
How can you get Michael Jordan looking like a member

of Z.Z. Top and keep drawing the conclusion that hes
still bald?
Look out! No brakes! Lets say you really ticked-off

your mother; how could she kill you and conclude that it
was a perfectly morally permissible late abortion?
The moral of the story: Dont use slippery slope arguments! Thats why they
are slippery slope fallacies!
85
So why are slippery slope arguments so popular, if theyre this bad?

Well, slippery slope arguments come so disguised that people dont realize that
theyre using them or falling for them. Youll never see anyone lay out their
slippery slope arguments with the degree of detail that I have in my discussion
above. They are always much more condensed. But as condensed as they are, all
slippery slope arguments can be formatted in the way Ive been presenting them.
And just how condensed and disguised can the slippery slope fallacy get? Well,
how often have you heard someone say Yes, but where do we draw the line?
You hear this whine almost every daymaybe even coming from yourself!
We cant allow this! Because, if we did, where would we draw the line? Wed
have to allow just about anything!
If we prohibited that, where would we draw the line? Wed soon be forced to
prohibit just about everything. No, lets not go down that slippery slope!
Basically one is arguing that since one cant draw the exact line between an
initially appropriate application of a vague term or concept and another obviously
inappropriate application of it, the inappropriate application would have to be
condoned if the appropriate one is; and one is unwilling to condone that
inappropriate application, so one is forced to refuse the initial appropriate
application.
This is the most popular form of the slippery slope fallacy: Rather than riding
the slippery slope to its supposed end, one refuses to take its first step, i.e., one
refuses to accept its first premise no matter how obviously true that premise
may be.
But always remember: You dont need to draw the line in order to
accurately and reasonably use vague terms to talk about clear-cut cases!
A slippery story: On late-night television one night, Martin

Mull was discussing how he was trying to stop smoking. He was
smoking about 20 cigarettes a day and thought, If I can get by
on 20, then I could get by on 19. And he found that he could.
Then he thought, If I can get by on 19, I can get by on 18.
And that worked too. Finally he was down to only 6 cigarettes a
day. When he told a friend of his success, his friend said,
Congrats! Youre down to 6, and thats not going to kill ya.
Mull thought, Youre right. And if 6 wont kill me, another one
86
wont either. And after a couple of weeks, he was back to 20 a

day.
Pop quiz: Think about various issues in the news and see if you
can spot slippery slope fallacies as they crop up during debates
in ethics and social policy. Here are some topics to get you
started.
gun control
alcohol consumption
religious freedom
freedom of speech
any safety issue

To practice and test your understanding of the material discussed so far, launch
the CT software and click the exercise list button for Whad Ya Know About
CT? This module has four levels. Select Level 1. Whad Ya Know consists of
True-False questions designed to review and assess your understanding of basic
concepts. If you dont do much better than a coin-flip, you need to do some
serious rereading. Keep track of the ones you miss and figure out why your
answers were wrong and why the computers were right. [Now that youre using
an etext, reviewing concepts is a snapjust do a Find on the concepts youre
screwing up on.]
87
Acceptability
6. ACCEPTABILITY
Lets start our detailed study of the ARG conditions for cogency, beginning with
the requirement that the premises of an argument be acceptable.
What is acceptability? First draft

What is it for our premises to be acceptable? Or, another way of phrasing this
question: What is it for our premises to be justified or rational to believe?
How about this? Acceptability is such that each premise is supported by a
cogent argument.
This is an interesting attempt at understanding what it is for our premises to be
acceptable. But, of course, thats just Minnesotan for saying I really dont like
it. Seeing why this account wont work will help us discover what having
acceptable or justified premises is in fact all about.
Lets say we have an argument consisting of premises P and conclusion C. We
could represent it as follows:
P
C
If premises P need to be supported by a cogent argument, in order to be acceptable, then P needs to be the subconclusion C* of an argument with acceptable
premises P*. That would leave us with an argument looking something like this:
P*
C* = P
C
But now, if premises P* are to be acceptable, they too must be the subconclusion
C** of a cogent argument, viz., an argument with acceptable premises P**. So,
that leaves us with an argument looking something like this:
. . . . P**
C** = P*
C* = P
C
And we are off and running on an infinite regress of needing further acceptable premises for our acceptable premises for our acceptable premises, and on
88
Acceptability
and on and on. And since this regress is infinite, it never gets completed; which
means that we could never have acceptable premises; which means that none of
our arguments is ever cogent.
But hold on there! Certainly not all of our arguments suck! Some of our arguments are cogent; so, some of our premises must be acceptable.
So this particular analysis of the acceptability requirement must be mistaken.
What is acceptability? Second draft

In order to avoid getting into an infinite regress in our attempts to have
acceptable premises, lets try a different analysis of acceptability: Acceptability
is such that each premise is either self-evident or [eventually] supported by a
premise that is self-evident.
Notice how this analysis of acceptability would end the regress, by anchoring all
our arguments ultimately on premises that are self-evident. This theory of acceptability or justification has come to be known as foundationalism; the attempt to
find the ultimate foundations for justifying our various beliefs.
Lets try this analysis on for size by applying it to the following example of an
argument:
The computer that Im currently using is a Mac, since it has an
Apple logo on the top of the screen and Macintosh PowerBook
G3 inscribed on the bottom of the screen.
We could more formally lay this argument out, including its implicit premise, as
follows:
P1. The computer that Im currently using has an Apple logo on
the top of the screen and Macintosh PowerBook G3
inscribed on the bottom.
P2. If a computer has an Apple logo on the top of the screen and
Macintosh PowerBook G3 inscribed on the bottom, its
[in all probability] a Mac.
C. The computer that Im currently using is [in all probability]
a Mac.
Are these premises acceptable under the foundationalist analysis of acceptability;
i.e., are these premises either self-evident or supported by self-evident premises?
89
Acceptability
What is it for our premises or beliefs to be self-evident, anyway? Well, thats a

little difficult to say, even for foundationalists; but its basically to be acceptable,
but not by virtue of any other beliefs. A self-evident belief wears its justification
on its own sleeve, so to speak. Lets see what this is supposedly like by looking
at our example some more.
Do I have a further reason for believing P1? Well, yes; Ive got all sorts of
perceptual reasons for believing P1: The thing in front of me looks like a computer, the top of the screen appears to have a white logo thats apple shaped, and
the white squiggles on the bottom of the screen look like the words Macintosh
PowerBook G3. And, generally, when things look like computers, they are
computers [as opposed to being cleverly disguised jewelry boxes, for instance].
So, on further examination, weve discovered two premises in support of P1,
which means that it was not self-evident:
P1*. The thing in front of me looks like a computer, the top of
the screen appears to have a white logo thats apple
shaped, and the white squiggles on the bottom of the
screen look like Macintosh PowerBook G3.
P1**. Generally, when things look like computers, logos, and
words, they are computers, logos, and words.
P1.
[In all probability] the computer that Im currently using

has an Apple logo on the top of the screen and
Macintosh PowerBook G3 inscribed on the bottom.
Are these premises acceptable under the foundationalist analysis of acceptability;

i.e., are these premises self-evident?
Well, P1* sure seems self-evident. After all, whats my reason for believing that
the thing in front of me looks like a computer, and that the top of the screen
appears to have a white apple-shaped logo, and that the white squiggles on the
bottom of the screen look like Macintosh PowerBook G3? Simply the fact
that it all looks this way to me!
The fact that Im forced to sort of spin my wheels here in my attempts to give a
reason for my belief, indicates [according to many foundationalists] that Ive
arrived at a self-evident foundational perceptual belief.
On the other hand, is P1** similarly self-evident? Certainly not!
90
Acceptability
My belief that, generally, when things look like computers and logos and words,
they are real computers and logos and words, is based on many, many other
beliefsperceptual beliefs and various other kinds of background beliefs. For
example:
Beliefs involving past experiences with computers, logos,

and words.
Beliefs about the good perceptual conditions Im currently

in.
Beliefs involving past experiences with objects generally.
Background beliefs about the nature of objects, as generally

being as they appear.
Background beliefs about how, if things werent generally

how they appear, we wouldnt be here, getting around so
well.
Based on this simple example alone, then, we can easily see that there will be no
such thing as a foundation of self-evident beliefs for us to use as our sole or
ultimate source of acceptability.
So, what is acceptabilityreally?! OK, this is my final answer.

We get a rough but helpful picture of what acceptability is all about by taking
another look at the above list of what justifies my simple little belief in
P1**that, generally, when things look like computers and logos and words,
they are really computers and logos and words.
All the beliefs in this list [and many, many others!] are reasons that support
and justify my belief in P1**. My huge stock of perceptual and background
beliefsabout my current experiences, my past experiences, the nature of
objects, how they appear, and the general reliability of my experiencesforms a
complex fabric of beliefs that support my belief in P1**.
This mutual support among the many beliefs justifying my belief in P1** is
similar to the relations among the many threads in a piece of fabriceach thread
is interwoven with and supports its neighbors and its neighbors neighbors, etc.
This is commonly called the coherence theory of acceptability or justification:
A belief is acceptable or justified to the degree to which it coheres with our other
beliefs.
91
Acceptability
The coherence of a belief can be illustrated by how disruptive it would be to our

overall stock of beliefs to give it up. For example, if I were to give up my belief
in P1** that, generally, when things look like computers and logos and words
they are computers and logos and words, I would be at a complete loss for an
explanation as to why its looking so much like theres a computer with a logo
and an inscription on it front of me now. And if my eyes are so deceiving me
now, in such good lighting conditions, then I can no longer explain why my
experiences are to be trusted at other times. And if my experiences arent
generally reliable, then its a mystery how Im getting around in the world so
well, since I might well be wrong about so many other beliefs about how the
world is, based on how it appears. My whole fabric of beliefs about the world
begins unraveling without my belief in P1**. My belief in P1** provides me
with a more coherent stock of beliefs than I would have if I were to give up that
belief.
Acceptability or justification of beliefs, then, is this relation of mutual support
among our beliefsmy stock of beliefs acts as a source of reasons for my belief
in P1**, and my belief in P1** may in turn add to my many reasons in support of
those various other background beliefs.
In light of this coherence view, we see that the most defensible analysis of the
acceptability condition is basically this:
Acceptability of premises is such that one has more reason to
believe the premises than to not believe them.
The relativity of acceptability

If the acceptability of premises is a function of whether or not one has more
reason to believe them than not to believe them, this makes acceptability a
function of ones background beliefs that are acting as reasons for believing those
premises.
We all have different background beliefs, since we all have led different lives,
experiencing and learning different things. So, indeed the acceptability of
premises is going to be relative to the individual recipient of the argument.
This is not to say that acceptability is arbitrary. Arbitrary beliefs are beliefs held
for no reason. Acceptable beliefs are held for reasonsreasons one has.
Because the acceptability of premises is relative to the recipient of the argument,
you must always Take your audience into account. Its just common sensea
premise you have every reason at your disposal to believe is true, might well be a
92
Acceptability
premise that a child, for example, has absolutely no reason at their disposal to
think is true. You have much more experience and background knowledge than
the child does, and you can call upon that background as justification for
believing your premises; the child cant.
So, if you want your arguments to meet the acceptability condition for a wider
audience than just yourself, your strategy should be to make your premises as
uncontroversial and as unquestionable as possible. Use premises that do not
presuppose extraordinary amounts of background knowledge.
This presents a challenge at times: You want to reach the widest possible
audience with your arguments, so you provide premises that even those with a
very moderate degree of background knowledge would find acceptable.
But you then run the risk of dummying down your argument to the point to
which your premises are so trivially true that youre insulting your audience by
talking to them as if they were children.
Use the Goldilocks strategy again as you pick your premises: Not too esoteric;
not too platitudinous; but rather, juuuust right for your audience.
The test for acceptability

Now that weve clarified what acceptability is, its time to develop the habit o f
requiring it of our premises and the premises of others.
When you either construct an argument or received one, ask the following
question: Are these premises acceptabledo I have more reason to think
these premises are true than I have for thinking they are false?
If the answer is No, the only rational thing to do is to withhold your belief in
the premises and reject the argument.
If you actually have less reason to belief the premises true than you have for
believing them false, then the rational thing to do is to believe that the premises
are false.
Being conscientious about the acceptability of premises, no matter whose
premises they are, is a large part of what puts the critical in critical thinking.
Its a Show me the reasons! point of viewa point of view we all need to
cultivate a bit more, to help us in our pursuit of the truth.
93
Acceptability
Subarguments
You might find your premises acceptable, by virtue of your background beliefs
that are acting as your reasons for thinking your premises are true. But if your
audience doesnt share that background, you must provide them with it. And
thats how subarguments get generated. You knowwhat we studied in the
Anatomy of an Argumentwhen conclusions become premises for further
conclusions.
Sometimes we need to provide reasons in support of our reasons for our
conclusions. And sometimes, we even need to provide reasons for believing
those reasons. And sometimes,well, you get the idea. Whatever it takes to
get your audience to finally have more reason to believe your conclusion than not
to believe it, thats what you need to provide.
The appeal to authority

Some years ago, on a camping trip in the pine woods of northern
Michigan, my friend Don brought along a copy of an outdoor
cookbook that appeared on the best-seller lists at the time. This
book contained many ingenious and easy-sounding recipes; one
that Don especially wanted to try was called Breakfast in a
Paper Bag. According to this recipe, you could take a small
paper lunch sack, put strips of bacon in the bottom, break an egg
into the sack on top of the bacon, fold down the top of the sack,
push a stick through the fold, hold the sack over hot coals, and
cook the bacon and egg in the sack in about ten minutes.
I watched as Don followed the directions exactly. Both he and I
remarked that we would naturally have thought the sack would
burn; the recipe, however, declared, grease will coat the bottom
of the bag as it cooks. Somehow we both took this to mean that
the grease, counterintuitively, actually made the bag less likely
to burn. Marveling at the who would have guessed magic of
it, we picked a good spot in the hot coals of our campfire, and
Don held the sack above them. We watched. In a second and a
half, the bag burst into leaping flames. Don was yelling for help,
waving the bag around trying to extinguish it, scattering egg yolk
and smoldering strips of bacon and flaming paper into the
combustible pines while people at adjoining campfires stared in
horror and wondered what they should do.
The wild figures that the burning breakfast described in midair as
Don waved the stick, the look of outraged, imbecile shock
94
Acceptability
reflected on our facesthose are images that stay with me. I

replay the incident often in my mind. It is like a parable.
Because a book told us to, we attempted to use greased paper as
a frying pan on an open fire. For all I know, the trick is possible
if you do it just so; we never repeated the experiment. But to me
the incident illustrates a larger truth about our species when it
ventures out of doors. We go forth in abundant ignorance, nearblind with fantasy, witlessly trusting words on a page or a tip a
guy wed never met before gave us at a sporting-goods counter
in a giant discount store. About half of the time, the faith that
leads us into the outdoors is based in advice that is half-baked,
made-up, hypothetical, uninformed, spurious, or deliberately,
heedlessly bad. [From Trust Me. In These Parts, Hot Dogs
Actually Repel Bears, Ian Frazier, Outside, Dec. 1999.]
Our most frequent and necessary source of acceptable premises, second only to
our own sense perceptions, is the appeal to authorityexperts and witnesses.
If you didnt agree with me on this claim, you wouldnt be bothering to read this
etext right now. In fact, you wouldnt bother reading anything for the purposes
of learning something. You wouldnt bother asking others for even the smallest
bit of information, such as what time it is. Youd rely on your own eyes and ears
as your sole source of knowledge. And, youd be screwed!
And yet all too often, when we do rely on the experts or authorities, were
screwedas the story above indicates. The question is: When is an appeal to
authority acceptable and when is it not?
Heres a typical appeal to authority, the kind we make many times every day:
Just last night on the news, there was a bit on how researchers
just published their findings that eating fish once a week reduces
the risk of heart attack. And, if thats the case, then I should start
eating fish more. So, Ill pick up some fish today, along with my
chips and brats, at the grocery store.
Theres a basic format underlying every appeal to authority such as this:
P1. Dr. Expert said X is the case.
P2. If X is the case, then Y.
C. Y.
Take a closer look at premise P1. P1 is actually a condensed subargument,
with the following implicit premise and conclusion.
95
Acceptability

P*. Dr. Expert wouldnt say X is the case,
unless X is the case.
C*. X is the case.
So, the typical appeal to authority is actually quite elaborate:
P*. Dr. Expert wouldnt say X is the case,
unless X is the case.
C*. X is the case.
P2. If X is the case, then Y.
C. Y.
Now that weve laid out explicitly the anatomy of an appeal to authority, it
becomes evident that the question to keep in mind when making any such appeal
is: Are P1 and P* likely true?!
Heres a checklist to see if P1 and P* are acceptable, i.e., for seeing if the
appeal to authority is acceptable.
Appeal to Authority Checklist
1. The expert appealed to in fact made the claim.
2. The experts claim falls within an objective area of
subject matter.
3. The expert is an expert on that subject matter.
4. Other experts in that area generally agree with this
experts claim.
5. The expert is a reliable, credible person when
commenting on this subject matter.
Fallacy of appeal to false authority

When any of these considerations fails, one commits the informal fallacy of
appeal to false authority.
Appeal to false authoritywhen a statement is believed
because it is expressed by an expert claiming to have
informed, privileged knowledge; however, one has reason to
96
Acceptability
believe that the expert is unreliable, viz., it is quite possible

that they would express the statement even if it were not
true.
I suppose it would be more accurate to call this fallacy the appeal to questionable authority, because no matter how bad the authority might flunk the considerations on our checklist, they still might be accidentally correct in their
claim. Even an authority like the Magic 8-Ball toy is right once in a while!
But this fallacy has a long history of being called appeal to false authority, so
lets go along with it, so that more people will know what were talking about
when we make the charge of committing this fallacy.
The appeal to false authority is fairly straightforward, but let me just make a few
comments about flunking the considerations on our checklist:
Re. 1. So often people are misquoted, quoted out of context, or
ascribed just plain fabrications. One really must make
reasonably sure that the authority cited is accurately represented.
Re. 2. Make sure the topic being discussed by the authority is an
objective issue and not a merely subjective issue. For if its
merely a subjective issuea mere matter of tastethen, as we
discussed earlier, there simply are no authorities or experts. All
opinions are equal.
Re. 3. Being an expert on one subject doesnt mean ones an expert on
other subjects. Being an authority in the field of science, for
example, does not make one an authority on matters of ethics or
critical thinking, and vice versa. And being a celebrity in one
area does not make one an expert in another.
Re. 4. There are people who still think that the Earth is flat. There is
even a Flat Earth Society, with a scholarly newsletter. There are
authorities within this group who labor to tweak the data and the
laws of physics to accommodate the Flat Earth hypothesis. For
virtually every weird view one can imagine, there is a so-called
authority advocating it [and who would gladly advocate it
publicly for a consultants fee, e.g., in a court of law, as one sides
expert witness]. But calling a person an expert doesnt make
them one. The experts claim must be the consensus view in an
objective area or discipline, within which the expert is
recognized as an expert.
Re. 5. The expert is relied upon to be telling the truth. To the best of
our knowledge, then, we must make sure that they would not be
97
Acceptability
saying what theyre saying unless it were true. With this in

mind, one must consider whether there are any other significantly
possible alternative explanations for the experts claiming what
they do, other than just a desire to express the truth. Might they
plausibly have another motive instead of just truth telling? If
they do, then they might not be speaking as an authority, but
rather merely as one who stands to make some personal gain.
Make sure the expert is impartial.
This is why, for example, its so worthless to appeal to Coca-Cola
researchers findings to support the claim that Coke is a better
cola than Pepsi, or to appeal to Pepsi researchers findings to
support the claim that Pepsi is a better cola than Coke. [BTW:
This really happened. And they sued each other, and both lost,
and both were forced by the FTC to stop making such goofy
claims in their ads about being better than the other.]
Whos an authority?
Sometimes the so-called experts arent the real experts and the real experts are
the least recognizable. Being able to know which are which often presupposes a
lot of background knowledge that the average person doesnt have. Catch-22!
For example, during the cold war, the so-called experts in the State Department
negotiating nuclear disarmament may not have been the real experts at all; we
just lucked out because of the USSRs failing budget. The real experts might
well have been the math geeks in decision theory, who had long ago proved that
the tit-for-tat negotiating strategy would result in total disarmament, assuming
merely that our two countries cared only for themselves and were not suicidal.
But who knew?!
The appeal to authority for one person is often the appeal to false authority for
another. For example, many people appeal to their religious leaders and religious
texts as guides on ethical matters. While these may be religious authorities, they
cannot function as authorities on ethics, for the general public. The general rule
about adopting premises is especially important when applied to appeals to
authority: Keep your audiences background beliefs in mind, so that your
premises are acceptable to the widest possible audience.
Unacceptable!
Seven of Nine never seemed to pass up the opportunity to make this charge on
Star Trek Voyager. Lets examine four of the various ways that premises can be
unacceptablewhen those premises are:
98
Acceptability
just plain false
inconsistent or imply an inconsistency
dependent on an implicit false assumption
simply no more believable than the conclusion
Premises that are just plain false

The most obvious way of blowing the acceptability condition is when a crucial
premise of an argument is flamingly false. For example, if someone were to
argue that:
Armed revolution against the U.S. Government is completely
justified on the part of U.S. citizens, because the Government
has struck a deal with aliens and is assisting these aliens in the
abduction of U.S. citizens for their use as hosts in the
regeneration process of the aliensall of which is well
documented on the X Files.
Most often, premises flunking the acceptability condition in this way are not
quite this bizarre. Sometimes a blatantly false premise is unknowingly advanced
by someone who is misinformed or out of the informational loop, but who
should have known better. As unintended as such a lapse in rationality may be,
it doesnt excuse ones premises from the acceptability requirement. One is
excused from having false premises only if one could not have known better.
Again, this is what makes acceptability relative to a persons background beliefs,
but not arbitrary.
So, as you survey arguments, ask yourself, Are all the premises true; to the best
of my knowledge, are any of these premises false?
Premises that are inconsistent or imply an inconsistency

Sometimes ones premises flunk the acceptability condition because they are
inconsistent or imply an inconsistency. The charge of being inconsistent is made
more often than people are actually guilty of it. Ive noticed that, often, when an
argument is called inconsistent, the person making the criticism merely means
that they cant quite follow the argument very welltheir complaint is merely
99
Acceptability
that they dont know what the argument is saying, rather than that they know that
the argument is both saying something and denying it simultaneously.
To assert and deny a statement simultaneously is a huge blunder. But it does
happen, so you have to be on the alert for it. Heres an example to illustrate [see
if you can find the contradiction].
[The problem with having pornography establishments in ones
community] concerns the tone of the society, the mode, or to use
terms that have perhaps greater currency, the style and quality of
life, now and in the future. A man may be entitled to read an
obscene book in his room, or expose himself indecently there....
We should protect his privacy. But if he demands a right to
obtain the books and pictures he wants in the market and to
foregather in public placesdiscreet, if you will, but accessible
to allwith others who share his tastes, then to grant him his
right is to affect the world about the rest of us, and to impinge
on others privacies. Even supposing that each of us can, if he
wishes, effectively avert the eye and stop the ear (which, in truth,
we cannot), what is commonly read and seen and heard and done
intrudes upon us all, want it or not. [Bickel, quoted in Majority
Opinion in Paris Adult Theatre I v. Slaton, Chief Justice Warren
Burger]
This was a rather ingenious argument for the States right to censor pornography
as it is shown in places of public accommodation, such as theaters: Establish
such pornography as a form of publicly displayed pornography, on a par with
pornographic billboards that have a captive audience. And virtually everyone
agrees with the States right to prohibit pornography that one is forced to see; so
therefore, everyone should agree to prohibiting pornography as shown in places
of public accommodation.
The problem with this argument is that there is a contradiction within the
premises: The pornography in places of public accommodation is called
discreet and yet shown in such a way that one cannot avert the eyeviz., the
author is saying one is not forced to look at it and forced to look at it.
Pop quiz: Find the contradiction:

There is no evidence for thinking that viewing even
violent pornography causes men to increase their
disposition to aggress against women. So there is no
reason to think that the censorship of pornography is
justified in the name of protecting the rights of women.
100
Acceptability
What the pornography industry should do is something

like the alcoholic beverage industry does. It donates to
MADDMothers Against Drunk Driving. The porn
industry should donate to shelters for abused women.
Another pop quiz: What conclusions follow from a

contradiction?
Hint: If youre unsure, go back and review our study of
the rather technical notion of validity.
While people seldom use contradictory premises within a single argument, they
quite often use contradictory premises across multiple arguments. Sometimes
this is unintentional absentmindedness, but sometimes it is done purely for the
devious purpose of rationalizing conclusions, as opposed to justifying them.
Dealing with this kind of intellectual dishonesty is very frustrating.
For example, I recently had to work with a committee that was deciding which
courses should be categorized as Different Culture for the Universitys new
general education program. All and only courses dedicated to the study of
different cultures were to admitted into this categorythats the only requirement. Classical Philosophy had always been in this category, and the committee
agreed that ancient Greek civilization was drastically different from our own in
terms of beliefs, values, and social and political practicesthe admitted core
aspects of a culture.
The committee agreed that African Americans do not differ markedly in these
respects; in fact, it would be rather racist to think so. The committee also agreed
that women do not differ markedly in these respects either; in fact, it would be
rather sexist to think so. While having a different language is sometimes an
indicator of a different culture, the committee agreed that, due to the
globalization of the economy and most other aspects of our lives, the U.S.,
Europe, and Latin and South America differ little overall in terms of beliefs,
values, etc.
From among the courses on the following topics, then, which should have been
granted Different Culture designation by the committee?
1.
2.
3.
4.
Courses on African Americans

Courses on women
Spanish, German, and French
Classical Philosophy
101
Acceptability
Here are the committees decisions and the reasons given:

Re. 1.
Re. 2.
Re. 3.
Re. 4.
Granted, because they are about African Americans

Granted, because they are about women
Granted, because they teach a foreign language
Rejected, because it is not about African Americans or
women and does not teach a foreign language
When all the inconsistencies of these arguments were pointed out [can you count
them all?], the committee gladly admitted that their decisions were inconsistent,
as if to say But thats not a flaw; thats a feature. Indeed, more and more I find
people taking pride in their inconsistencies; in fact, they regularly quote Walt
Whitmans Leaves of Grass, as their contradictions are pointed out to them: Do
I contradict myself? Very well thenI contradict myself.
What can you do with such a person who refuses to honor the most fundamental
law of thoughtconsistency? There is really nothing you can do. Just walk
away, recommends Aristotle: We cannot be expected to argue with such a person, any more than we can be expected to convince a vegetable. And so, walk
away is just what I did.
Pop quiz: One committee member pointed out that: The new
title for the different culture category is actually Multicultural
Perspectives; so, since Classical Philosophy studies only one
culture, instead of multiple ones, it should be rejected. I had to
laugh out loud. Why?
Premises that involve an implicit false assumption

Another way that ones premises can be unacceptable is by involving or
depending on a false assumption that is crucial to the task of grounding the
conclusion. For example, say that someone is arguing that visiting Rome is a bad
idea, because the traffic is unbearable, finding gas stations is impossible, parking
is almost nonexistent, and car theft is a frequent occurrence. These reasons
against visiting Rome, however, presuppose that one is going to be driving in
Rome, and that is in all probability falseone usually uses public transit or taxis
[unless youre into extreme sports!].
Advertising tries to persuade people into believing that various products would
be good to buy, but sometimes it does so by getting its audience to make false
assumptions. Looking at some famous cases will illustrate this and remind us to
be on our guard when someones making a sales pitch.
102
Acceptability
A classic is a 1970 magazine ad for Volvo. The ad claimed that 9 out of every
10 Volvos registered here in the last eleven years are still on the road. This
naturally set up the inference that Volvos are pretty darn reliable. And since
most people in the market for a car are searching for a reliable one, the practical
conclusion to draw is I better seriously consider buying a Volvo. But this
argument rested on a false assumption; an assumption that was almost unavoidable in light of how the ad was worded.
The false assumption was that Volvos were sold in the U.S. for quite some time.
They werent. They were sold in the U.S. as of 1959. They had only been sold
in the U.S. for eleven years, with the vast majority of them sold in those years
just prior to the appearance of the ad.
And now that were wise to the ad and stop relying on the false assumption, we
realize that Volvos record of reliability was no better than the competitions.
Sometimes, the arguments false assumption is so well hidden that one could
barely be expected to find it. A great example of this is a television ad that aired
during the late 1980s. [I remember watching the ad, but I prefer to mention no
names.] The ad consisted of a long description of how brand MH coffee was
overwhelmingly preferred over brand F in a huge double-blind taste test that was
carried out all over the country at various locations such as shopping malls. The
inference that was so natural to make at the end of this ad was, If so many
people preferred brand MH over brand F, then I should buy MH next time Im at
the grocery store shopping for coffee, since I want the better-tasting coffee.
The false assumption working behind this ad is that the coffee used in the taste
test was the same as the shopper finds on the grocery store shelves. While that
assumption held true for the brand F, it was not true for the MH coffee. The
corporation owning brand MH bought gourmet coffee to use as its coffee in the
taste test. No wonder it was preferred!
Pop quiz: Name that fallacy:

The owners of MH coffee later argued as follows: We made no
false claims in our ad. Indeed, we did use MH coffee in our taste
tests. After all, we bought that coffee legally; it thereby becomes
MH coffee.
103
Acceptability
The trick revealedlogical implication vs. mere intimation

Lets examine some more ads. [These examples, and many others, are rounded
up in a great book by Ivan Preston, The Tangled Web They Weave: Truth, Falsity, and Advertisers.]
A pain reliever, Efficin, claimed that it contains no aspirin.
Those prone to suffering side-effects from aspirin naturally
assumed that if Efficin contained no aspirin, it would not have
aspirins side-effects. False: Efficin was chemically so like
aspirin that it had identical side-effects.
Black Flag roach killer claimed that it killed roaches the other
leading brand couldnt. The natural assumption was that the ad
was talking about average roaches. False: The roaches tested
were specially bred to be resistant to the competition.
Kroger food stores advertised its Price Patrol, that would go to
its competitors stores and compare prices of a shopping cart of
items. The natural assumption was that the shopping cart
consisted of the average collection that the average shopper
would gather. False: It consisted of only those items Kroger had
on special sale for the week.
In their ads, products claim to whiten teeth or freshen breath, etc. The natural
assumption is to think that these ads are claiming something special and unique
about their products, e.g., if the product claims to be fat-free, that must mean the
competition is not. False: Just keep in mind that Apple could claim that its
computers are fat-free and contain no additives!
Advertisers make it so easy for their audience to make the false assumptions
necessary to draw the favorable conclusions sponsors want. Once you see how
they do it, though, maybe you wont be so prone to make such false assumptions
in the future:
The makers of Efficin, for example, didnt s a y that Efficin was medically
different from aspirin, and nothing they said logically entailed that Efficin was
medically different from aspirin. So they did not make or imply a false
statement. But they sure intimated or suggested that Efficin was medically
different from aspirin. After all, why did they state that Efficin contained no
aspirin unless it was relevant information to the consumer (and not just the nitpicking chemist)?!
104
Acceptability
This illustrates a distinction between logical implication and mere intimation

the distinction between what a statement says or implies and what is merely
suggested by the saying of the statement. [FYI: This distinction was first noted
by Paul Grice, as the difference between statement implication and conversational implication.]
Heres another example of how people use intimation to get others to make false
assumptions. Lets say I know full well that Mark is the culprit. When asked if
Mark did it, however, I reply, Well, I know this much anywayit was either
Bob or Mark or Mary. Have I lied? No, but I did probably get my audience to
falsely believe that I think these three are all equally likely candidates as the
culprit. And thats the trick of intimating as opposed to saying or implyingto
get people falsely assuming what they really have no good reason to believe.
Pretty sneaky, isnt it? But without it, many professions would look a lot
differentadvertising, lawyering, any others?
Premises that are simply no more believable than the conclusion

Perhaps the most frequent way in which ones premises can be unacceptable is
when they are simply no more believable than the conclusion is in the first place
the premises give you no more reason for believing the conclusion than you
had prior to receiving them.
If the premises are desperately in need of reasons to believe them, they can
hardly function as reasons to believe their conclusion. Arguments are, after all,
supposed to function as conduits of acceptability from premises to conclusions,
and if the premises have no more acceptability than the conclusion, they cant
pass it on. [Especially when the argument is an inductive argument, with its
inherently imperfect ability to preserve acceptability.]
How often have you heard the following argument: Abortion is wrong, because
it is the killing of an innocent person? This argument fails miserably, however,
because the whole issue of abortion revolves around whether or not the fetus is a
person like you and I, with rights such as the right to life. This argument just
asserts that the fetus is a person. So the argument just pushes the controversy
back one step.
If youre ever in this predicamentif you find yourself using premises that are
no more reasonable to believe than your conclusion was in the first placethen
youre going to have to supply your audience with reasons to believe your
premises. Youre going to have to supply a subargument for your premises. And
maybe this subargument will be in need of its own subargument. As we
discussed before, it will depend on your audience and the amount of background
knowledge they have or need.
105
Acceptability
If you leave your premises no more believable than your conclusion, and if your
conclusion happens to be quite unbelievable, then be prepared to have someone
reverse the argument on you, arguing that one of your premises is suspect.
The following little example will show you what I mean:
Everywhere that Mary went, her lamb was sure to
follow. And Mary was at the party, so the lamb was
there too.
Now imagine that someone objects by saying the following:
Well, I was looking for Mary at the party all night and
didnt bump into her. And Im allergic to sheep. So I
would have noticed a lamb at the party! But I agree with
you that Mary and that lamb of hers are always
togetherits weird! And thats why I dont think Mary
was even there.
Here someone has argued against my second premise. Alternatively, heres how
someone might argue that my first premise is suspect.
I doubt that Marys lamb was therethey dont even
allow cats in the house! But I do think I saw Mary. So I
bet Mary and that lamb of hers have finally parted ways
a bit.
So make sure you provide acceptable reasons for thinking your conclusion is true,
or your audience might well reverse the argument on you and provide you with
stinging reasons for thinking your premises are false.
Begging the question

You should beware of a special way of leaving premises no more believable than
their conclusions. This is called the fallacy of begging the question.
Begging the questionwhen one of the premises is either a
disguised version of the conclusion or presupposes the truth
of the conclusion.
Certainly, when your premise is your conclusion, its not going to be any more
believable than that conclusiona statement cant be more believable than itself.
106
Acceptability
This is also called, appropriately enough, circular reasoning: You should

believe it, because I told you about it, and you should believe me because Im
telling you the truth. When you sense that the argument is circling back on
itself, relying on the very thing its supposed to be supporting, thats your
indicator of circular reasoningbegging the question.
Here are a couple frequently used question-begging arguments.
We Christians rightly believe in the Christian God as described
in the Bible, because the Bible is completely accurate. It is, after
all, the revealed Word of God. And God, in His perfection,
would not lie or mislead anyone. So the Bible can be trusted
completely. Other religions, on the other hand, are not to be
trusted completely, because their teachings are not those of the
Holy Christian Bible.
Here, the question is begged in favor of Christianity and begged again against
other religions: This argument for Gods existence presupposes Gods existence;
and this argument against other religions presupposes they are incorrect by
assuming that Christianity is correct.
Abortion is morally wrong, because it is the murder of an
innocent unborn child.
Here, the wrongness of abortion is begged by being part of the very definition of
murdera wrongful killing of a personwhich is ascribed to abortion in the
premise.
How about this one?
My advice is to invest in a tech-stock sector fund, because their
stocks are predominantly in the NASDAQ instead of the Dow.
The NASDAQ will do much better than the Dow, because it
consists more of nibbler small-cap stocks than does the Dow,
with its big cumbersome Blue Chips. Yes indeed, small-caps are
the stocks of the high-tech future for Wall Street.
The premises ultimately presuppose or restate the conclusion: The final appeal to
Wall Streets high-tech future is pretty much just another way of stating that the
tech sector will do well. Some circular arguments make quite a trip, but big
circles are still circles. They still beg the questionthey just take a little longer
to do it.
107
Acceptability
Accepting premises for the sake of argument

We have been discussing the issue of whether or not premises are in fact
acceptable. I want to change the subject just a bit now and talk about a couple
special forms of argument that result when one doesnt in fact accept or believe
the premises of an argument, but instead merely accepts them purely for the
sake of discussionone makes believe they are true and then investigates their
implications. This happens in two noteworthy ways.
Conditional arguments
1. Assume that the University once again fails to get a
real budget increase.
2. In that case, there will be a freeze on staff hirings.
3. If that happens, computer service will only get worse.
4. If the University doesnt get a substantial budget
increase, then computer service is going to
deteriorate.
This is called a conditional argument, because of the form of the conclusion;
its a conditional statementan if, then statement. On the basis of these
premises, could one conclude that the Universitys computer service will be
getting worse? No. One can draw that conclusion only under the assumption
stated in the first premise: If that assumption is true, then computer service goes
into the toilet.
Whats the good of such a limited conclusion? Well, sometimes you dont know
exactly what to believe about your circumstancesit could be one thing or
maybe another. And so you dont know exactly what to conclude. But that
neednt stop you from investigating various implications of various possibilities,
for planning purposes. Sometimes, in times of uncertainty [and which times
arent?!] conditional conclusions are all you can draw, in preparation for when
the truth of your premises is finally settled.
Reductio ad absurdum
A reductio ad absurdum, or reductio, as its called for short, is best understood
as a type of criticisma criticism against either an argument or a position.
Reductio ad absurdum is Latin for reduced to absurdity. And thats what one
is trying to do to the targeted argument or positionshow that its absurd to
adopt it.
108
Acceptability
It begins by assuming, purely for the sake of discussion, that certain statements
are true. If the reductio is directed at an argument, assume for the sake of
discussion that its premises are true. If the reductio is directed at a position,
simply assume for the sake of argument that that position is true.
Next, the reductio details some of the implications of those assumptions; more
specifically, it details some of the implications that are inconsistent or just
patently false. Lets illustrate the reductio by using an example drawn once again
from the well of the abortion debatea deep well of bad reasoning by all sides of
the issue. Lets look at the popular Sanctity of Life Argument [SLA] against
abortion on demand.
This argument claims that all human life is sacreda gift of God. And, as a gift
of God, only God can take back such a life. That is why the taking of a human
life during an abortion on demand is morally impermissible.
Assuming these premises are true, for the sake of our investigation, what else is
implied by them? What other conclusions are equally well supported?
Well, lets trace some out.
The sanctity of human life is not a function of how that life was brought about.
The origin of the fetus has no bearing on its sacred statuswhether it comes
from Cleveland, for example, makes no moral difference. And that it came about
as the result of rape or incest should be equally irrelevant, then. In light of this,
however, abortions in the case of rape and incest should be construed as being
just as morally objectionable as abortions on demand. But this contradicts the
usual position adopted by the advocates of the SLA, which makes an exception to
the moral impermissibility of abortion in cases of rape and incest.
What if the pregnancy becomes life-threatening in such a way that, tragically,
one but only one can survive the situationif the pregnancy continues, the
woman will die and the fetus is rescued; if the abortion is performed, the fetus is
sacrificed and the womans life is saved. There are two lives of equal sanctity
now in the balance. The SLA has no means of distinguishing between the two,
favoring one over the other. In such cases of equal claims, each claim must be
given an equal chancesuch an impartial means of deciding would be, for
example, flipping a coin. But this too contradicts the usual position adopted by
the advocates of the SLA, which makes an exception to the moral impermissibility of abortion in the case of life-threatening pregnancies. Moreover, to
conclude that abortions in the case of life-threatening pregnancies should be
decided by means of a coin-flip is patently false, wildly absurd. Thats, however,
where the SLA leads us.
109
Acceptability
Things get even worse for the SLA. It ultimately implies that all lethal cases of
self-defense or the use of lethal force in the life-saving defense of others is
immorala taking of a sacred life, which only God is allowed to do.
But, of course, this is absolutely absurdthere are perfectly permissible cases of
self-defense and defense of others by lethal means. In fact, some such cases
would arguably be our moral duty!
So there must be something wrong with the SLA. It may have first appeared
reasonable, but we now see that it is a loose cannon, spraying out patently absurd
conclusions. We see that the SLA is completely unreliable as a means to the
truth.
The victim of our reductio is left with three options:
1. Keep their original argument, and deny the new implication.
But: If the new implication is as well supported by the
argument as their original conclusion is, then this option
would entail that they are inconsistentusing their argument
when it leads them to conclusions they like, but not using it
when it leads them to conclusions they dont like.
2. Keep their original argument and avoid being inconsistent by
embracing the new implication toobiting the bullet, lets
call it. But: This would mean adopting a patent
falsehoodadopting something that is wildly, absurdly
false. [Hence the name!]
3. Avoid being inconsistent and avoid being forced to adopt an
absurd implication by simply giving up the argument.
What is getting them between the rock of inconsistency and
the hard place of implying an absurd falsehood is their
argument. By getting rid of the argument, they get rid of the
problem. But: Giving up ones argument means that ones
conclusion is thereby reduced to a mere assertion.
So, when someone hands you an argument, ask two very important
questions:
What are the implications of this argument? Find
implications that are indeed as equally-well supported by
the argument as the original conclusion is. This
forecloses on the first option listed
aboveinconsistency is not an option!
110
Acceptability
Are any of these implications patently false? Make

sure these implications are patently false. This
forecloses on the second optionmake the bullet so
unpalatable that no one could bite it! Get as absurd as
you can!
A successful reductio leaves its recipient with only the third optionditching
their argument.
One way of giving up an argument is to supply it with an amendmentthis is an
admission that the argument does not work reliably on its own. If the recipient of
your reductio does this, you simply repeat the process on their new, amended
argument. Such is the never-ending process of critically reviewing arguments.
Pop quiz:
The SLA is usually amended in an attempt to avoid the
reductios we just constructed. One amendment has
classically been called the Double Effect Argument,
which stipulates that of all the effects that are caused by
our actions, we are to be morally judged on only the
intended effects, rather than the unintended [and often
regretted] effects. Adding this argument to the SLA, one
can then avoid the implications that abortion in cases of
rape or incest or in cases of life-threatening pregnancy
are morally objectionable, because in those cases the
intention behind the abortion is to save the woman from
further pain of the sexual assault or to save the life of the
pregnant woman; the intention is not to kill the fetus,
even though one knows this will regrettably result.
How many reductios can you find against the Double Effect
Argument?
If reductio ad absurdum criticisms sound familiar, they shouldall of the

criticisms I trotted out against the slippery slope argument were reductios. Thats
why slippery slope arguments are so fallaciousas reasonable as they first may
sound, they slip the leash and equally support absurdly false conclusions.
So far, weve just been looking at reductios of arguments. Now lets take a look
at a reductio directed at a position.
111
Acceptability
Quick reminder: There are two types of criticisms: a criticism

of an argument and a criticism of a position. The former,
when successful, just removes a reason for believing that a
claim [viz., a conclusion] is true. The latter, when successful,
actually provides a reason for thinking a claim is false.
As our target, lets use the Golden Rule: Do unto others as you would have them
do unto you. This has long been believed to be a pretty good guide to ethical
behavior and to promoting happiness in society. And, indeed, it works a fair
portion of the time. But, as a principle, it has some major flaws which work as
reductios to this position on how to promote happiness and morality.
For example, the first time I was with my wife, Ann, when she became ill, I
wanted to make her as comfortable and as happy as possible, so I consciously
applied the Golden Rule, asking How would I like to be treated when Im
feeling like crap? I like to be left totally alone; I dont want to bother with
people wondering how Im doing. So thats how I treated her. Big mistake! She
likes lots of attention when shes sick; she likes to be checked on frequently; she
likes lots of poor baby!s.
The first time I was sick, she too consulted the Golden Rule to figure out what
she could do for me. I thought Id die!
Other reductios can be generated by simply citing some pretty nasty and masochistic things people might like done to themselves. The Golden Rule would
deem them morally permissible to do to others, and thats just wrong.
Here, then, is the basic format for building reductios:
Find implications:
Cite absurdity:
Reject target:
If that were true, then

this would follow.
But this cant be true.
Therefore, that must be incorrect.
Confession: One of the most refreshing things you can do when applying your
critical thinking skills is to deliver a stinging reductio. It is the most powerful
tool of criticism at your disposal. Long-reigning theories in science, religion,
ethics, you name it, have come tumbling down because of reductios. It doesnt
matter who you are, who you know, or how much money you have, this critical
tool is always at your disposal to help you cut through the errors and get closer to
the truth. Andit feels pretty good too, when you can reduce some pompous gas
bag to absurdity.
112
Acceptability
Pop quiz: Have some fun finding reductios to these rather

popular arguments or positions:
1. Correlation experiments in artificial conditions are regarded
by many competent critics as an unilluminating and
unreliable way of investigating complex behavior, even in
many other species, let alone in human beings. [The
Williams Committees (1979) argument against using
laboratory research as evidence for the theory that viewing
violent pornography increases mens disposition to express
aggression against women.]
2. Even though it might arise in a genuine effort to promote the
general welfare and to protect certain rights, officials and
groups might use the power to censor as a means to advance
their own interests and values and to suppress the rights,
interests, and values of others. [Mark Wicclairs argument
against the censorship of pornography, from Feminism,
Pornography, and Censorship.]
3. Prior to the point of viability, the fetus is totally dependent
on the pregnant woman for its very existence, and thus owes
its very existence to her continued assistance. At that time,
therefore, she is due total veto power. But after viabilityat
the point at which, were the fetus removed, it could survive,
albeit with artificial aidthe fetus is no longer totally
dependent on her, and she thus loses her total veto power
over the continued existence of the fetus. Hence, prior the
viability, abortions are morally permissible, but not after.
[The Independence Argumentimplicit in Roe v. Wade.]
4. Abortion is immoral after approximately the tenth week into
the pregnancy, because the fetus is sentient by that pointit
has the capacity to feel pain. And all sentient creatures have
the right to life.
5. I know my meter was expired! But did you see the cop who
gave me the ticket? She looked right at the next car, and its
meter was expired too just like mine, and she walked right
by it! I was treated wrongly because I was treated unfairly,
since fairness is equal treatment under equal circumstances.
6. Abortion in the case of rape or incest is perfectly moral,
because the woman was wronged, and thereby should be
permitted to avoid living with the aftereffects of that wrong.
113
Acceptability
7. Withholding medical treatment, i.e., passive euthanasia, is

morally permissible [given the patients and familys
consent], unlike active euthanasia, because the intention in
the former case is merely to save the patient from further
suffering.
8. What counts as desecration of the flag is vague [territory
having no discernible or defensible boundaries, Justice
William Brennan]. And so laws banning it would be vague
too. And thats why they should never be permitted and
should always be ruled unconstitutional when they are
proposed.
A few answers [the rest I leave for you to discover and

discuss]:
Re. 1. Almost all experiments in all the sciences are done under
artificial conditions, but it is absurd to think they are
unilluminating and unreliable for that.
Re. 2. It is true of every law, that it might be used to advance
the self-interests of those enforcing them. But it is absurd
to think that these laws should be abolished for
thatwed be left with no laws!
Re. 3. If the fetus total dependence on the pregnant woman
gives her the right to end that relation by killing the fetus,
the same would be true of the newborn or even the young
child, who is totally dependent on someone or other for
its very existence. But it is absurd to think that one can
kill, e.g., a two-year-old the moment it becomes totally
dependent on one, simply because it is so dependent.
Re. 4. If sentience is what wins creatures rights, then livestock
have rights to the same degree as people do, and those
raising livestock are morally equivalent to the Nazis. But
that is absurd, even if sentience is a sufficient condition
for lesser degrees of moral consideration.
114
Acceptability
Another use for reductios

Weve been studying reductios as a form of criticism. But they also have a
positive function. You can use them to argue for a position. This is especially
important when you find yourself trying to argue for a position or belief that is
basic or fundamental to your whole world view and your entire stock of beliefs.
Everyone has certain beliefs that are so fundamental that, if you were to try to
give reasons for them, you would have to beg the question of their truth.
For example, you have basic beliefs about math; for instance, that 2 + 2 = 4. Try
arguing for them though! If you were to appeal to the fact that when you take
two blocks and put them in a box with two other blocks you get four blocks, that
might seem persuasive, until someone starts pressing you on what a small sample
you have in your studyyoure drawing a universal conclusion on the basis of a
little experience with putting blocks in a box? Thats as lame as when, as a kid, I
concluded that all uncles chew snuff, on the basis that Uncle Henry did.
In every discipline, there are going to be fundamental assumptions or positions,
such as 1 + 1 = 2 for arithmeticfundamental principles that are presupposed
by whatever else you would appeal to to argue for them. When you find yourself
unable to argue for a belief without begging the question of that very belief, what
should you do?
The first thing to do is to fess up to the fact that you have reached a fundamental
assumption. Intellectual honesty demands this, as opposed to offering only
question-begging reasons for your assumption.
But you can do more than thatyou can try arguing for your fundamental
belief by doing a reductio on its denial. Hows that?
Heres what I mean: Assume for the sake of argument that 1 + 1 2; lets say,
for the sake of discussion, that 1 + 1 is something else, say 3. But if 1 + 1 = 3,
then, since 1 + 2 = 3 too, 1 + 1 = 1 + 2. But thats just plain false! [And it would
lead to countless other contradictions, such as 1 + 1 + 2 = 1 + 2 + 2, viz., 4 = 5.]
So, the position that 1 + 1 2 must be false, viz., the position that 1 + 1 = 2 must
be true.
We have argued for the most fundamental belief in mathematics, a belief that all
other mathematical beliefs presuppose, by arguing that its denial leads to a
falsehood.
Again, this situation crops up within every disciplineevery discipline has its
fundamental planks in its platform, planks that everything else is built upon.
Lets examine one more case, this time in the physical sciences.
115
Acceptability
Here is an argument for the conclusion that there is no soul that is responsible for
our actions; there is no spiritual entity that is causing our bodily movements. It
consists of two premises: 1) Only physical objects can cause physical events,
such as the movement of the body. 2) The soul is a spiritual objecta nonphysical object. And thats why its just not the kind of thing that can cause
actions, unlike the kind of thing the brain is.
The flaw with this argument is that the first premise, about what alone can cause
physical events such as bodily movements, already rules out souls as possible
causes. So the argument commits what fallacy? Thats right, the fallacy of
begging the question. But how would one argue for that first premise? We
would run into the same problem as we witnessed earlier, if we just appealed to
our vast experience of physical objects that do cause physical eventsits not a
sample big enough to rule out souls doing it too.
We soon realize that this premise is a fundamental assumptiona basic position
in a world view we call physicalism. It is the view that all physical, natural
phenomena are the result of physical, natural laws and relations and that there are
no external, supernatural influences on these phenomena. This is the basic belief
of science. The denial of this premise is what we call dualism. Dualism is the
position that there are two kinds of entities, physical and spiritual, and that they
can interactphysical events causing spiritual events and spiritual events
causing physical events.
The physicalist usually argues for physicalism by claiming that the dualist is
unscientific. And the dualist usually argues for dualism by claiming that the
physicalist is closed minded. But, of course, this just amounts to each camp
begging the question against the other during this name calling. What each camp
needs is enough intellectual honesty to at least admit their respective fundamental
assumptions are just that, fundamental assumptions.
But the physicalist can do more than that. They can do a reductio on the denial
of their first premise, and thereby do a reductio on dualism. Heres how.
Assume for the sake of argument that dualism is trueassume that not only
physical objects can cause physical events, but non-physical objects can too.
The soul or spiritual self is non-physical, and thus, according to the dualist, is not
fettered with the same limitations that physical objects are. For example,
spiritual objects dont have mass, evidenced by the absurdity of wondering how
much ones soul weights [an ounce; a pound?!]. Moreover, they dont have
spatial location either, evidenced by the absurdity of wondering where your soul
is [behind your ear; in your left foot; up your butt?!]. The soul, according to the
dualist, doesnt need mass to cause events in the physical world, and it doesnt
116
Acceptability
need to be in contact with the physical object it affects in order to affect it. And
now we are ready to begin drawing implications.
If indeed the physicalist is wrong and the dualist is right, then not only should my
soul be able to cause my body to move at my will and your soul cause your body
to move at your will, but, with equal ease, my soul should cause your body to
move at my will, and your soul should cause my body to move at your will.
After all, my soul shouldnt just move my body because its near it; my soul is
not near any bodyits nowhere at all.
But, of course, these across-individuals causal relations are not equally probable.
[I order you to send me all your money immediately! See? Nothing.] So
dualism cant be right; only physical objects can be causing physical events such
as our actions. And thats why our minds must be, in some sense, identical to our
brains, in order to be in control of our bodies.
So, if you cant argue for your position without begging the question, try
doing a reductio on its denial.
Pop quiz: Can you think of any other absurd implications of

dualism, now that the reductio is all set up?
Hint: The dualist has admitted that the soul has no mass.
That concludes our discussion of the first of our cogency conditions, the acceptability of premises. Its time to move on to the second condition: The relevance
of those premises.
117
Relevance
7. RELEVANCE
An argument is supposed to be a conduit of truth and acceptability. And thats
the job of the relevance condition for cogencyto make sure the acceptability
of the premises, to some degree at least, gets transferred to the conclusion.
Premises are relevant to the degree that they provide reason to believe their conclusions. So premises can get by rather cheaply sometimes and still meet the
relevance condition, since all they have to do it to provide some reason to
believe the conclusion.
Remember, seeing one measly chickadee eat one measly sunflower seed enables
me to meet the relevance condition as I draw the broad conclusion that chickadees prefer sunflower seeds over any other food. Its more reason than I had
prior to that observation [when I had no reason at all!].
Another way of characterizing the relevance condition is that the truth of the
premises increases the probability that the conclusion is true. Again, it
doesnt have to increase it by much; it just has to increase it.
Irrelevance
So whats it like to flunk the relevance condition? Embarrassing.
Its when the truth of the conclusion is independent of the truth of the
premises.
Its when the connection between the premises and the conclusion is so out of
whack that the truth or falsity of the premises doesnt matter to the truth or
falsity of the conclusion.
An argument falling victim to this plight is said to be a non sequitur. This is
Latin for That doesnt follow! Here are some examples:
I should not be given such a low grade for this course. This
course is only an elective, its only a 100-level course, and I get
high grades in all the courses for my major.
You shouldnt do that! What if everyone did that? Wed be in a
fine fix then, wouldnt we?
In response to recently proposed closures of sensitive sand dune
areas on Bureau of Land Management tracts to protect endemic
118
Relevance
plants and animals, [off-road vehicles] groups said impacts were

exaggerated. There seems to be a concerted effort to lock this
country up with respect to outdoor activities, Mark Harms,
owner of an off-road tire store and a member of the American
Sand Association, told the Associated Press. Theres going to
be nothing left to do if this keeps going. [From an article by
Todd Wilkinson, in National Parks, March/April 2001]
These last two examples are especially interesting because so many people think
these kinds of appeals to hypothetical cases are relevant. But, what would result
if things were taken to extremes is totally irrelevant if things never will in fact be
taken to such extremes. Just think about it: What would happen if everyone
became the likes of Mother Teresa? The world would grind to a haltno
doctors, no farmers, no police, etc. But thats no reason why the rare individual
who decides to dedicate their life to helping people shouldnt do so.
Some premises are so irrelevant, its as if the author has completely changed the
subject. These are called red herrings. People swim them by to try to distract
you into accepting their positions.
Instructor: And why should I give you a passing grade, when
you never achieved even minimal competency in any of the CT
modules?
Participant: I just want you to know how worthwhile I think
critical thinking is, and philosophy in general, for that matter.
Whether and how an argument flunks the relevance condition depends
crucially on the kind of argument it is. Lets examine this with respect to four
types of arguments:
deductive arguments
arguments by analogy
inductive generalizations
arguments by subsumption
Relevance and deductive arguments

With deductive arguments, the premises must meet the highest of standards in
order to meet the relevance condition: They must be able to provide sufficient or
conclusive reason for believing the conclusion. The truth of the premises must
be able to do the work of guaranteeing the truth of the conclusion.
119
Relevance
Remember how deductive validity by itself does not guarantee that the
premises are relevant? Validity is just a relation between the premises and the
conclusion such that at no time could the premises be true and the conclusion
false. But that relation can hold with no help from the truth of the premises.
Bill Clinton was president; therefore, 2 + 2 = 4, is a valid argument; but its
premise does not give us a reason for believing the conclusion.
So when we claim that an argument is valid because its premises guarantee its
conclusion, we are saying that it is both valid and meets relevance condition.
For deductive arguments, relevance is going to be all or nothingeither the truth
of the premises guarantees the truth of the conclusion or it doesnt. Either the
premises entail the conclusion or they are irrelevant. This is because of the
strong claim of a deductive argument. It doesnt claim to make its conclusion
probably true, given the truth of its premises; it claims to make its conclusion
true, given the truth of its premises.
So if a deductive argument is invalid, its also a non sequitur.
Relevance and arguments by analogy

The argument by analogy is perhaps the most persuasive, accessible, allpurpose argument available. [BTW: Its as frequently called the argument
from analogy.]
Its especially useful when arguing about ethical issues. Here is perhaps the most
famous argument by analogy in contemporary ethics. Its from an article by
Judith Jarvis Thomson entitled A Defense of Abortion.
You wake up in the morning and find yourself back to back in
bed with an unconscious violinist. He has been found to have
a fatal kidney ailment, and the Society of Music Lovers has
canvassed all the available medical records and found that you
alone have the right blood type to help. They have therefore
kidnapped you, and last night the violinists circulatory system
was plugged into yours, so that your kidneys can be used to
extract poisons from his blood as well as your own. The director
of the hospital now tells you, Look, were sorry the Society of
Music Lovers did this to youwe would never have permitted it
if we had known. But still, they did it, and the violinist now is
plugged into you. To unplug you would be to kill him. But
never mind, its only for nine months. By then he will have
recovered from his ailment, and can safely be unplugged from
you. Is it morally incumbent on you to accede to this situation?
120
Relevance
No doubt it would be very nice of you if you did, a great

kindness. But do you have to accede to it? What if the
director of the hospital says Tough luck, I agree, but youve
now got to stay in bed, with the violinist plugged into you,
Because remember this. All persons have the right to life, and
violinists are persons. Granted you have the right to decide what
happens in and to your body, but a persons right to life
outweighs your right to decide what happens in and to your
body. So you cannot be unplugged from him. I imagine you
would regard this as outrageous, which suggests that something
really is wrong with that plausible-sounding argument [from the
hospital director].
Consider this story, about you being kidnapped by the Society of Music Lovers
and hooked up to a dying violinist, to be one lonnnnnng premise. What
conclusion about abortion follows from this story?
What follows from this story is that, even if the fetus were a person, like the
average violinist, abortion in the case of rape-caused pregnancies would be tragic
but morally permissiblehaving the abortion would be the pregnant womans
ethical right.
Some of you might be asking what the relevance of this little bedtime story is to
the issue of abortion. That is exactly what should be asked when considering
whether or not the relevance condition has been met by an argument from
analogy!
With an argument by analogy, the relevance of the premises is a function of
the degree of the similarity between the subject matter of the premises [the
analog] and the subject matter of the conclusion [the target].
Huh? Admittedly, this is too abstract right now to give you a good feel for how
the premises of an argument by analogy are relevant towards supporting the
conclusion, but a little more discussion will remedy that.
Arguments by analogy have the following basic format [lets stick with the area
of ethics, just for the purpose of illustration]:
1. Action A is morally [im]permissible.
2. Action B is just like A in all relevant respects.
3. Action B is morally [im]permissible too.
If you indeed agree with these premises, then you must agree with the
conclusion, or be guilty of an inconsistency. After all, you have agreed that the
two actions are alike in all relevant respects. Refusing to agree to the conclusion
121
Relevance
would mean you are making a distinction between these two actionsa
distinction where you have just confessed there is no difference.
Thats the power of an argument by analogy: If your audience agrees with your
premises, youve gotem!
Lets go back to J.J. Thomsons famous argument and see how it fits into our
schematic.
1. Unplugging yourself from the violinist in these circumstances

would tragically kill the violinist, but you would have an
ethical right to do itit would be morally permissible.
2. Abortion in the case of rape-caused pregnancy is just like the
case of unplugging yourself from the violinist in all relevant
respects:
analog:
target:
you
the violinist
fatal kidney ailment
the kidnapping
the Society of Music Lovers
plugging in
unplugging
the pregnant woman

the fetus
total dependence
the rape
the rapist
impregnating
the abortion
3. It would be morally permissible to have an abortion in the

case of a rape-caused pregnancy, even assuming that the fetus
is a person with the same moral status as the pregnant woman.
What if someone were to point out that violinists and fetuses are so different that
this just doesnt make any sense; after all, violinists play the violin and fetuses
dont? The proper response is that this difference is irrelevant. An argument by
analogy just requires that the two cases be alike in all relevant respects; they
dont have to be alike in all respects, period. Here the only relevant respects are
those that have to do with ones moral status and ones right to be treated in
certain ways. And certainly, being able to play the violin does not win one any
more or less moral status. Similarly, violinists generally have more hair than
fetuses; but thats not a relevant difference either.
122
Relevance
As you construct your arguments from analogy, use our basic format as a
checklist:
Regarding premise #1: Find a clear-cut case. For example, if
you are arguing for the conclusion that a certain action is morally
impermissible, find an analogous action that is clearly morally
wrongone that everyone except the pathologically amoral
would agree with. How about this: Shooting the elderly for
sport is wrong. We have hunting season on deer and it has its
benefits, especially for us rural drivers [collisions with deer
being the most frequent type of auto accident]. And, from a
macabre point of view, having hunting season on seniors would
have its social benefits tooit would go some way towards
solving the social security issue; it would speed up the lines at
the supermarket, etc. But no matter; it would be just plain
unethical! And anyone who would even hesitate to admit
this,well, dont invite them to your partytheyre beyond the
pale!
Regarding premise #2: Make sure your analogy is a good
one. Make sure your analog and your target are alike in all
relevant respects. Make sure your analogy is on all fours.
As you critique arguments from analogy, use this same checklist to form your
critical questions:
Regarding premise #1: Is this a clear-cut case? If not, the
argument by analogy doesnt get off the launch pad.
Regarding premise #2: Is there any relevant difference
between the two cases? This question is important because, if
there is such a relevant difference, one could very well agree
with the arguments claim about the analog but yet disagree with
the conclusions claim about the target, without being
inconsistent.
False analogy
If either of these premises is false or unacceptable, the argument commits the
informal fallacy of false analogythe premise regarding the so-called analogous
case is irrelevant to the conclusion about the target case.
For example, what about the following argument?
123
Relevance
The Governments failure to substantially increase healthcare

benefits is morally bankrupt. You might as well just take our
poor seniors out back and shoot em, in light of how many of
them die annually for lack of adequate healthcare!
The problem here, of course, is that the argument is talking apples and oranges:
All things being equal, failing to provide someone life-giving assistance is not on
a moral par with killing them. This is a false analogy.
Pop quiz: If we were ever to think there is no relevant

difference between letting die and killing, then you and I would
both be no better than mass murderers, as we sit here at our
computer screens! For, instead of sitting here, we could be
running down to the Post Office, sending a check to some
organization for world-hunger relief. Im sure its quite true that
with the money from a small check, a few of the worlds starving
could be fed for a week or a month, So, as we sit on our dead
butts, those people die! But certainly, that is different from our
sending them boxes of poison food which kills them. So the
killing / letting die distinction must be correct. What kind of
criticism did I just use?
FYI: The longest analogy that Im aware of is Platos Republic.

Its a book-length analogy, giving an analysis of what justice is
for the individual, by analogy from what justice is for the State.
Even if you dont agree with the final theory of justice Plato
assembles by the end of the book [and I doubt if anyone could],
the Republic is a marvel of literature and critical thinking.
Relevance and inductive generalizations

One of the most frequently used types of arguments in science is the inductive
generalization. For example, we easily conclude that the sun will rise tomorrow,
on the basis of seeing the sun rising in the past. We quickly learned that touching
hot objects hurts, on the basis of getting burned a couple times. Inductive
generalization is basically just learning from experiencebasic conditioning.
Such learning happens so quickly that we dont normally think of it as involving
any argumentation, but our [implicit] inferences from experience are indeed
captured by the following schematic for inductive generalizations.
124
Relevance
1. A certain degree of regularity holds in the

past. For example, one event regularly
brings about another event [e.g., acid causes
corrosion], or one state of affairs is regularly
accompanied by another [e.g., mammals
give birth to their young live].
2. The Regularity of Nature
Assumptionpresent or future regularities
resemble past regularities.
[This premise is usually left implicit.]
3. Probably, the same degree of regularity cited
in Premise #1, will hold also in present or in
the future.
Notice that this is an inductive argument, so the premises can at most make the
conclusion probablenot certain.
The past evidence cited in Premise #1 is relevant only so long as Premise #2 is
true, i.e., only so long as the present and the future resemble the past. If this
Regularity of Nature Assumption is false, all bets are offwhat happened in
the past would be no indication of what will happen afterward. And thats how
premises involving empirical evidence can flunk the relevance condition.
Bertrand Russell inspired the following nice example of how our empirical
evidence can become quite irrelevant: A turkey begins to notice a certain pattern
to the farmers behaviorwhen the farmer appears, the turkey gets corn thrown
at it. This happens time and time and time again. But then one special morning,
all this evidence about the past correlation between the farmer and the corn is
irrelevant, and it isnt probable at all that the farmer is going to be throwing
corninstead, hell be throwing an ax!
Relevance and inductive arguments by analogy

The arguments by analogy that we studied a moment ago were classified as
deductive argumentsif their premises are true and one still refuses to accept
the conclusion, one is guilty of an inconsistencya logical inconsistency. But
not all arguments from analogy are intended to guarantee the truth of their
conclusions; some are only intended to make their conclusions probably true.
An example will illustrate this difference:
125
Relevance
Bob sat on his CD and bent it so badly that the disk is

unreadable; so stop sitting on your disk or its likely to become
unreadable too.
Note the implicit premise that the instances are alike in all relevant respects
Bobs disk and your disk. But the analogy is false [and, thus, the conclusion is
left unsupported] if these disks differ in some relevant respect; for example, if
you kept your disk in a protective case, while Bob did not.
Relevance and arguments by subsumption

What Im calling subsumption arguments are the kind in which one draws a
conclusion about a particular case by claiming it is covered under a general rule
what goes for the general rule, then, applies to the particular case being subsumed under it.
Many arguments about what ought or ought not to be the case have this format.
This is a very frequently used method of argumentation in ethics. For instance,
one could very well argue that I ought not to stroll up to some stranger standing
on the street, minding their own business, and poke them in the eye with a sharp
stick just for grins, because one simply ought not to cause others unnecessary
harm [and its certainly not necessary that I get my grins, especially this way!].
Notice the schematic of this typical argument by subsumption:
1. Statement of a general principle, to the effect
that one ought [not] to do actions of a certain
type.
2. Claim that a certain action in question is a
token of that type.
3. One ought [not] to do that token action.
Subsumption arguments flunk the relevance condition when the particular instance in question just isnt subsumed under the cited rulewhen the rule just
doesnt apply to it after all, i.e., when Premise #2 is just plain false.
Take the case of the guy who was charged with desecrating the U.S. flag when he
used it as a grease rag as he changed his oil. His lawyer argued that he should
not be so charged with any offense because of the unconstitutionality of any
ordinances prohibiting the desecration of the flagall such ordinances violate
ones right to free speech.
126
Relevance
But this appeal to the general principle that ones right to free speech ought not to
be infringed upon is totally irrelevant, because this guy was not saying anything
by wiping off his dipstick with the flag; he was just changing his oil. Wiping off
his dipstick was not a form of speechhe was not expressing a political protest.
There are plenty of reasons why he should be allowed to use a discarded U.S.
flag as a grease rag; its just that the exercise of his right to free speech isnt one
of them here.
It seems to me that this sort of thing happens quite oftenpeople appeal to
ethical or legal principles that in fact are irrelevant. For purposes of discussion,
let me give you a more controversial example:
The most popular argument against the censorship of pornography is that it
violates its producers right to freedom of speech. And yet this ethical or
constitutional right is relevant only if pornography is a form of speech. Is it?
What is said by pornography?
Not, what is intimated by pornography? For pornography intimates many
things, one of which is the view that women are the fit targets of sexual abuse.
[And the producers of pornography have consistently denied intending that
message.] Many actions intimate or indicate things without stating them; for
example, my neighbors failure to take in his garbage can indicates that hes
lazy, but his leaving his garbage can out is not his saying to me or to anyone
Im lazy. So let me repeat the question: What is pornography saying, or
what are the producers of pornography saying, to the consumers? Kind of
puzzling, once you think about it!
Well, maybe this will help: What is the function of pornography? Is it like Time
magazine, a clear-cut case of a form of speech, telling its readers things? Thats
a stretch! Whats its function, then? Well, basically to sexually arouse its
consumers. In light of this discovery, its no mystery why those other items are
usually sold in porn shopsthose novelties or marital aids. Their sole
function too is sexual arousal.
Now, make believe that the government wanted to confiscate such novelties. [I
can hear the likes of Charlton Heston now: The only way the governments
going to get my dildo is to pry it from my cold, dead hands!] Would one appeal
to the right to freedom of speech to argue against it? Hardly! Theres no speech
involved here [other than perhaps the directions on the box about how to put the
batteries in, but thats not what were interested in protecting].
And so it is with pornographyits as silly to appeal to ones right to free speech
to protect it as it is to appeal to the right free speech in order to protect ones use
of marital aids. What appeal would be relevant? Well, how about an appeal to
ones generic right to libertythe right to own and do what one wishes so long
127
Relevance
as doing so does not infringe on the rights of others. This is indeed an important,
applicable ethical and legal principle. The debate about pornography, then,
focuses where it shouldon the interesting, complex empirical issue of whether
or not the consumption of some forms of pornography increases the disposition
of some men to aggress and discriminate against women. The censorship issue is
not thereby settled; rather, the debate about it is finally relevant.
Something to think about: Be on the lookout for cases in

which people inappropriately argue by subsumption. Sometimes
people use loaded or euphemistic language [spin] so as to
make it appear that their particular case correctly fits under an
acceptable rule or regulation or principle, when in fact it doesnt.
When youre inclined to say, But that doesnt apply! you have
pretty good evidence that someones argument by subsumption
has violated the relevance condition.
Informal fallacies of irrelevance

There are some ways of using irrelevant premises that are so popular and s o
persuasive, and yet so fallacious, that they have even acquired names; and you
need to study them in detail, so that you wont be fooled by them any longer.
Here is the list of the ones well cover:
Straw man
Ad hominem
Guilt by association
Appeal to ignorance
Appeal to popularity
Appeal to force
Straw man
Whats fishy about the following little argument?
The legislation prohibiting cigarette machines and requiring
elaborate proof of age is not a good idea. Do they really expect
this to stop every adolescent from getting cigarettes? How
ridiculous!
If indeed the advocates of this legislation did think this was a way of stopping
adolescent smoking, they would be ridiculously wrong. But that was not their
128
Relevance
reason or their expectationthats a gross misrepresentation of their argument.

All they wished to do, and all they expected to do, is to decrease adolescent
smoking. And indeed that was likely accomplished by these new regulations.
Straw manto criticize or attack not someones actual
argument or conclusion or position, but rather a
misrepresentation of it.
[BTW: Im a real stickler about sexist language, if you havent already noticed.
For instance, Ive been using plural pronouns (such as they) instead of singular
pronouns (such as he), or havent you even noticed? But I dont think calling
this fallacy the straw person fallacy is necessary. In fact, I think thats as
awkward as straw woman would be. No, straw man will be most recognized
and least offensive, so lets stick with it.]
The strategy behind the straw man is to attack a weaker facsimile of an opponents argument or position, and then pass it off as a successful attack on the real
thing. But, of course, even if the criticism against the weaker misrepresentation
is successful, the real opponent is still untouched. Thats why an argument
against a straw man flunks the relevance condition.
When the National Rifle Association argues against gun control by arguing
against the prohibiting of all firearms, they are flailing away at straw, since those
arguing for gun control are advocating only the likes of gun locks, some gun
restrictions [e.g., on fully-automatic weapons], or merely gun registration for
some types of guns or some types of gun purchasers. The misdescription of
ones opponent, as painting with broader brush strokes than they actually are, is
the straw man fallacy.
Heres another straw man that I found, in a letter to the editor in Money magazine [July 1999]. See if you can diagnose the fallacy:
Jason Zweig wrote 200 years of corporate history show that
the early leaders in a dynamic industry almost never turn out to
be the victors in the end. Since when have people been
investing with a 200-year horizon?
The best way to diagnose the straw man fallacy is to lay out both sidesthe
actual one and the critics misdescription of it, which they are attacking. The
straw man here is the view that one should not buy stocks of early industry
leaders because those corporations dont last 200 years. But that was not
Zweigs reason for cautioning against buying the stocks of early leaders.
[Indeed, virtually no U.S. company has lasted 200 years.] Zweigs argument,
based on looking at 200 years worth of evidence, is that the early leaders almost
always get bettered by their competition; so, it might be prudent to wait for the
129
Relevance
industry to mature before investing long-term. Hard to argue against that

argument!
Pop quiz: Here is a straw man thats become a classic, included

in many critical thinking texts, and for good reason. See if you
can diagnose it.
Wrong for many. Thats the reality of soft
energymassive, often unsightly projects. But the
dream is appealing partly because it seems small-scale
and spread out, like another fantasy of the back-to-nature
movementdo-it-yourself farming for everybody. Yet
to give every American family of four a 40 acre farm
would take more landincluding deserts and
mountainsthan there is in all of the lower 48 mainland
states. And such a program would surely mean goodbye wilderness. Besides, what about people who like
cities, or suburbsrather than the constant ruralism
in between? There may be a lot of good in soft energy
to supplement conventional power. But were uneasy
with people who insist it will do the whole job and who
then insist on foisting their dreams on the rest of us.
Especially when their dreams cant stand up to reality.
[Originally from a column written by the Mobil
Corporation]
Hint: This is an argument by analogy. Based on the analogy,
what is the straw position of soft energy being attacked?
Answer: The actual soft energy position is that alternatives to
fossil and nuclear fuels, e.g., solar and wind power, should be
used to supplement traditional public utilities. The straw soft
energy position is that these alternatives would replace
traditional public utilities. Ironically, what does the author think
of the actual soft energy position?
Two popular straw man fallacies

There are so many ways to commit the straw man fallacy that its impossible to
note them all, but two of them bear special attention.
Criticisms should be directed at the strongest, most current versions of an argument or position; and then, if they are successful, they also function as criticisms
130
Relevance
of any weaker versions of their targets. A straw man fallacy is committed when
one successfully criticizes a weaker version of ones opponent, claiming those
criticisms work equally well against all stronger versions too, when in fact they
dont. This variety of straw man has all the intellectual honesty of claiming you
climbed a mountain, when you only walked its foothills.
Another way of committing the straw man fallacy is to give an unsuccessful
reductio ad absurdum criticism. Heres why a fizzled reductio is properly diagnosed as a straw man: When your reductio fails, its because the position or
argument youve targeted either fails to have the implication you claimed it does
or its implication is not absurdly false as you claimed it is. In either case, you
have misrepresented your opponents position or argument as having an absurd
implication when in fact it does not.
Ad hominem
Literally translated from the Latin, this is Attack on the man.
Bill Clintons China trade policies were completely bankrupt.
The only foreign affairs he would have been competent at was
having sex in the Oval Office with an Asian intern.
This is a typical, blatant case of an ad hominem; and now you see why its so
named.
Ad hominemwhen derogatory charges are made against
the advocate of a position or argument, using those charges
as reasons to claim that the position is false or the argument
is not cogent.
What do ya mean, pornography is disrespectful of women?
Where do you get that feminist clap-trap anyway? Youve been
in that Women Studies course too long; its starting to affect
your brain. Youre in that course just to meet chicks, I know it!
There are a couple ad hominems in this example: The guys mental health is
questioned and his motives are being questionedhe is being called an opportunist for learning a thing or two in a Womens Studies course. The critic is
lobbing insults at him rather than critically reviewing his claim that pornography
is disrespectful of women. The author of this claim is attacked instead of the
claim itself. And thats the ad hominem.
The ad hominem fallacy is a way of flunking the relevance condition, and heres
how you can tell: Even if the derogatory charges against the person were
131
Relevance
true, it would not provide any evidence whatsoever against the persons
position or argument.
Lets illustrate: Assume, just for the sake of argument, that the guy in our
example is thinking with his heart instead of his head, and lets also assume that
indeed the main reason he took the Womens Studies course is to meet chicks;
especially chicks that aint afraid to spring for the check now and then. Is that
any evidence against his claim that pornography is disrespectful of women? Not
in the least; that explanation as to why hes making his claim is totally independent of the truth of his claim.
Those who favor putting restrictions on abortion are just a bunch
of old fart conservatives who think women ought to be on their
backs or by a stove!
Lets illustrate how this ad hominem is irrelevant. Even if it were true that all
those who think that abortion is unethical are old farts who also think that women
should be relegated to purely domestic roles, that would not be any reason to
believe that abortion is ethical.
Men would not be so anti-abortion if they got pregnant!
Another ad hominem. Indeed it is probably true that if men got pregnant too,
they would be more inclined to advocate abortion rights, but that is no reason to
think that abortions are ethical.
To improve your ability to spot ad hominems, there are two things to keep firmly
in mind:
The truth of ones position is independent of ones motive or
explanation for having that position.
The cogency of ones argument is independent of ones motive
or explanation for giving that argument.
Thats why attacking the motives or background of the advocate of the
position or argument is irrelevant.
Species of ad hominem
FYI: There are various ways of committing the ad hominem. Whats crucial is
that you simply be able to recognize an ad hominem when you bump into one
and, of course, to not commit one! Being familiar with some of the specific
kinds of ad hominem arguments, however, can only help you do this.
132
Relevance
Poisoning the Well: A pre-emptive ad hominem attack on

someones position or argument.
I dont care what kind of arguments hes going to use, they all
just stem from his Catholicism. Hes been trained to do nothing
but argue against abortion.
Even if every argument and statement the person will advance in the future is
motivated merely by their Catholic upbringing, those statements and arguments
stand on their own and must be independently tested for truth and cogency.
The Genetic Fallacy: Criticizing a position or argument
solely because of its origin.
His criticisms of the welfare program are hollow. How many
men are on welfare? Its women that need to be heard from on
this issue!
This fallacy has gained in popularity recently, under the guise of gender- or
culture-specific ways of knowing, as its often called. The passage above, for
example, boils down to the claim, His opinion is worthless because it comes
from a man. Certainly this sort of criticism was indefensible when it was expressed by men who claimed, Her claim is worthless because it comes from a
woman. Its equally indefensible when the ad hominem is lobbed in the other
direction. Those who live by the ad hominem, die by the ad hominem.
Evidence and justification are origin neutral. Truth and cogency dont care
where they come from.
There appears to be an exception to this rule; but there really isnt, when you
think about it a little bit: Knowing what its like to have a certain experience
depends on the origin of the knowledge, viz., only one whos actually had the
experience can know what its likee.g., only one who has tasted lemons can
know what lemons taste like. So maybe someone who has cancer knows best the
anxiety of having it. But that certainly doesnt make one a knowledgeable
oncologist. And certainly, being fortunate enough not to have cancer doesnt
disadvantage one from being an authority on it.
Seldom is it true that being there puts one in a privileged position of authority
on something. All too often, being there is so fraught with emotion and trauma
that its not the way of knowing its cracked up to be.
Tu Quoque: To respond to a criticism by pointing out that its
author falls prey to it also.
133
Relevance
Your defense of the federal governments threat to withhold

funds from states with lax seat belt laws is outrageous! Why,
just the other day I saw you drive your garbage down to the road
for pickup, and you didnt have your seatbelt on!
You can tell that this tu quoque fallacy is a violation of the relevance condition,
because the appropriate reply to it would be: Youre right, I did drive down to
the road without my seat belt on, and it was wrong; it was stupid!
The tu quoque boils down to the claim, Well, you do it too; so you should not
be criticizing others! But, of course, the proper response is, Indeed I do it too,
and its equally wrong when I do it! A persons hypocrisy and the fact that they
cant or dont live up to their own claims or standards, does not act as a criticism
of those claims or standards. In making an argument or a claim, a person might
be the height of hypocrisy, but their argument is no less cogent for that and their
statement is no less true for that.
To keep this firmly in mind, think about the following extreme example: Lets
say that Hitler was somehow resurrected and walked up to you and said, One
ought never to mistreat anyone Jewish! Im sure youd drop your jaw, and more
for what Hitler said than for the fact that youre chatting with a dead man. Youd
be so tempted to say, You bastard! You of all people should be lecturing us on
the respectful treatment of Jews! And yet this would be an ad hominem,
because what Hitler said is perfectly true, despite the fact that it qualifies as the
most hypocritical thing ever uttered.
In poker, the hand speaks for itself, no matter whos holding it. So too with
arguments and statements.
Worthy of embroidering on your sofa pillow:
The truth of a statement and the cogency of an argument are
independent of the author and their motivation and their
hypocrisies.
Distinction alert!
There is a difference between being hypocritical and being
inconsistent. The former just means one doesnt live up to ones
own claims, which does not entail that those claims are false.
The latter, however, means ones claims are contradictory, which
does entail that some of those claims are false. Thats why
134
Relevance
pointing out someones inconsistencies is not an ad hominem

attack.
It looks like an ad hominem, but its not [eeewww!]

Some attacks on a person are relevant. For example:
Bill Gates claimed that his Windows operating system and his
Internet Explorer browser will not function well unless they are
integrated. He claimed that building the browser into the OS is
the only way to maximize innovation with respect to internet
access. But of course Bill would make such claimshe would
have lost billions if Judge Jacksons ruling were to have stood;
he would have lost half his company and at most become CEO
of one of the baby bills.
Here, these attacks on the motives of Bill Gates do n o t amount to an ad
hominem. They are attacks on the credibility of the authority being appealed
to. Here, Bill Gates was appealing to himself as an authority, and thats
especially why criticizing what he says looks like an ad hominem.
Lets return to the basic format of an appeal to authority, to examine some subtle
differences.
Case: Imagine an instance in which someone argues that it would be a good idea
to start eating Shaker Oats in the morning. They claim this on the basis of
reading an article by a medical authority, Dr. E, who says that a balanced breakfast with Shaker Oats reduces heart attacks.
Question: What if we were to object by pointing out that the person giving this
argument had just purchased a bundle of stock in the Shaker Oats Company
would that be an ad hominem?
Answer: Yes. The persons possibly questionable motives [profits, not truth]
for appealing to Dr. E as an expert are irrelevant to Dr. Es status as an authority.
That Dr. Es claim serves the persons interests does not undermine that claim.
Question: What if we were to object by pointing out that Dr. E has a bundle of
stock in Shaker Oats Companywould that be an ad hominem?
Answer: No. Lets see why by detailing this appeal to authority, including its
crucial implicit premise.
135
Relevance
Dr. E said Shaker Oats reduces heart attacks.

[Dr. E would not have said this unless it were true.]
It would be a good idea to start eating Shaker Oats
And now we can see why talking about Dr. Es stock portfolio is relevantwe
are offering an alternative plausible explanation for Dr. Es saying what theyve
said. Weve given a reason to think that this implicit premise in this appeal to
authority is unacceptable. Were given a reason for thinking the argument
commits the fallacy of appeal to false authority.
Attacking the persons motivation for giving the argument, then, was an ad
hominem. But attacking the motivation of the authority appealed to was not
that was merely an argument that the person has committed the fallacy of
appeal to false authority.
Lets look at some more examples, using the case above : Ad hominem or not?
Question: What if we were to object to the argument by saying, What are you
doing reading articles on breakfast food? You should be writing those thank-you
notes!
Answer: Yes. Even if the motive of the person is to avoid writing the thank-you
notes, pointing this out is irrelevant to the cogency of their appeal to authority.
Question: What if we were to object by saying, Yes, I read that article too, but
wasnt the study funded by the Shaker Oats Company?
Answer: No. Here we would be offering a pretty good reason to think that the
fallacy of appeal to false authority has been committed. If the research was
funded by Shaker Oats, funds for further research would thereby likely be
contingent upon results beneficial to the funding source, and then the researcher
is not motivated solely by truththere is ample reason to believe that
impartiality might well have been sacrificed, such that there is a significant
possibility that Dr. E would have said what they did even if it were false.
Pop quiz: Ad hominem attacks or not?
Shakers Oats? Thats the cereal with Cindy Crawford on the

box, aint it? Im sure youre thinken about nutrition here,
dude!
And now what did you have for breakfast this morning?
Two donuts and a pot of coffee, thats what!
136
Relevance
So Dr. E says that a balanced breakfast with Shaker Oats

reduces heart attacks, does he? And what did you expect
from a member of the Board of Directors of the American
Grain Producers of America?!
Answers: The first two are ad hominems, the third is notits a

pretty good reason for thinking that Dr. E is a false authority.
Something to think about: Youve probably noticed how

people often take criticisms very personally, no matter how
constructive and warranted those criticisms might be. We may
have an explanation for this now: In view of how frequently
people use ad hominem attacks, its no wonder that those who
are criticized begin to overgeneralize a bit.
Guilt by association
This informal fallacy is a close cousin to the ad hominem, because it too involves
mudslinging. But there is a subtle difference.
Guilt by associationto place an author or a position in a
group of disreputable standing; by so doing, the audience is
tempted to attribute that same disreputable characteristic to
the author or position.
This fallacy is a favorite of politicians, especially during elections. Two classic
examples happened during the Bush-Dukakis presidential campaign.
George Bushs comment that Michael Dukakis was a card-carrying member of
the American Civil Liberties Union, was a very effective way of driving voters
away from his opponent. This comment was made to the media at a time when
the ACLU, a legal-defense organization, was arguing before a high court for
peoples right to child pornography. This invited the average person to make the
following inference:
1. The ACLU is currently defending child pornography.
2. Dukakis is a member of the ACLU.
3. Dukakis defends child pornography!
But, of course, one can be a member of a group and still not advocate everything
that group doesin fact, thats usually the case. In the context of Bushs comment, however, many listeners werent going to remember this truism. They
137
Relevance
were too busy thinking: That bastard! How can he defend something so
horrible! The dirt of the group was successfully rubbed off on Dukakis.
There was another, more subtle, instance of guilt by association occurring in
Bushs comment. Can you discover it?
Heres a hint: Who else were called card-carrying members? That phrase is
strongly associated with Communism. Senator Joe McCarthy kicked off many
hearings before his House Committee on Un-American Activities with the
question Are you now or have you ever been a card-carrying member of the
Communist Party? That phrase is so strongly associated with Communism, that
many people find it psychologically impossible to keep from dragging that notion
along when the phrase is applied to other matters. Check it out: Are you now
or have you ever been a card-carrying member of the Nobel Prize Laureates?
Even as great an honor as the Nobel Prize sounds dastardly now.
Another example was the famous Willie Horton television ad. The ad included a
mug shot of Willie Horton, a rather tough-looking black man. The ad described
how Horton had received week-end passes from prison through a furlough
program, while he was serving a life-sentence for murder. During a weekend
release, he kidnapped a couple, stabbed the man and repeatedly raped the
woman. The ad also pointed out that Dukakis was an advocate of the program.
This, then, invited the average viewer to make the following inference:
1. Dukakis favors a furlough program.
2. This program enabled Willie Horton to rape and attack.
3. Dukakis favors programs that enable prisoners to rape and attack.
Obviously, one can be in favor of a program or a policy and still not favor everything that unintentionally and unforeseeably results from it. Many viewers,
however, forgot thisthey were too busy falling victim to the fallacy, thinking:
That SOB! How can he favor what Willie Hortons done?!
BTW: You can watch all the classic campaign ads on the Web
at http://www.ammi.org Especially after completing this
chapter, critically review these ads and see if you can spot all the
fallacies.
People who use the fallacy of guilt by association are taking advantage of the fact
that people are not careful with their inferences. That a disreputable party is
associated with a belief or value or action is, by itself, irrelevant to its truth or
worth.
138
Relevance
Heres a strange, extreme case that may help you keep this in mind: Hitler was a
firm advocate of breathing and supported the idea that 2+2 = 4; but certainly, we
would never conclude that breathing and arithmetic are suspect as a result of this
association.
Disrepute is in the eye of the beholder, come time to judge whether or not the
fallacy of guilt by association has been committed. An example will illustrate
my point. I remember hearing a U.S. legislator give the following little speech
early in the Clinton years:
The issue before us today is whether to pass the Presidents
health-care bill, a copy of which I have in my hand. I want you
to know that I am against this bill. I dislike the way the First
Lady has gotten involved in this legislative process. Who
elected her, anyway?
By snuggling the bill up to the Hilary Clinton, the legislator was hoping that the
audiences dislike for her would rub off on the bill. The fallacy worked only on
those who viewed the First Lady unfavorably. And so it is that youll find
religious groups arguing against something by pointing out that the atheists are in
favor of it, and atheists arguing against something by pointing out that religious
groups are in favor of it. In either case, though, its still the same fallacy.
Beware: When people point out cases of strange bedfellows, they are often
committing the fallacy of guilt by association.
BTW: There is such a thing as praise by association. Its just

as fallacious as guilt by association; its just never become as
famous.
Pop quiz: Explain the following cases of guilt by association:

1. RU-486 should not be permitted in the U.S. Roussell-Uclaf,
the developer of the drug, is controlled by the giant
international firm, Hoechst. Its parent company produced
cyanide for the Nazi death camps. This is nothing less than
chemical warfare against the unborn.
2. From a script for a television ad:
ANNOUNCER: Mark Taylor has some more problems
to clear up before he runs for any office. First, Taylor
fought to preserve discriminatory racial quotas.
(Footage of Taylor with black mayor Campbell.)
139
Relevance
ANNOUNCER: Then, he was solidly endorsed by the

homosexual newspaper Southern Voice. (Visual of the
cover of Southern Voice.) [Harpers, May 1999]
3. From an April 22, 1999 press release sent by the Republican
National Committee to talk-radio stations across the country:
Today, as America celebrates the twenty-ninth
anniversary of Earth Day, were reminded of the
writings of two of the scariest environmental extremists
around: Vice President Al Gore and the Unabomber,
Ted Kaczynski.
Gore: Industrial civilizations great engines of
destruction still seduce us with a promise of fulfillment.
Unabomber: Very widespread in modern society is the
search for fulfillment. . Butfor the majority of
people, [technology] does not bring completely
satisfactory fulfillment.
Gore: The increased productivity of assembly
linesrequires many employees to repeat the identical
task over and over until they lose any feeling of
connection to the creative processand with it their
sense of purpose.
Unabomber: A theme that appears repeatedly in the
writings of the social critics of the second half of the
twentieth century is the sense of purposelessness that
afflicts many people in modern society. [Harpers, July
1999]
4. In response to a debate on establishing a National Biological
Surveya federal agency that would deploy volunteers as
part of an effort to monitor plants and wildlife
nationwideRepresentative Jack Fields (R-TX) said the
following:
More than likely you are going to have self-interested
groups coming in as volunteers. And in essence we
are creating an environmental Gestapo that will go on
peoples private property. [The Amicus Journal,
Summer 1999]
140
Relevance
Appeal to ignorance
Until someone proves to me that there was no conspiracy to kill
John F. Kennedy, Im going to continue believing that the
Warren Commission was wrong in their finding against the
conspiracy theory. In fact, Im inclined to think theyre covering
up the conspiracy, because theres no reason to think otherwise.
This is a typical instance of our next fallacy.
Appeal to ignoranceto argue for a position by appealing to
ones lack of reasons for believing its denial.
The fallacy of appeal to ignorance has two basic formats. Here is one of them.
We dont know that statement S is false. [or]
We have no reason to think that S is false. [or]
Its not been proven that S is false. [etc.]
S is true. [or]
S is probably true.
One can go out on the limb with an appeal to ignorance and conclude that something is the case, or one can play it a bit safer and merely conclude that something is probably the case. Either, however, is fallacious.
Its surprising how often people use the appeal to ignorance. Sometimes its so
quick, its hardly noticeable:
Interviewer: And why should we hire you from among
our pool of candidates?
Interviewee: Well, why shouldnt you?
Bob: Why should we go to Shortys for lunch?
Mary: Well, why shouldnt we?
The interviewee is in essence arguing that a lack of reasons against hiring them
would itself be a reason for hiring them. Clarified like this, however, it really
does sound suspiciousand rightly so!
The other variation of the appeal to ignorance is just as popular; for example:
You havent a shred of evidence for thinking that I
cheated on this, so it must be my own work.
141
Relevance
The basic format of this variation is:

We dont know that statement S is true. [or]
We have no reason to think that S is true. [or]
Its not been proven that S is true. [etc.]
S is false. [or]
S is probably false.
Notice that both formats of the appeal to ignorance do exactly that they
appeal to ones ignorance, ones lack of knowledge, ones lack of evidence,
ones lack of reasons for believing something.
This is essential to the fallacy of appeal to ignorance: Its not just an ignorant
appeal. Its not just an ignoramus making an appeal. Its an appeal to the fact
that one is ignorant.
Whats exactly wrong with the appeal to ignorance?

Let me briefly detail some of the major problems with the appeal to ignorance.
This will enable you to see just how totally fallacious it is.
If indeed the appeal to ignorance were a way of meeting the relevance
conditioni.e., if such appeals actually provided some evidencethen the
following might be a cogent argument.
I have no reason to believe that there are not exactly
1111 bottles in the recycling shed.
There [probably] are exactly 1111 bottles in the
recycling shed.
Lets say that I havent even looked into the recycling shed! Certainly, if I
havent even looked yet, that would explain why the premise above is true. And
yet Im in a position to draw conclusions about the exact number of bottles in
there? Not a chance!
Things get worse. The following would also be a reasonable argument:
There [probably] are exactly 1112 bottles in the
recycling shed.
Combining these two arguments, wed get:
142
Relevance

There [probably] are exactly 1111 and 1112 bottles in
the recycling shed.
And this is a contradictionthe appeal to ignorance entails a contradiction.
But things get even worse. Remember, I havent even looked inside the shed.
My ignorance about the contents of the shed is quite robust. As a result, the
following argument would also be possible, if the appeal to ignorance were permissible:
I have no reason to believe that there are exactly
There [probably] are not exactly 1111 bottles in the
recycling shed.
By now, you probably see where Im going:
I have no reason to believe that there are exactly
There [probably] are and are not exactly 1111 bottles in
the recycling shed.
And we have yet another contradiction.
The appeal to ignorance is a completely loose cannon, pumping out all sorts of
contradictions.
And another thing!

The appeal to ignorance cites ones ignorance as a reason to believe that something is either true or likely true. Just imagine if this actually worked! Just imagine if ones ignorance were actually a source of knowledge! This would change
everything, especially your strategy for learning. If ignorance were a source of
knowledge, then the way to maximize your knowledge would beto stay as
stupid as possible.
Add that to the list of reductios weve heaped on the appeal to ignorance.
143
Relevance
I hope its clear now why the appeal to ignorance is a fallacy. From ignorance
can only flow ignorance. The only rational thing to do in a state of ignorance is
to withhold belief. If one has no reason to either believe or disbelieve something,
the only rational thing to do is to not believe.
WOW, that was fast!

Appeals to ignorance can happen so quickly, you can hardly see them. But after
a bit of practice, you should be able to spot even the sneakiest, most condensed
ones. Here are a couple of good ones that I bumped into recently:
I dont see any reason why I shouldnt hold students accountable
to the basic rules of grammar as I grade their philosophy papers.
Vendor: Would you like to sign up for a credit card?
Student: Ive got one already.
Vendor: Theres no rule that says you cant have two.
So be on the lookout for phrases such as I dont see any reason why, I dont
see why not, or It didnt say we couldnt. Theyre great indicators of the
fallacy of appeal to ignorance.
For all I know, it may be true

We frequently hear people say, For all I know, it may be true. People intend
this to be shorthand for, For all I know, and I dont know anything, it may be
true.
There are those who would argue that this appeal to ignorance is not fallacious,
because the conclusion drawn is so minimal. We arent claiming here that something is true or probably true or even that there is a significant chance of it being
true. Only that it might be.
Lets investigate such minimalist appeals to ignoranceare they cogent?
We have no reason to think that statement S is false.
S may be true. [or]
S is possibly true.
We have no reason to think that statement S is true.
S may be false. [or]
S is possibly false.
144
Relevance
Ironically, the best person to argue against these appeals to ignorance is a fiveyear old. Have statement S be, The sum of the interior angles of a [Euclidean]
triangle equals 180 degrees. About the most a five-year-old knows about triangles is that they look a certain waypointy. Kids can use this knowledge to
distinguish triangular blocks from the square ones and the round ones. But
they certainly have no reason to believe that the sum of the interior angles of a
triangle equals 180 degrees. And now were ready for our reductio from the fiveyear-old:
I have no reason to think that the statement The sum of
the interior angles of a [Euclidean] triangle equals 180
degrees is true.
It may be false. [or]
Its possibly false.
But, of course, it is not possibly false that the sum of the interior angles of a
triangle equals 180 degreesits a necessary truth; its logically impossible for
this to be false.
Let me give you another example: You dont know anything about my father;
youre completely ignorant about him. If minimalist appeals to ignorance were
permissible, however, you could make the following inference:
I have no reason to think that Kevins father wont visit
him next weekend.
Kevins father may visit him next weekend. [or]
It is possible that he will visit Kevin next weekend.
But this just doesnt follow. Eds dead; its physically impossible for him to
visit.
Nothing follows from ones ignorancenot even knowledge about mere farflung possibilities. In a sense, you need just as much reason to believe that a
possibility exists as you need to believe that a probability exists.
Perhaps when people say For all I know, it may be so, they just mean to be
saying, For all I know, it may be so or it may not beI just dont know. If
so, then they are not arguing by appeal to ignorance. Instead, they are just
expressing their ignorance in a very misleading way.
Discussion: Why does the minimalist appeal to ignorance sound

so tempting? For example, lets say Im standing in front of the
township recycling shed again [a little utility shed, about the size
145
Relevance
of single-car garage] and I havent opened the door yet. Why

does it seem so cogent for me to argue as follows: For all I
know, there may be 1111 bottles in there?
Answer: The reason my inference seems so cogent is that it is
not an appeal to ignorance at all. Im not arguing from the
premise that I have no reason to believe that there are not exactly
1111 bottles in the recycling shed, to the conclusion that there
may be exactly 1111 bottles in there. Instead Im appealing to
what I know about sheds and, in particular, the one in front of
me. And, based on all I know about it [its dimensions, its
physical properties, its physical laws], I know that its logically
and physically possible for there to be exactly 1111 bottles in it.
Namely, for all I know about basic science and logic, I know it
would break no law of physics nor be a contradiction for there to
be 1111 bottles in the shed.
So, what may appear to be an appeal to ignorance is actually an
appeal to basic background beliefsbackground beliefs that
are so basic, we dont even have a sense that they are acting as
our acceptable reasons for our conclusions. But thats quite
permissible.
Something to think about: One of the most fundamental views

in our judicial system is that one is presumed innocent until
proven guilty beyond a reasonable doubt. But this looks like a
simple case of appeal to ignorance: We have no reason [yet] to
think the accused is guilty; therefore, the accused is [probably]
innocent. Does our entire legal system rest on the fallacy of
appeal to ignorance? Or can you find an analysis of our
innocent-until-proven-guilty policy that will save our legal
system?
Appeal to popularity
How often have you seen something advertised as the leading brand or the
nations number one or the best selling? How often has the person taking
your order at the restaurant answered your questions about the quality of the food
by saying A lot of people order it? How often do you bump into the reply,
Oh, no, weve always done it this way! to your request for change in policy? I
would wager, a lot! These are all instances of our next fallacy:
146
Relevance
Appeal to popularityusing the fact that many people

believe or value or do something as a reason to think that its
correct.
The appeal to popularity is fallacious because there are too many alternative
explanations for why a belief, value, or action is popular besides its being correct.
Democracy can only tell you what people in fact believe, value, and do; it does
not reliably tell you what people ought to believe, value, and do.
Indeed, the worn-out parental reply shows exactly whats wrong with the appeal
to popularity: If everyone decided to jump off a bridge, would you do it too?!
Pop quiz: What kind of criticism was that?
Here are some more examples:

Other countries are free to change their constitutions when it
becomes necessary. In fact, with the exception of Luxembourg,
Norway, and Great Britain, there is not one advanced industrial
nation that has not thoroughly revamped its constitution since
1900. If they can do it, why cant we? Why must Americans
remain slaves to the past? [Your Constitution is Killing You,
Harpers, Oct. 1999]
VIEWER: What qualifies you to comment on political affairs?
RUSH LIMBAUGH: One simple thingratings. [Dec. 1995]
The judicial system is one of our most consulted sources for
direction concerning values and ethical judgment; so legal
judgments and distinctions are definitely worth considering
when trying to figure out whats morally right or wrong.
There are variations on the appeal to popularity that you should be on the alert
for:
Appeals to tradition:
Marriage has historic, religious, and moral content that
goes back to the beginning of time, and thats why I
think a marriage is as a marriage has always
beenbetween a man and a woman.
147
Relevance
The fallacy of appeal to popularity can involve a claim about what a lot of people
do currently or what a lot of people have done over timeno big difference.
Appeals to past practices:
Oh, Im afraid weve never done it that way before.
Thats just not the way we do things around here.
Its our policy to do it this way.
Its not been our policy to do it that way.
Appeals to precedent:
Weve already began doing it this way; so we
better stick with it.
We did it this way last week; if were going to
do it that way, we should have begun last week.
We allowed this last time; so we must allow it
again.
Once again, we see that informal fallacies can be very condensed and happen
very quickly. We really must be on our guard.
The appeal to precedent is especially compelling to those who think the judicial
system is a reliable indicator of whats true or right. But when an action or belief
is just plain wrong, it should be abandoned, no matter its current or past popularity. Remember: Slavery was once popular, commonly practiced, and a legal
precedent.
Appeal to force
Yeah? You got a problem with that?!
Appeal to forceto use a threat as a reason for someone to
adopt or change their beliefs, values, or intentions.
A threat is a claim that one is conditionally placing an undesirable price on someones belief, value, or intention. The classic example is If you dont give me
your wallet, Ill kill you. The mugger has put an undesirable price on your
intention to keep your wallet. The mugger certainly has not given you a reason
148
Relevance
as to why they rather than you ought to have your wallet; they have not given you
a reason to think that you are somehow the wrong person for the money and that
its been a mistake all along that youve had it instead of them. Thats why the
appeal to force is a violation of the relevance condition for cogency.
Sometimes the threats involved in the appeal to force can be quite veiled. For
example, when a boss says something like A few years ago we had an employee
that thought the same way as you dohe wasnt here long. The lone fact that
employees are getting fired for having a certain belief is no reason to think that
the belief is false.
Pop quiz: Find the fallacy:

I am writing in response to the editorial submitted by the
Womens Resource Center, objecting to the storage of
nuclear waste on Prairie Island. Since none of the
personnel at the WRC has the least bit of education,
training, or experience in the issue of nuclear power
safety, their unsupported opinion on the issue is of little
value to the debate. Second, as a long-time contributor
to the WRC, I would prefer its personnel to stick to
editorializing about things they know best, viz., the
terrible harm of domestic violence. The WRC must
decide between maintaining their current position on
Prairie Island or maintaining their current donor list.
Answer: As much as the first half looks like an ad hominem,
its not; its simply pointing out that the original editorialist from
the WRC is a false authority on the issue of nuclear waste
storage. The fallacy of appeal to force occurs in the latter half,
where the respondent threatens to stop making donations unless
the WRC changes, or at least stops expressing, its opinion on the
Prairie Island case.
Informal fallacy round-up

Weve studied quite a few informal fallacies in the process of discussing the
acceptability and relevance conditions of our ARG conditions for cogency. Lets
gather them all up into a nice list for final review.
To maximize your understanding of these fallacies, add your favorite
examples of each one.
149
Relevance
Informal Fallacies
EquivocationSupporting a conclusion by shifting from one
meaning of an ambiguous term to another.
Slippery slopeTo argue from the fact that one cannot determine
the exact boundaries of a terms extension, to the conclusion that a
quite liberal use, or even any extension, of the term is appropriate.
Appeal to false authorityWhen a statement is accepted because
it is expressed by an expert claiming to have informed,
privileged knowledge; however, one has reason to believe that the
expert is unreliable, viz., that it is quite possible that they would
express the statement even if it were not true.
Begging the questionWhen one of the premises is either a
disguised version of the conclusion or presupposes the truth of the
conclusion.
False analogyWhen there is a relevant difference between two
cases, such that what is claimed to be true of one case is no reason
to believe it is true of the other.
Straw manTo criticize or attack not someones actual argument
or conclusion or position, but rather a misrepresentation of it.
Ad hominemWhen derogatory charges are made against the
author or advocate of a position or argument, using those charges
as reasons to claim that the position is false or the argument is not
cogent.
Guilt by associationTo place an author or a position in a group
of disreputable standing. By so doing, the audience is tempted to
attribute the same disreputable characteristic to the author or
position.
Appeal to ignoranceTo argue for a position by appealing to
ones lack of reasons for believing its denial.
Appeal to popularityUsing the fact that many people believe or
value or do something as a reason to think that its correct.
Appeal to forceTo use a threat as a reason for someone to
adopt or change their beliefs, values, or intentions.
150
Relevance

Its time to assess your understanding of the acceptability and relevance
conditions and the informal fallacies associated with flunking them. So launch
the CT software and click the menu button for Whad Ya Know About CT?
This time, work on Level 2, to verify your grasp of the fundamental concepts
weve covered in the last four chapters.
Working on Level 2 will also verify your knowledge of what the informal fallacies are. Being able to spot them as they sneak up on you in everyday discourse,
however, is a skill that takes time to develop.
Identifying informal fallacies is tough! But, then, thats only naturalotherwise
they wouldnt work so well as rhetorical tricks, faking people out, getting them to
believe things they have no good reason to believe!
So when youve reached competency at Level 2, click the Informal Fallacies
menu button and just keep plugging away until you can reliably identify the fallacies. The better you are at spotting them, the less often you will fall victim to
them.
151
Groundedness I: Categorical Logic
8. GROUNDEDNESS I: CATEGORICAL LOGIC

Lets now examine the third and last condition for cogencythe groundedness
condition.
What is it for the premises to provide adequate grounds for the conclusion?
Its when the premises provide adequate reason to believe

the conclusion.
Its when the truth of the premises would make the

conclusion at least probably true.
So how do premises do this?

From the most coarse-grained point of view, there are two ways in which
premises can provide adequate grounds for their conclusions.
They can make conclusions probably true by virtue of guaranteeing the truth of
those conclusions. These, of course, are deductive arguments. The groundedness relation between premises and conclusion in deductive arguments is, as you
well know by now, validitydeductive implication or entailment.
On the other hand, premises can make their conclusions probably, but not
necessarily, true. These, of course, are the inductive arguments.
We will be studying the groundedness condition as it applies to deductive arguments first and then, later [viz., when I write it], dedicate an entire section of this
etext to the study of how premises ground conclusions by means of inductive
arguments.
Our research into how premises of deductive arguments entail their conclusions
will be broken into two parts, corresponding to the two classic forms of deductive
argumentationcategorical logic and propositional logic.
Categorical logic is the study of deductive arguments that are made up of a
certain kind of statements, viz., categorical statements. Propositional logic is
the study of deductive arguments that are made up of another certain kind of
statementsviz.,well, duh!
Now that weve completed a birds-eye view our topics, lets begin our examination of categorical logic.
152
Categorical logic
Everyone at the party had a good time.
No one at the party got sick.
Although, some people celebrated a bit too much.
But most people did not.
Everyone got an invitation, telling them to bring a dish
to pass. And once you know, youve got no excuse. So
no ones got any excuse for coming empty handed.
We typically talk this way. I bet you never realized that it is the language of
categorical logic.
If categorical logic is the study of deductive arguments that are made up of categorical statements, what exactly are categorical statements?
Categorical statements are statements of relations of
inclusion or exclusion between two categories of things.
OK. But now, what are categories?
Categories are classes or types or kinds of objects, events, states of affairs or
propertieskinds of things in general.
Categories can be very broad, e.g., the category of all things in the universe. Or
even broader, such as the category of all possible things in the universe, for all
timewow, thats a pile of stuff.
Categories can be much narrower too, e.g., the category of furniture. Or the category of chairs. Or of wooden chairs. Of wooden chairs sat on by you. Of
wooden chairs sat on by you on a Tuesday. With gum under them. Put there by
you [and cut that out]! A category can get so specific in its description that
nothing fits it; so the category has nothing in it; it has no members.
OK, thats what a category is. What, then, are relations of inclusion and
exclusion? Between two categories, there can be only four possible relations
of inclusion and exclusion. Lets use the symbolic categories of things that are
A and things that are B to illustrate:
All As are Bs.
No As are Bs.
Some As are Bs.
Some As are not Bs.
153
This covers all the possibilities. Lets take a closer look at each type of categorical statement and introduce some terminology and some notation.
All As are Bs.

This is the relation of total inclusionall things that are A are included in the
class of things that are B. This categorical statement is the universal affirmative, and for good reasonit makes a universal claim about all As, affirming
that they are Bs. This categorical statement is called the A categorical statement.
It is symbolized in the following way: A a B. The lowercase a indicates the kind
of categorical statement, and the uppercase letters indicate the categories
involved in the relation.
No As are Bs.
This is the relation of total exclusionthings that are As are excluded from the
class of things that are Bs. This categorical statement is the universal negative;
again, for good reasonit is making a claim about all As, that all of them are
excluded from being Bs. Its called the E categorical statement, and its symbolized as A e B.
Some As are Bs.

This is the relation of partial inclusionsome things that are As are also Bs.
This is the particular affirmativethe I categorical statementand it is symbolized as A i B.
Some As are not Bs.

This is the relation of partial exclusionsome things that are A are not also B.
This is the particular negativethe O categorical statementsymbolized as
A o B.
Basic format and review

Youve probably noticed the shared format of all four categorical statements:
A v B, where v stands for one of the four representative vowels. The first
category of the relation is called the subjectits the topic of the categorical
statement. The other category is called the predicate. The categorical statement
154
is said to predicate or attribute a property to the subject. For example, All

humans are mammals predicates being a mammal to human beings.
Statement
Symbol
Type
Relation
All As are Bs.
AaB
Total inclusion
No As are Bs.
AeB
Total exclusion
Some As are Bs.
AiB
Partial inclusion
Some As are not Bs.
AoB
Partial exclusion
From ordinary language to standard categorical formand back

There are dozens of ways of making categorical statements in ordinary language.
Our goal here is to be able to identify as many of them as possible, so that we can
easily translate ordinary language into our standard categorical form. Being
good at this will help you to understand exactly what relation is being expressed
by a statement. This skill is also the first step towards being able to assess the
validity of categorical arguments as you encounter them in daily life.
Here, then, is a good start at a translation manualits not exhaustive, but it
should enable you to extrapolate to ones Ive missed. Ill discuss some of these
phrases in a moment. For now, let me just warn you that you should eventually
be familiar with all of these phrases and know how to put them into their
standard categorical form and notation.
A Statements
Inclusive Statements
All As are Bs.
Any As are Bs.
Every A is a B.
As are all Bs.
As are always Bs.
AaB
AaB
AaB
AaB
AaB
155
Each A is a B.
Whenever something is A, its also B.
If its an A, then its a B.
An A is a B. [the is of predication]
Anything is an A unless it is a B.
A until B.
Its not an A, unless its a B.
Its not an A, without being a B.
AaB
AaB
AaB
AaB
non-B a A
non-B a A
non-B a non-A
non-B a non-A
Exclusive Statementsfor multi-membered category

Only As are Bs.
BaA
None but As are Bs.
BaA
Only if its an A is it a B.
BaA
As alone are Bs.
BaA
The only thing thats a B is an A.
BaA
None other than As are Bs.
non-A e B
None except As are Bs.
non-A e B
Exclusive Statementsfor single-membered categories
Only A is B.
A a B and B a A
A is the only B.
A a B and B a A
Exceptive Statements
All except As are Bs.
All but As are Bs.
non-A a B and A e B
non-A a B and A e B
Identity Statements
An A is identical to B.
An A is a B. [the is of identity]
A a B and B a A
A a B and B a A
There are some terms in ordinary language that you would never think of as
technical, but they very much are. The first one on our list is is.
Remember how silly it sounded when Bill Clinton, during his deposition in the
Paula Jones case, said, That depends on what the meaning of is is? Ironically, Clinton was right! Compare A Homo sapien is a mammal and A
bachelor is an unmarried male. The first is is called the is of predication.
All that is stated is that all Homo sapiens are mammals, not that all mammals are
also Homo sapiens. Whereas the second is the is of identityall bachelors are
unmarried males and all unmarried males are bachelors too. The is of predication states a single A categorical statement; the is of identity states two A
categorical statements.
The next technical term on the list is unless. If I say, The dishes are clean,
unless they were used for lunch, Im making a claim about all the dishes that
156
have not been used for lunch, viz., theyre all clean. If I tell you that the picnic is
on unless it rains, Im saying that at the time its not raining, the picnic is a still
proceeding as planned. Here, then, is a rule for how to translate the unless into
standard categorical form: The unless negates the subject of an A statement;
and the other category, then, must be the predicate.
Without often works the same way as unless: To say, One cannot graduate
without taking 128 credits, is to say that all those who do not have 128 credits
are not able to graduate. You cant make an omelet without breaking eggs,
means that all the times when youre not able to break eggs, youre not able to
make an omelet.
Only, only if, and none but are extremely tricky and important terms to
keep straight, especially around tax timeOnly those below such-and-such
income are eligible for this deduction, None but business expenses can be
deducted on this form, One is eligible for this deduction only if one paid taxes
the previous year.
The rule to follow with only, only if, and none but is that they refer to the
predicate. So, Only the strong survive, means that all those that survive must
have been strong. [It does not mean that all those that were strong survivedit
takes more than strength to surviveit takes some dumb luck too.]
The exception to this rule [as noted on our list above] is when one is talking
about categories in which there is only one member. For example, if I say that
only Bob showed up for the meeting, I mean that all people who are Bob
attended the meeting and all the people who attended the meeting are Bob. This
sounds kind of funny, but it clearly states the point, and thats the important
thing; so get used to sounding funny when putting statements into standard
categorical form.
A strange cousin to the none but is the all but. This behaves rather uniquely,
so be awareits another one of those cases in which there are two categorical
statements packed into one sentence. If I tell you that all the members except
Bob have paid their dues, Im telling you that all those other than Bob [all the
non-Bobs] have indeed paid, and Bob has notthats why all except Bob have
paid.
Rules and regs: Take a look at the instructions to the 1040

Form or the language used in any kind of regulations. What
youll find is a shower of categorical statements expressed in
every which way. If you cant tell which categorical statement is
being expressed, you wont know, for example, if youre eligible
for the deduction, if youre in violation of the law, or if youre in
157
line for serious cash or serious jail time. If more people knew
their way around categorical statements, fewer people would be
heading to H&R Block every year. Sometimes, wording is
everything, and this is one of those times.
E Statements
No As are Bs.
Never are As Bs.
As are never Bs.
As are not Bs.
There isnt a single case of A that is B.
AeB
AeB
AeB
AeB
AeB
I Statements
Some As are Bs.
A few As are Bs.
At least one A is B.
There are ABs.
Frequently As are Bs.
Most As are Bs.
AiB
AiB
AiB
AiB
AiB
AiB
O Statements
Some As are not Bs.
Few As are Bs.
Seldom are As Bs.
Hardly any As are Bs.
AoB
AoB
AoB
AoB
I and O Statements
Only some As are Bs.
Only a few As are Bs.
Almost all As are Bs.
Not quite all As are Bs.
[Exactly] XX% As are Bs.
Only XX% As are Bs.
Almost no As are Bs.
A i B and A o B
A i B and A o B
A i B and A o B
A i B and A o B
A i B and A o B
A i B and A o B
A i B and A o B
These various ways of finding E, I, and O categorical statements in ordinary

language are quite straightforward. I only want to draw your attention to the last
158
little group, which are unique because they are two categorical statements rolled
into one phrase. Be on the lookout for these cases: What makes a statement such
as Only some As are Bs true is that indeed some As are Bs, but only some
some are not.
Public notice: Have this translation manual down coldbe

able to identify categorical statements no matter how you find
them disguised in ordinary language.
Some important things to note about some statements

There are some unique features of categorical statements that you must keep
track of.
If someone says, Some aardvarks are blind, they are saying that at least one
aardvark is blind. More exactly, they are saying that there exists at least one
aardvark and it is blind.
When one makes an I categorical statement, one is going on record as claiming
that there exists at least one member in the subject category, who happens
also to be a member in the predicate category.
This is the existential commitment behind every I categorical statement.
To keep this in mind and really capture the meaning of Some As are Bs, get
into the habit of unpacking it as, There exists at least one A and its a B.
You dont have to say it this wayyoud sound a bit snooty if you didbut try
to think of it this way.
Question: Isnt there also an existential commitment with the A categorical
statement?
Answer: No. And heres why: We want to claim that unicorns have only one
horn, right? Thats true by definition of a unicorn, after all. But notice what
wed be saddled with if existential commitment held for the A categorical
statement as well. Wed be committed to the existence of unicorns [and then,
wed simply be committed!]. Many of the universal claims we make are purely
hypothetical; we make them completely cognizant of the fact that there are no
members in our subject category.
Soexistential commitment holds for the I statements, but not for the A statements.
159
To highlight another feature of the I categorical statement, let me ask you this:
On the basis of the fact that some As are Bs, can one validly infer that some As
are not Bs?
Answer: No. And why not? Well, even though the frequent explanation for
why some As are Bs is that only some are, thats not always the case. It could
very well be that what explains the fact that some As are Bs is that all of them
are. Surely, if all existing As are Bs, then some of them are! Some dinosaurs are
extinct, arent they?! OK; so you just cant infer an O categorical statement from
its corresponding I categorical statement.
The third feature to note about I categorical statements is the tremendous loss of
information that occurs when you use them to translate ordinary language
into standard categorical form. Take the following list of statements
commonly found in ordinary language:
A few As are Bs.
At least half the As are Bs.
Most As are Bs.
Many As are Bs.
From the point of view of categorical logic, these all get translated merely as
Some As are Bs. Information about differences in the proportions of As that
are Bs is lost.
Moreover, if I were to say, Some things are As despite their being Bs,
categorical logic only picks up on the fact that some As are Bs; it totally misses
the explanatory information carried by my saying that these As are As in spite
of, not because of, their also being Bs.
There is nothing that will ever be more powerful in its capacity to carry
information than ordinary language. No language of computers, mathematics, or
logic is ever going to come up to ordinary languages potential for telling us
about the world and relaying subtle differences and details.
So why are we studying this categorical language of logic, if its so incredibly
poor at capturing the ordinary language of daily discourse and debate? Good
questionI only hope I can give you a good enough answer:
While ordinary language has the greatest capacity for conveying information,
what means does it have for assessing the validity of arguments? Well, none,
really. The only assessment tool at your disposal while using only ordinary
language is whether an argument sounds good or not. And certainly, how
someones argument sounds, how it strikes your untutored intuitions, is hardly a
reliable gauge of validity.
160
On the other hand, categorical logic will provide us with a very accessible and
reliable set of tools for assessing the validity of arguments. In order to apply
those tools to arguments we encounter in daily discourse, however, the
arguments must first be translated into standard categorical form. And then, after
the assessment of validity is made, the arguments must be translated back into
ordinary language.
As in all of life, there is a trade-off: In order to better assess the validity of our
daily arguments by means of categorical logic, we must sacrifice some
information. If were unwilling to do the latter, were unable to do the former.
And similarly for some are not statements

The O categorical statement has the same existential commitment behind it as
does the I statement; so get in the habit of thinking of A o B as, There exists at
least one A and its not a B.
And just as one cannot infer A o B from A i B, so too one cannot infer A i B
from A o B. And why not?
Answer: Because what explains why A o B might well be that no As are Bs.
And if A e B, it surely cannot be the case that A i B. Remember: Its certainly
true that some dinosaurs are not living, but that does not entail that some of them
are living.
And lastly, be aware of the loss of information when putting ordinary language
into standard categorical form for O categorical statements:
A few As are not Bs.
At least half the As are not Bs.
Most As are not Bs.
Many As are not Bs.
Some things are As despite not being Bs.
These all get rendered as simply A o B.
Make sense with your symbols!

So far, weve been discussing the four categorical statement types and weve
become pretty familiar with our way of symbolizing them: A a B, A e B, A i B,
and A o B.
161
But lets not get too set in our ways here, thinking that there is something
special about the letters A and B, and about having A always stand for the
subject and B always stand for the predicate! That couldnt be farther from the
truth and function of symbols!
What is the function of a symbol? To stand for something and help us
remember what it stands for. The moment a symbol fails to do this for you is
the moment you should ditch it. So make your symbols make sense, so that
its easy to see what they stand for. Lets discuss some examples of putting
categorical statements into notation:
Aardvarks are boisterous.
AaB
Think of your old alphabet book, as youre choosing your symbols: A is for
aardvark, with its long, ugly snout; B is for boisterous, must you shout like a
lout?
Boisterous creatures are all aardvarks.
BaA
Symbolizing this one as A a B would be asking for troublelike using the green
light to stand for Stop.
The aardvark is one ugly animal.
AaU
Weve already settled on A for aardvark; so switching to some other symbol

now would be just plain misleading. U is the obvious choice as a symbol for
the predicate: B isnt going to help you remember ugly, and, besides, its
already standing for boisterous, in our context.
The aardvark peed on the rug.
AaP
What better symbol is there for peed?! Even if the original sentence were,
The aardvark urinated on the rug, ours would have been a better choicewhy?
Alice peed on the rug too.
AL a P
When your first choice is already spoken for, include or move to the second
letter. If thats taken, move to the third, etc.
Slammers are slackers.
SLAM a SLAC
See? M a C was not a great choicewhy?

Baboons are authoritarian.
162
BA a AU
And now youve got the hang of it!

Remember, logic shouldnt be some completely abstract geeky game of jerking
symbols around. Logic is a tool that enables us to assess the cogency of daily
argumentation and inference. So get into the habit of using the symbols and the
logical notation were studying, to accurately and meaningfully represent the
flood of arguments you must deal with.
A new symbolthe bar

When we were discussing the A statement earlier, we studied what were called
exceptive statements, e.g., All the animals except the aardvarks were outside. This sentence actually has two categorical statements packed into it: All
the non-aardvarks were outside and No aardvarks were outside. If we put
this into our notation: non-A a O and A e O.
We can get even more abbreviated than this, however. And we definitely want to
develop speedy notation, because we are going to be using it to record arguments
as fast as people rifle them off. So lets introduce a bit of notation for non-.
Lets use a bar written over the symbol weve chosen for the category. Non-A,
then becomes A.
This will really come in handy when working with unless, and its amazing
how often you encounter arguments using this word. For example: I wont
refinance my home unless the interest rate drops at least two points. Instead of
symbolizing this as non-D a non-R, we can simply use D a R.
Lets practice a little notation together

Im going to pick some strategic examples here and discuss some of the quirky
things that can crop up as you put sentences into standard categorical form
notation. Lets begin with:
Every computer comes with the OS preinstalled at the factory.
Whats the predicate? Is it the category of things with an OS? Is it things with
an OS preinstalled? Or is it things with a factory-preinstalled OS? Its the third.
Note how these are three very different categories, increasing in order of specificity. Our job is to accurately and completely represent the statement and its categories. So a good way to symbolize this one would require a bit longer notation:
C a OSFP.
163
A DNR worker dropped by today and asked permission to take a

survey of our woods.
What type of categorical statement is this? Once again, the trick is to pinpoint
the predicate. Is it the person who dropped by today? The person who asked
permission to take a survey of our woods? No, its the person who jumped
through both hoopsthey asked permission and did it by dropping by today. So:
DNR i D. [We can use D here to stand for the whole enchilada in the
predicate.]
A person who doesnt exercise and yet has a high caloric intake,
smokes, and has a history of heart trouble in their family is
risking early death, unless they alter their life style considerably.
Wow! This is a long-winded subject category. But dont let that scare you;
just keep track of all the aspects of the category by means of your notation:
ECSHA a R.
Watson and Crick won a Nobel Prize.
WC a N. Categories can easily have unique membersonly Watson and Crick
are in the category of things that are Watson and Crick. Sometimes you have to
talk a little funny, then, when you put things into standard categorical formAll
persons identical to Kevin are those who wrote this critical thinking etext.
Now is a good time to act, would be All times identical to now are times that
are good to act.
Only you know and I know.
This line from an old Dave Mason tune is good practice on how to deal with
only. Remember, when it is establishing a relation between the categories,
only refers to the predicate. So, we should symbolize this lyric as K a U&I.
But we should also notice that you and I and those who know are categories
with identical members, viz., you and I. So, to capture the full meaning of this
little phrase, we would need K a U&I and U&I a K. This was a tricky one!
[Leave it to a philosopher to suck the lyrical value right out of the lyrics!]
Watson and Crick discovered the helical structure of DNA.
WC a D and D a WC. Can you explain why this sentence expresses both these
categorical statements?
He who knows only his own side of the case, knows little of that.
164
This is a famous quotation from John Stuart Mill. I include it to illustrate that,
like with most rules, our rule about how to handle only has exceptions. In the
case of this famous line from Mill, the only is not functioning to establish a
relation between the categories; its instead part of the description of one of the
categories. The subject of Mills categorical statement is, Those people who
know only their own side of the case. The predicate category is, Those people
who know little of their own side of the case. The claim, then, is that all those
in the first category are also in the latter: KOT a KLT.
No friend is better than a fair-weather friend.
Heres another tricky one, similar to Mills one-liner. We are accustomed to the
word no working to make the statement into an E categorical statement. But, if
we treat this no as doing this, then we would interpret the sentence as saying,
None of ones friends are any better than just fair-weather friends. [Thats
depressing, being stuck with nothing but such lousy friendships.] But thats not
what this famous one-liner is sayingwe havent captured it correctly. What
this is driving at is the having no friends at all, viz., being friendless, is actually
better than just having fair-weather friends. Notice that the no was functioning
as a nonas part of the title of the subject category: F a BFWF.
A rose by any other name would smell as sweet.
Lets put Shakespeare into standard categorical form: All roses that are named
other than rose, e.g., Jesse, are roses that smell as sweet as roses that are
named rose. From a literary point of view, this sucks. But from a logical point
of view, this is actually better, and clarifies the original sentence quite well.
Pop quiz: You cant swing a dead cat without bumping into
a categorical statement!
In fact, the preceding sentence illustrates its own point.
So if you want more practice at identifying categorical
statements in the field, just do your daily reading routine and
youll be looking at one categorical statement right after another.
Seriously! I just grabbed the closest book to me right now and
Ive randomly opened it and picked a paragraph as Im keying
this inI kid you not! The book happens to be Word 98 for
Macs for Dummies. On page 269, under the heading Installing
the miniprograms, is the following paragraph, which consists of
nothing but categorical statementssometimes packed morethan-one to a sentence. See if you can translate the paragraph
into standard categorical form.
165
Microsoft has developed two versions of many of its

programsone to run on PCs and one to run on Macs.
When you run the Mac version of such a program, its
often all too apparent that Microsoft expended most of
its programming energies on the PC version, and then
simply reworked that version to run on the Mac. Thats
why most of us Mac devotees get a perverse satisfaction
out of criticizing Microsoft. With Word 98 (actually,
with the whole suite of Office 98 for Macintosh
programs), Microsoft promised to change things. And
Bill & Co. actually came through. Word 98 for
Macintosh is more Mac-like than any previous efforts by
Microsoft. This improvement is reflected in the
installation process: To get all the Word files on your
Mac, you simply drag a single folder from the
installation CD-ROM to your hard drive.
Introducingthe Venn diagram [dot, da da daaaaah]!

To represent all of our categorical statements, we can use whats called the Venn
diagram. If its not obvious already from my heading, I think Venns are great.
Well get gobs and gobs of work out of the Venn.
The first thing it will do for us is vividly capture the relations between the categories expressed by the four basic categorical statements.
Heres the Venn we will use as our bare canvas:
This little Venn represents the entire universestrange but true. It represents all possible relations of inclusion and exclusion between the two categories, A and B.
166
Heres how: There are four areas of this Venn:

AB: The football-shaped area where the circles
overlap [AB-land] is the area where all the
stuff in the universe that is both A and B
is.
B: The left-hand crescent, to the left of the
AB
football, is where all the stuff that is A but
not B is [A non-Bs live here].
AB: The right-hand crescent is where all the
stuff thats B but not A is [the non-A Bs].
AB : The area outside of both circles is for all
the stuff thats not A and not B [the nonA non-Bs].
Pop quiz: Where would the following things live on our Venn?
[Remember, my symbols will always make sense.]
red ants
brown ants
red cats
brown cats
Diagramming categorical statements

Now that you know the lay of the land for Venns, we can begin diagramming our
four basic categorical statements.
167
A a B All As are Bs.
We begin with our bare Venn on the left, and shade in the left crescent, to get
the diagram on the right. We do this because, in that crescent were any As that
were not Bs. But, A a B requires that there be no such creatures! Any As that
were in there, then, had to be made into Bs, so that all As are Bs. So we just
shoved them into B-land, and that did it.
Important tip: Think of shading as way of herding things into the right areas
so as to meet the descriptions expressed by the statement youre diagramming.
This is the only sort of counterintuitive aspect about using the Venn diagram
usually, with pie charts and bar graphs, etc., shading represents the existence
of things. But with Venns, shading represents the nonexistence of things!
A e B No As are Bs.
Here, we had to make sure that all As get evicted from B-land, so that theres
absolutely no chance for an A to be a B. Thats why we shaded the AB football,
thereby installing a complete divide between the As and the Bs.
168
A i B Some As are Bs.
Instead of using shading to represent the I statement, we use an asterisk, to

indicate that there exists at least one A that is also a B. Here especially its
helpful to unpack the I statement in this wayit then tells you exactly where to
put the thing, viz., in AB-land, so that there is something that is both A and B.
A o B Some As are not Bs.
Similarly, unpack the O statement, and it will tell you right where to put the
asterisk: There is at least one A and its not a Bits an A thats outside of Bland. The asterisk must stay out of B-land and yet must stay in A-land, and thats
why it plops down in the left crescent.
Pop quiz:
On the Venn below, diagram the following [dont peek at the
answers until youre done!]:
1. All Bs are As.
2. As are identical to Bs.
3. Some non-As are non-B.
169
4. Something is an A unless its a B. [This is a tough one.]

5. Only some As are Bs. [This ones tough too.]
6. All except As are Bs. [Way tough!]
1. B a A
2. A a B and B a A
3. A i B
4. B a A
170
5. A i B and A o B
6. A a B and A e B
Contradictory categorical statements

For each of our four basic categorical statements, there is a corresponding contradictory.
Contradictory statements are such that both cannot be true at the same time
and both cannot be false at the same time either.
The truth of one statement entails the falsity of the other. And the falsity of
one entails the truth of the other. So, from the truth of one statement, one can
immediately infer the falsity of the other, and vice versa.
To really drive our point home: The truth of one statement is logically
equivalent to the falsity of the othersaying that one is true is just like saying
that the other is false, and vise versa.
The contradictories are mapped out on The Square of Opposition: The contradictories are the diagonals of each other.
So, if you know that all cats are mammals, you immediately know that its false
that some cats are not mammalsthe truth of an A statement is equivalent to the
falsity of its corresponding O statement. If you know that some students passed
171
the course, you know that its false that none of them did. These are just two
different ways of saying the same thing. And if you know that not all the food
was good, you know that some of it was not.
Pop quiz: What are the contradictories of the following?

Some of the tennis was interesting.
Some players were not friendly though.
The matches were all televised.
None of the announcers were informative.
Not all the calls were accurate.
But most were.
External vs. internal negation

Put the following in standard categorical form:
All the mail is not sorted.
Its ambiguous. How you interpret this statement will depend on its context.
If its stated in response to the question, Youve been at that for hours now;
have you finished sorting the mail? then one probably means to say:
It is not the case that all the mail is sorted.
If, on the other hand, its stated in response to the question, Were the emergency crew you sent for; did you get a chance to even start sorting the mail?
then one probably means to say:
All the mail is non-sorted.
This is the difference between external and internal negation.
In the first case, one is merely saying that the statement, All the mail is sorted
is false. This is a case of external negation. And if this A statement is false,
then its contradictory O statement must be true, viz., Some of the mail is not
sorted.
In the second case, however, one is saying, All of the mail is unsorted. This is
a case of internal negation.
172
Certainly, the two statements, Some of the mail is not sorted, and All of the
mail is unsorted, are not equivalent. So, its very important to keep track of
this difference between external and internal negation.
Pop quiz:
After a century in the shadow of Newtonian ideas, according
to which light is composed of small particles, wave models of
light were revived about 1800, first by an Englishman,
Thomas Young, and then by a Frenchman, Augustin Fresnel.
Fresnels model was submitted for a prize offered by the
French Academy of Sciences. One of the judges for the
academy, S.D. Poisson, deduced that according to Fresnels
model, the shadow of a small circular disk produced by a
narrow beam of light should exhibit a bright spot right in the
center of the shadow. Poisson and the other judges are
reputed to have thought that this refuted Fresnels hypotheses
because they had never heard of there being such a
phenomenon and regarded it as highly unlikely to exist. No
known particle models predicted such a spot. But when
the experiment was carried out in carefully controlled
circumstances, there was the spot, just as required by
Fresnels model. Fresnel received the prize in 1818. [Ronald
Giere, Understanding Scientific Reasoning.]
The sentence above in bold can be constructed three ways, due
to the distinction between internal and external negation.
Find all three ways.
And which way must be true in order for the results of the
experiment to work best as evidence for Fresnels wave
theory of light and evidence against the particle theory?
BTW: Did you catch the misuse of deduced? And who really
deserved the prize here?
Different ways of saying the same categorical statements

Weve already seen that theres a bundle of ways of expressing categorical
statements in ordinary language. Well, theres also a bundle of ways of
173
expressing categorical statements merely within the confines of standard categorical form.
Traditionally, this has come under the title of immediate inference. But that
title is a rather misleading understatement. When two statements are two ways of
saying the same thing, then of course you can infer one from the other; after all,
they are simply two ways of saying the same thing!
So, rather than calling these immediate inferences, lets call them equivalents.
Equivalent converses
There are certain classic ways of manipulating categorical statements so as to
generate new ones. We will study three different ways of generating categorical
statements, the first of which is the converse.
The converse of a categorical statement is created simply by switching the
subject and the predicate. So the converse of All members have contributed
is All the contributors are members.
BTW: The converse, as we are discussing it, is not necessarily

the way others might mean it in ordinary discourse. When you
hear someone say, Blah blah blah; and, conversely, blah blah
blah, they might merely mean to contrast two things, as if to say
something like, Blah blah blah; on the other hand, blah blah
blah.
The question now is, which converses are equivalent to their originals?
We will use the Venn diagram to discover and prove the equivalence or nonequivalence of categorical statements.
Weve already used the Venn to represent categorical statements and the relations between their categories. So, if we are wondering if two statements are
equivalent [namely, just two ways of saying the same thing], then, if we diagram
both of them and end up with identical diagrams, thats our proof that the
statements are equivalent.
Identical Vennsidentical statements. Different Vennsdifferent statements. Its that simple.
174
Lets begin our hunt for equivalent converses, by looking at the A statement. We
first diagram A a B. Remember, we herd all the As into B-land by means of
shading.
AaB
And now, how would you diagram the converseB a A? Youre going to herd
all the Bs into A-land. Heres a fresh Venn for you to use:
BaA
Compare the two Vennsare they identical?
175
AaB
BaA
No, theyre not identical. So, the converse of an A statement is not equivalent to
its originalyou cant infer one on the basis of the other. Weve proven it!
Just think about it tooto say that all cats are mammals is for sure not to say that
all mammals are cats. And likewise, we cant infer that all mammals are cats
[which is flamingly false] from the statement that all cats are mammals [which is
obviously true].
Lets test the next oneis the converse of an E statement equivalent to its
original?
Begin by diagramming A e B. Rememberyou herd all the As out of B-land, in
order to get it so that not a single A can be a B:
AeB
And now, on this fresh Venn, how would you represent the converse, B e A?
176
Herding all the Bs out of A-land now, you get the following:
BeA
Compare the two Venns:

AeB
BeA
Identical diagrams; equivalent statements. [Different processesherding to the

left vs. herding to the rightis not what matters; having identical diagrams is
what matters for equivalent statements.] So the converse of the E statement is
equivalent to its original. Just think about it a minute: To say that no cats are
reptiles and to say that no reptiles are cats, are indeed to say the same thing;
theres just a difference in whos getting top billing in the discussion, but the
same claim is being made in either case.
177
How about the I statement; is its converse equivalent to its original? Lets get
A i B represented on the Venn and see. Remember, we now shift to the asterisk
to represent the claim that at least one A is a B:
AiB
Using the same Venn this time, how would you represent the converse, B i A
that there is at least one B and its also an A?
BiA
Comparing the two Venns, we see theyre identical; so the converse of an I

statement is equivalent to its original.
Just a quick example will cinch the point: To say that some cats are brown is the
same as saying that some brown things are cats. The second way of putting it
sounds a bit weirdwed rather talk about our cats than our brown thingsbut
its equivalent nonetheless.
How about the O statement?
Remember, if you express the O statement in the following way, it will tell you
exactly where to place the asterisk: There is at least one A, and it is not a Bits
an A living outside of B-land.
178
AoB
And now, how would you represent B o A, [There is at least one B, and its
outside of A-land.]?
BoA
Comparing the two Venns, we see that they are Close, but no cigar!
AoB
BoA
Thinking through an example will make it obvious that conversion does not hold
for O statements: To say that some animals are not mammals is one thing; to say
that some mammals are not animals is quite another!
179
Our list of equivalents

Start keeping a list of equivalents to use as a kind of thesaurus, so that youll be
able to identify categorical statements no matter how theyre expressed.
So far, we have the converses of the E and the I categorical statements on our
list.
Equivalent obverses
Another way of generating categorical statements is by means of obversion.
Making the obverse of a categorical statement is a two-step process:
Obverse:
1. Change the quality of the statement. That is to
say, change it from an affirmative to a negative or
from a negative to an affirmative. So, if its an A
statement, change it to an E; if its an O
statement, change it to an I. Think of it as
changing the charge of the statement.
2. Negate the predicate.
So, the obverse of All As are Bs, is No As are non-Bs. Read through the list
and youll quickly get a feel for it:
AaB
AeB
AiB
AoB
AeB
AaB
AoB
AiB
Some obverses get a bit strange sounding, e.g., the obverse of the I statement
Some aardvarks are not other than butt ugly. With all the negatives, it gets
a little hard to follow. That will turn out to be one of the benefits of knowing the
equivalent ways of expressing categorical statementsso you can translate those
statements into more positive, understandable terms.
Lets see, then, which obverses are equivalent to their originals. Lets start with
A a B and its obverse, A e B. Heres the Venn for A a B. On the fresh Venn on
the right, diagram A e Bthat there are no As that are outside of B-land. [All
B -land, must get herded out of there and into B-land, so
those As that are in AB
that none of those As are in B-land anymore.]
180
AaB
Your verdict? You should have gotten the same Venn for both. Herding all the
As over to make Bs out of them [A a B] and herding them all over so that theyre
no longer non-Bs [A e B] results in identical Venns.
These may be different motives for shading, but thats not what matters here.
What matters is results. These are two different ways of expressing the same
statement.
So the obverse of the A statement is equivalent to its original. And, sure
enoughNo cats fail to be mammals, is just a less straightforward way of
saying All cats are mammals.
How about the obverse of the E statement; are A e B and A a B equivalent?
[Dont peek; see if you can find the answer before you scroll down.]
AeB
You should get the same diagram againA e B = A a B. We can say that no cats
are reptiles, or we could say that all cats are other than reptilesits the same
thing.
How about the obverse of the I statementare A i B and A o B equivalent?
181
AiB
Same diagram againto say that some cats are brown and to say that some cats
are not a color other than brown is to say the same thing. The latter way is just a
bit clumsier.
And lastly, what do you think about the equivalence of the O statement and its
obverse?
AoB
Yep; the same againA o B = A i B. To claim that some cats are not brown is
to claim that some cats are colored other than brown.
So add all the obverses to your list of equivalents!
Equivalent contrapositives
The contrapositive of a categorical statement is generated by a two-step process
thats sort of a hybrid of the converse and the obverse.
Contrapositive:
1. Switch the subject and the predicate.
2. Negate the subject and the predicate.
182
The mantra of contraposition is: Flip & NegateFlip & NegateFlip &
Negate.
Heres the list of contrapositives, then. Say them to yourself and get used to the
funny-sounding ones:
AaB
AeB
AiB
AoB
BaA
BeA
BiA
BoA
How about that last one! Some non-brown things are not other than alligators?!
Henry Kissinger talks like that! Mumble in a deep, gravelly voice: It is not
ruled out that I would fail to be absent at such an occasion. Hank, are you
coming or not?! And thats why we need to know our equivalents: To translate
the Henry Kissingers into ordinary language.
Lets begin with the A statement and its contrapositiveare A a B and B a A
equivalent? Shade the Venn on the right so that all the non-Bs are herded into
non-A-land.
AaB
You should be getting the same diagram as A a B. The process will be a bit
different, but the result will be the same: With A a B, you herded all the As into
B-land; with B a A, you herded all the Bs out into the big surrounding area of
AB.
So, the contrapositive of the A statement is equivalent to its originalyou can
say, All cats are mammals, or you can say, All animals other than mammals
are not cats. [Question: Before knowing what you do nowviz., that the
contrapositive of an A statement is equivalentwould you have correctly
identified these statements as saying the same thing?]
183
Lets see if the contrapositive of the E statement is equivalent: A e B and B e A.

Shade the Venn on the right so that none of the non-Bs are in the non-A area.
AeB
WowVenns dont get much different than that!

BeA
So, saying that no cats are reptiles and saying that no animals that arent reptiles
are not cats is to say two different things. We know this now for sure.
Before, when all we had to rely on was just our general intuitions on how these
statements sounded, I doubt if any of us could tell whether these statements were
equivalent or not.
The Venns are teaching us things already!
Next, lets test the contrapositive of the I statementA i B and B i A.
Place an asterisk on the Venn on the right to indicate that there is at least one
thing that is both a non-B and a non-A.
184
AiB
You should get a Venn that looks like this:

BiA
So the contrapositive of the I statement is not equivalent. [Although, once again,

without the Venn to prove it, wed be hard-pressed to say.]
Lets try our last one: The contrapositive of the O StatementA o B and B o A.
Place the asterisk on the Venn on the right to indicate that there exists at least one
thing that is not a B and not a non-A. This thing must be a non-B and yet avoid
being a non-A. Wheres it go?
AoB
The asterisk will go in the same place as it did with A o Bits definitely a nonB, but to keep from being a non-A, it must remain in A-land.
185
So the contrapositive of an O statement is equivalent to its original. To say that

some cats are not brown is really the same thing as saying that some things that
arent brown are not things other than catsbelieve it or not.
We have now proven, by means of our Venn diagrams, that the contrapositives
of the A and the O statements are equivalent to their originals.
Rounding up all the equivalents

Lets complete our thesaurus of equivalents weve proven.
Remember that we began our investigation with a discussion of what contradictory statements were and how to use the Square of Opposition to help you
keep track of which statements were contradictories of each other. With contradictories, the truth of one was equivalent to the falsity of the other, and the falsity
of one was equivalent to the truth of the other.
We can now compile our knowledge of equivalent statements into the following:
Logical equivalents
Converse
Obverse
Contrapositive
Contradictories
E and I
A,E,I, and O
A and O
A and O
E and I
The truth of one =
the falsity of the other.
Be familiar with these equivalents.

Be so familiar with them that moving from one way of expressing a
statement to all the other ways comes so easily that you dont really have to
think about it.
Thats pretty hardcore, but thats what it will take to be able to deal with all the
things people say to you in categorical form and all the arguments people throw
at you using categorical statements.
186
Practice makes it easylets work on some together

Lets take a categorical statement and put it through its paces. For example, the
other day I got the following in a fortune cookie: Happiness is in the mind.
Lets play with itfind all its equivalents.
Original:
HaM
Converse
none
Obverse
HeM
Contrapositive
MaH
Contradictory
Its false that H o M
Lets play around with Those who live by the sword, die by the sword.
Original:
LaD
Converse
none
Obverse
LeD
Contrapositive
DaL
Contradictory
Its false that L o D
Another way to practice finding equivalents is to let our list take us to wherever it
leads us. This is more fun. Lets do it with L a D:
We cant take the converse of L a D, but we can always take the obversewhich
gives us L e D. And now we can take the converse of thatwhich would give
us D e L. The converse of that would give us nothing new, but the obverse does,
namely D a L . And now, how does D a L compare to our original, L a D?
Thats right; its the contrapositive. See? [FYIthe contrapositive [D a L] of
L a D is just the obverse of the converse of the obverse of L a D, all of which are
equivalentso, no wonder its equivalent!]
Practice, practice, practice: All you need is a little scratch papera gum
wrapper or a napkin will do. Lets do: Some profs are a pain in the butt.
P i PB = PB i P = PB o P = P o PB
[original => converse => obverse => contrapositive]
187
Lets play with this little gem from Socrates: The unexamined life is not worth
living.
A few new issues crop up with this example. The first issue is the question of
how to interpret the statement.
My advice is to always give the statement the most literal representation
possible, to capture all of its aspects and information.
Is Socrates making a universal claim or just a particular claim? Is he launching a
claim about all unexamined lives or just some of them? He seems to be making
a universal claim [or else his claim sort of fizzles out as a lofty pearl of wisdom,
and becomes a mere platitude]. So Socrates is making an A statement about
unexamined lives, claiming that all of them are not worthy of being lived. And
so now we are ready to put it into our notationE a W.
You may be asking why I didnt symbolize the category of unexamined lives
with U. The reason I didnt, is that then the negation [the un] would be harder
to keep track of. This is especially a problem with statements such as, The
unexamined life is uninteresting. Both categories are negative categories, and
so referring to them by means of their negative aspects wouldnt distinguish
them. A related problem crops up with, The unexamined life is one of
urgency.
My recommendation is to use the core positive aspect of a category to determine
your choice of notation. Thats why the unexamined life is best thought of as the
non-Examined life [and the uninteresting as the non-Interesting].
So lets put Socrates little observation through its paces:
We begin with E a W. We know the converse is not equivalent; so, whats the
obverse [which we can always do]? Taking the obverse of E a W brings us to
our next little issue.
When we try to negate the predicate of E a W , we get a double negationwe
would be putting a bar on top of W. When this happens, the double negations
just cancel each other outnon-non-W and W are the same. So the obverse of
E a W is E e W.
And now, with these little issues cleared up, lets follow the equivalents of E a W
to where they lead us:
EaW = EeW=WeE = WaE
[original => obverse => converse => obverse = contrapositive of the original]
188
Notice how we were able to translate Socrates famous one-liner into positive
categoriesall lives worth living are the examined ones.

OK, youre ready to begin working with the CT software again. This time,
launch the module called Standard Categorical Form. This will hone and verify your knowledge of everything weve covered so far on categorical logic.
Your project in Standard Categorical Form is to select the answer that best
represents the statement in the passage box.
Keep the following in mind, as you work on the exercise:
The proper use of symbolsthey should always

make sense.
The many ways of expressing categorical

statements in ordinary language.
The equivalent ways of expressing categorical

statements.
Strategy: I recommend that you give the sentences in the

passage box the most literal representation first, by means of our
notationso that you keep track of all the nons and all the
other subtleties of the statements. And then, if necessary, apply
the appropriate rules of immediate inference for conversion,
obversion, contraposition, and contradictories, thereby working
your way to the answer[s].
Use scratch paper. There is no way you can reliably work
through all the equivalence relations in your headI know I
cant! So use scratch paper.
Contraries vs. contradictories

Just a reminder: Contradictory statements are such that both cant be true and
both cant be falsethe truth of one implies the falsity of the other and vice
versa. There is also a near cousin of this relation; its the relation of being
contraries.
189
With contraries, both statements cant be true at the same time, but both
statements can be false at the same time.
Take the statements, Kevin is bald, and Kevin is hairy. These are contraries
the predicates of baldness and hairiness are contrary predicates. Theres no
way I can be both bald and hairy at the same time. But both statements can be
false; in fact, they are false [at least for a few years yet]. Im sort of fuzzy on top.
The contradictory of bald is simply non-bald. If you want to make sure
youre stating the contradictory [the simple denial] of something, there is one
fail-safe way to do itjust slap a non or a not in front of it. So, the
contradictory of Kevin is bald, is It is not the case that Kevin is bald, or
Kevin is not bald, or Kevin is non-bald.
Confusions and misrepresentations can easily occur when one loses track of the
difference between contraries and contradictories. Heres an instance:
Take the statement, Some people are not happy with the candidates. We would
put that into standard categorical form as: P o HC. And now that we know that
the obverse is always equivalent to its original, we know P o HC is equivalent to
P i HC. But now, if we translate that back into ordinary language, we might be
tempted to express this as, Some people are unhappy with the candidates. But
that would be a mistake. Happy and unhappy are contrary predicatesone
could be neither happy nor unhappy with the candidatesone could simply be
totally blah about them. So, the proper rendering of P i H C would be, Some
people are non-happy with candidatesa klutzy phrase, but an accurate
representation of our original.
To treat contraries as if they were contradictories is the commit the fallacy of
false dichotomy. This happens so frequently that many cases have obtained the
status of slogans: e.g., If youre not part of the solution, youre part of the
problem., or Either youre for us or against us. In both of these cases, there is
a neutral positiona middle ground that is mistakenly ignored.
When trying to figure out if two statements or predicates are either contraries or
contradictories, you need to ask the following questions:
1) Can both be true simultaneously?
2) Can both be false simultaneously?
Lets go through a few examples, to get a good conceptual feel for the difference.
190
heavy
rich
near
fast
empty
loud
Contrary
Contradictory
light
poor
far
slow
full
quiet
not heavy
not rich
not near
not fast
not empty
not loud
Pop quiz: Are these contraries or contradictories?

polite
possible
probable
intelligent
appropriate
approved
valid
impolite
impossible
improbable
unintelligent
inappropriate
unapproved
invalid
Another pop quiz:

Are the predicates of being contraries and being
contradictories themselves either contraries or
contradictories?
While were on the topic of contraries and contradictories, what are the relations
among the following, where A is an action?
The duty to do A
No duty to do A
The duty to not do A
Answer: Having the duty to do A and having no duty to do it are contradictories.
Having the duty to do A and having the duty to not do it are contraries. One
cant have a duty to do something and, in the same sense, have a duty to refrain
from doing it too. But one can have neither a duty to do something nor a duty to
refrain from doing itone is simply at liberty to do it, if one chooses. [To see
this, have the action be, e.g., eating a cheese sandwich for lunch.]
What about the relations among the following, in which p stands for any statement, e.g., Winona is in Minnesota?
191
To believe that p
To not believe that p
To believe that not p
Answer: Believing that something is the case and not believing that it is the case
are contradictory states. Believing that something is the case and believing that
it is not the case are contraries. One cant believe that Winona is in Minnesota
and believe that Winona is not in Minnesota, but one who has never heard of
Winona could easily fail to have either belief.
And now you try one:
To see that p
To not see that p
To see that not p
See a pattern? Its the distinction we discussed earlier, between internal and
external negation. The failure to properly distinguish between internal and
external negation is the source of many instances of the fallacy of false
dichotomy. For example, when one hastily thinks that someone is an atheist, on
the basis of theyre admitting that they dont believe in a God. All one can
conclude, on the basis of their admission, is that they are agnostic. If someone
doesnt believe in global warming, that doesnt imply that they believe there is no
global warming. Between the belief that p and the belief that not p is the huge
option of withholding beliefexactly what one ought to do when one lacks
sufficient evidence for believing one way or the other.
Subcontraries
A cousin of the contrary is the subcontrary. Two statements are subcontraries
when they can both be true, but they cannot both be false.
I and O categorical statements are subcontraries. Some men have big feet,
and Some men dont have big feet, can both be true [in fact, they are], but they
cannot both be false.
Pop quiz:
Why cant Some men have big feet, and Some men
dont have big feet, both be false?
192
Another pop quiz:

Whats the relation between A and E categorical
statements?
Syllogisms
Weve studied categorical statements and the equivalent ways of expressing them
so as to make immediate inferences from one to another. So weve already
been studying various categorical arguments, albeit really short onesones with
only one premise and one conclusion. Its time to study some longer arguments,
ones with multiple premises.
The type of argument thats most frequently found in categorical form is the
syllogism. It has quite a few specifications:
Its a deductive argument, with two premises, one conclusion, all in the form of
categorical statements, concerning exactly three categories. Lets list all these
features, so we can keep track of them better:
Syllogism:
Deductive argument
2 premises
1 conclusion
Categorical statements
3 categories
Syllogisms have so many specs that youd think theyd be quite rare, but theyre
not. Heres a typical example:
People who know the proper rules of punctuation and grammar
do well on the composition part of the Law School Aptitude
Test. Which is why I think Mary doesnt know her way around
grammar and punctuation very wellshe bombed the comp part
of the LSAT.
The basic format of the syllogism looks like this, once symbolized in our
notation:
AvB
BvC
AvC
193
Its easy to distill the argument above into this format. The first premise, in
standard categorical form, is the A statement that all people who know the proper
rules of grammar and punctuation do well on the LSAT. The second premise is
the last clause of the passage: The E statement that no one who is Mary did well
on the LSAT [in fact, she did poorly, but we dont need that information to infer
the conclusion]. And the conclusion is the E statement that Mary doesnt know
proper grammar and punctuation. Choosing our symbols wisely, we would put
this into our notation as:
KaW
MeW
MeK
Note how efficient our notation is: We can distill our argument down to
thisnine measly letters. Thats amazing! The power and efficiency of our
notation will enable you to record an argument as fast as one of those court
stenographers. As fast as a presenter can be throwing an argument at you, you
can be jotting it down in the margins of your program, or some other scrap of
paper [or on your palm or Palm!], so you can quickly assess it for validity. Pretty
nifty tool, I must say.
Validity and the Venn

We now know what a syllogism is and how to quickly symbolize one using our
notation. Its time to begin our study of how to determine whether a syllogism is
valid or not.
Once again, we will be using the Venn diagram as our tool. Earlier, we used a
two-circled Venn to test the equivalence of statements, i.e., to see if we could
validly infer one statement from the other. Since syllogisms have three statements, involving three categories, we will have to beef up the Venn by adding
another circle:
194
The three-circled Venn will serve as our canvas. It represents the entire universe
and everything in it, with respect to being included in or excluded from categories A, B, and C.
For example, where are those things which are A and B and C? They are right in
the center of the Venn, where all three circles overlapABC.
Where are those things which are A and B, but not C? They will be in the
overlap of the A and B circles, but in the upper half of that footballABC.
Where are the things that are C but not A and not B either? Well, they are found
in the C circle, but in an area of that circle that doesnt overlap with the A circle
and doesnt overlap with the B circle. So theyre found in the odd looking area
labeled ABCthe lower-most half-circle that looks as if someone sat on it.
Representing categorical statements on the Venn

Weve already learned how to represent categorical statements on two-circled
Venns. Well, diagramming them on the three-circled Venn is not much different.
Lets start with our basic four:
AaB
To indicate that all As are Bs, we use shading, as usual. We have to herd all the
As into B-land, and we find that As can come in two varietiesthose that are
also Cs and those that are non-Cs. No matter; we have to make all those As into
Bs whether they are Cs or not. [Its sort of as if the C circle isnt there.]
[You probably noticed that I started using a trimmed-down version of the Venn
as our bare canvasone without the ABC area. We arent forgetting about those
things that are ABC; were just making it a bit easier on ourselves, working with
a cleaner-looking Venn.]
195
AeB
To represent the claim that no As are Bs, you have to shade all the As out of Bland, whether those As are Cs or not. [Again, its as if the C circle isnt really
there.]
AiB
To diagram an I statement, we use an asterisk. Here, the third circle does make a
difference thats important never to forget.
We must represent the claim that there is at least one A, and that A is also a B.
We know already that the asterisk will be going in the AB overlap, but exactly
where? If we put it in the top half of the AB football, then we are specifying that
the A that is a B is definitely not a C. Was that stated in the original claim,
A i B? No! If we put the asterisk in the bottom half of the AB football, then we
are specifying that the AB is also a C. And was that stated in the original claim?
No! We are told only that the thing is an AB; we are told nothing about the
things C-ness. To indicate that it is yet undetermined whether this AB is a C or
not, you put the asterisk on the fence.
196
Your job is to capture all the information in the statement, when you diagram it,
but not read anything into it. Which is what youd be doing if you put the
asterisk above or below the C fence.
AoB
To indicate that there is at least one A that is not a B, just think about all of the
specifications in the statement and it will tell you exactly where to drop the
asterisk: The thing is an A, so itll be in the A circle, but it is also outside the B
circle. Its not stated whether the thing is a C or not, so it must be placed on the
C fence. And there you are!
Representing arguments
Were ready to begin working with syllogisms! Lets begin with perhaps the
most famous little syllogism of all:
Socrates is a man.
All men are mortal.
Socrates is mortal.
The first thing to do is to symbolize it, using our notation:
SaM
M a Mo
S a Mo
This establishes the labels for the Venn [note that weve cleaned up the Venn a
bit more, by moving the labels outside the circlesbut each of the eight areas
still retains its unique specifications]:
197
And now were ready to diagram the premises on the Vennlets do the first
premise firstS a M:
After herding Socrates into the circle of men, we next diagram the second
premise, that all men are mortalM a Mo. Herd all men to their death in Moland.
We have both premises diagrammed on the Venn now. How can we tell if the
argument is valid?
Well, keep in mind what the groundedness condition of validity is all about.
Validity is the relation between premises and conclusion such that given the
premises are true, the conclusion must be true. Given the premises hold, the
conclusion must hold too. So, given that we have accurately diagrammed the
198
premises on the Venn, we ought to be able to read the conclusion right off the
Venn, if the argument is valid.
Can we read the conclusion off the Venn; is the area representing Socrates forced
into the area of mortal things? Indeed it is! Where is Socrates? The area of
things that are S suffered a huge loss of real estatethe premises herded all of
the S area into just the innermost triangle, totally within Mo-land.
And thats how the Venn is used to test the validity of arguments: If the argument is valid, the premises force the conclusion to be true; you will be able to
read the conclusion right off the diagram. If you cant read the conclusion
off the diagram, its not valid. Its that simple.
The dynamics of the Venn

Look again at the argument about Socratesthe Venn does a great job of illustrating how premises can deliver a one-two punch, to force their conclusion
to be true: The first premise [S a M] shoves Socrates into M-land, and then, by
virtue of the second premise [M a Mo], hes swatted down into the into Moland and doomed to death. What can I sayVenns are action figures!
Lets work through a couple more examples

Ive picked some examples to bring out some of the special issues that crop up
when using the Venn to test syllogisms for validity.
Lets test the following argument:
Stocks in Corp Inc. are over-the-counter. And, as everyone
knows, OTC stocks are different from Treasury Bills. So, Corp
stocks are different from T Bills.
199
CaO
OeT
CeT
Diagram the first premiseC a O [try not to peek]:
Now, diagram the second premiseO e T:
200
Now, can you read the conclusionC e Tright off the diagram? The diagram
must make it the case that not a single C can be in T-landdoes it?
Yes! All thats left of C-world is the little iron-shaped area up top. C-world is
totally cut off from T-landtheres just no way a C could possibly be a T.
So the argument is valid.
Lets do another one:

Some rich people are Catholics; and a number of Catholics are
Democrats; so there are at least some rich Democrats.
RiC
CiD
RiD
Diagram the first premiseR i C:
[Make sure youre able to explain exactly why the asterisk was placed just so.]
201
Now diagram the second premiseC i D:
And now, can you read the conclusionR i Doff the diagram? Is there an
asterisk that fits the exact specifications of the conclusion? The conclusion
requires that there be something that is definitely an R and definitely a D. Is
there?
Well, the first asterisk we drew is definitely an R, but its not definitely a D. The
second asterisk we drew is definitely a D, but its not definitely an R. So we
cant read the conclusion off the diagram; its not a valid argument.
But wait a second!

We know damn well that some rich people are
Democratsa lot of Hollywood bigwigs are, for
example. Very true, but quite irrelevant here. The issue
is not the actual truth of the conclusion. The issue here
is validitywhether the truth of the premises can
guarantee the truth of the conclusion. So, when testing
for validity, dont be distracted by the accidental truth of
the conclusion or the accidental falsity of the premises.
Learning from ones mistakes

You just learned how to test the validity of syllogisms using the Venn diagram.
Another thing the Venn can do for you is to tell you how to fix invalid
syllogisms once theyre detected. Ill show you by using the invalid syllogism
we were just discussing.
RiC
CiD
RiD
202
The conclusion [R i D] claims that there is definitely an R that is definitely a D.

The first premise established the existence of an R, which the conclusion
requires. The problem is, the R doesnt have to be a Dbecause the R is a C,
and only some Cs are Ds, as the second premise states. So what would it take
to make the R into a D?
Well, if all Cs are Ds, instead of merely some Cs being Ds, that would force the
R to be a D. So changing the second premise from C i D to C a D would do the
trickthe second premise would knock the first premises asterisk down off
the D fence and into the middle triangle, where it would indeed be an R that is
also a D [R i D]. Follow the dynamics of the new argument on the Venns below:
RiC
RiC
CaD
Beware: Even though the Venn showed us how to correct our syllogism so as to
make it into a valid one, it may not be wise to make that correction. Think again
about the new second premise which is necessary to make the argument valid
C a D. What statement would that represent? All Catholics are Democrats.
203
But that premise is just plain false! Here, the price of having a valid argument
would be to flunk miserably the acceptability condition. There is just no hope for
this argument with respect to cogency.
But often, the change in premises [or perhaps conclusion] suggested by the Venn
is quite acceptable. This, then, makes the Venn more than just a validity
gaugeit makes the Venn a wonderful diagnostic tool.
Existential commitment revisited

Lets test the validity of another argument, this time one in which the issue of
what we earlier called existential commitment generates a bit of a problem.
No lawyers are illiterate, but they are all educated; so
some educated people are not illiterate. [Adapted from
Goviers A Practical Study of Argument]
LeI
LaE
EoI
How do you diagram the first premiseL e I?
204
How do you add the second premiseL a E?
Is the argument validi.e., can you read the conclusion off the diagramE o I?
Answer: Not exactly. The conclusion states that there exists at least one E that
is not an I. An O statement such as this requires an asterisk to represent it, and
the premises have placed no asterisk on the Venn.
We mentioned early in our discussions of the A and the E categorical statements
that they involve no commitment to the existence of even a single member in
their subject categories. [This was important for our ability to make universal
claims about hypothetical things.] And now we have a syllogism with two universal premises and a particular conclusionan I or an O statement.
In this situation, however, to call the argument invalid because its premises did
not place the necessary asterisk on the Venn, is merely to catch the argument on a
technicality instead of really putting the argument to the test.
When you are testing the validity of a syllogism like thisone with universal
premises and a particular conclusionyou should further investigate whether
it is valid given the obvious existence of at least one member in one of its
categories.
Let me show you what I mean.
LeI
LaE
EoI
The subject of the first premise is lawyerswould you grant me the existence of
at least one lawyer? Sure! After the O.J. Simpson trial, youd probably grant me
the existence of too damn many lawyers! [Youd probably grant me the existence of at least one illiterate person and at least one educated person too, but
lets work with our obviously existing lawyer.]
205
Knowing nothing else about that lawyer, where would they live on our Venn?
[This is a tricky one.]
Heres where they would bedefinitely an L, but undetermined as to whether

they are illiterate or not and educated or not. Theyre riding two fences at once:
Now, diagram the first premiseL e I:
The lawyer sort of shinnied down the E fence, as they got knocked off the I
fence.
Next, diagram the second premiseL a E:
206
And now, can you read the conclusion off the diagram, after the shading knocks
the lawyer off the E fence too? You bet!
So, the proper, robust answer in this case is that the argument is valid, given
the obvious existence of a lawyer.
When you have a syllogism with universal premises and a particular conclusion, dont just sneak by on a technicality and call it invalid. You must
test it further to see if it is valid given the known existence of at least one
member in any one of its categories. This is called existential import. This
investigation can get a bit involved, but thats what it will take to give the
argument its day in court.
Pop quizsome easy ones

Test the following syllogisms for validity using the Venn
diagram. Use scratch paper!
Remember:
First, identify the premises and conclusiondont
forget what you learned about the anatomy of an
argument!
Second, put the syllogism into standard categorical
form, locating its three categories and wisely selecting
your symbols for them.
Third, draw your Venn, add your labels, and diagram
the premises.
Lastly, see if you can read the conclusion off your
diagram, to decide whether the argument is valid or not.
Remember: After youre done diagramming the
207
premises, its pens down. The Venn has to speak for

itself, just like in poker, where the hand speaks for itself.
The argument is claiming that its premises can do the job
of guaranteeing the truth of the conclusion, all on their
own. Except in cases of existential import, then, there is
no helping the premises.
The answers will be provided below, but dont peek!
1. In view of the fact that many important church
officials held fiefs from kings, there were bishops
who held such fiefs, since bishops were important
church officials.
2. Not all rhombuses are similar. Hence, given that
figures A and B are not similar, they must be
rhombuses.
3. Foods that can be stored or kept in stock for long
durations without spoiling must contain additives.
However, foods containing additive are harmful to
ones health. We must conclude, then, that foods
with a shelf life of many weeks are harmful to ones
health.
4. All that glitters is not gold. Because some jewelry
doesnt glitter, one can infer that some of it must not
be gold.
5. No true believer is condemned, because the heretic is
condemned, and no true believer is a heretic.
6. None of woman born shall harm Macbeth. And,
Macduff was from his mothers womb, untimely
ripped. In light of this, Macduff was to be the
demise of Macbeth.
7. Some of the poets in our text are romanticists. Since
the romanticists idealized life, then, some of our
texts poets idealized life.
8. Plants in which photosynthesis occurs need water.
There are plants, however, in which photosynthesis
does not occur. Therefore, there are plants which
dont need any water.
208
Answers:
1. Invalid
2. Invalid
3. Valid
[Tip: When choosing titles for categories, first
take a birds-eye view of the whole argument, to
see what the general topics are. Thats why I
picked shelf life rather that stored or stocked
for many weeks. Also, foods would not a
good choice, since all three of the categories are
foodsfoods with a shelf life, foods with
additives, and foods that are harmful to health.]
209
4. Invalid
5. Invalid
6. Invalid
7. Valid
210
8. Invalid
Worst case scenarios

To really put your knowledge to the test, lets look at some examples where
youll have to apply all the skills youve learned regarding categorical logic.
Many people from Arkansas are not without property. And that
is why its reasonable to believe that not all citizens of Arkansas
are entitled to charity, in light of the fact that property owners
should not be getting charity.
The first order of business is to identify the premises and the conclusion [never
forget your anatomy of an argument!]. This argument goes, premise-conclusionpremise.
Then, put the argument into standard categorical form. Note that the conclusion
is stated as the denial of an A statement. With that in mind, we can put the
argument into the following notation, as we symbolize it most literally so as to
make sure we are accurately representing it:
AoP
PaC
AoC
At this point, the argument is not strictly a syllogism. Why? Because it has too
many categoriesit has five. But these are easily distilled down to three, by
obverting both premises.
AoP = AiP
PaC = PeC
AoC AoC
And now were ready to use the Venn diagram as usual to test for validity. [Ill
leave that to you.]
211
Be able to use your thesaurus of equivalents like this, to reduce the categories
down to the requisite three, so that the syllogism is ready for testing on the Venn.
Positive and negative categories

Negative categories are usually much harder to keep in mind. We are the only
species around that can pull it off half-way decently. For example, dogs can
quickly learn to fetch the ball, but have a much harder [impossible?] time
getting their little doggie brains around the concept non-ball and learning to
fetch any object other than the ball.
So, to make your syllogism easier to follow and easier to work with, distill it
down to positive categories whenever it is possible and beneficial, to help you
remember and assess the argument. In short: De-non whenever possible.
Sometimes the negative category is actually the one that people are more familiar
with. We studied just such a case a moment agothe example of the lawyers
and illiteracy. We often discuss the issue of illiteracywe dont debate the
problem of literacy. We argue about what to do about the homeless, not what to
do about those with homes. When this happens, its best to use the negative
category. The primary consideration is to facilitate peoples understanding of the
argument. De-non accordingly.
The best tool for de-non-ing is obversion. Rememberit applies to all four basic
categorical statements, and it removes the non of the predicate [by creating a
double-negation].
How do you remove a non on the subject? Well, sometimes you can use conversionviz., on the E and the I statements. This will put the non in the predicate
position, where you can use obversion to remove it. Sometimes you can use
contrapositionviz., on the A and the O statements.
But: Sometimes one is simply stuck with a negative category.
For example: How would you try to de-non K a L? If you take the contrapositive, that would give you L a K. The non is like a wrinkle in the carpet that
pops up elsewhere no matter how you step it down. You cant take the converse
of it, so as to be able to apply your de-noner. And obverting K a L as it stands,
just makes matters worse! No, youre just stuck working with a negative category. Just try to make it the easiest one to think about.
212
Pop quiz: Distill the following arguments down to three

categories, to make them ready to diagram on the Venn. [Ill
leave it to you to test them for validity.]
CaS
CeH
SaH
ZeP
HaP
HaZ
FYI: There are so many specs to a syllogism that there end up

being only 256 types of them. And out of these, there are only
15 valid ones. I can tell what youre thinking already. Dont
bank on it! If syllogisms were assembled randomly, then the
odds would grossly favor invalidity; but, so many of the invalid
syllogisms are so wacky that no one would ever use them
seriously. So dont play the odds and just call them invalid;
do the Venn.
FY further I: Long ago, all the syllogism types were given

names. Strange, but true. These names were supposed to help
people remember the format of the argument. For example, the
famous syllogism about Socrates, which we used to kick off our
study, is named Barbara. Do you see why? Names were picked
with exactly three vowels, the order of which represented the
order of categorical statements in the argumentin this case,
AAA.

By assembling and practicing all your skills at categorical logic, you should be
able to represent and test syllogisms almost as fast as someone can rattle them
off. Thats the project in the Venn Diagrams module of the CT software.
If things dont go well on this module, try to identify a pattern in your errors.
You may need to brush up on Standard Categorical Form, if youve forgotten
the logical equivalents weve studied.
Remember to use scratch papertheres no way you can do these in your head
reliably.
213
Enthymemes
An enthymeme is a syllogism with a portion left unstated.
Usually whats left implicit is a premise:
In order to graduate, one needs to meet the residency
requirement; so it looks as though Bobs not going to be
graduating this semester.
But sometimes, the conclusion is left implicit:
In order to graduate, one needs to meet the residency
requirement. And Bobs just transferred in. So there you are.
And sometimes, the context of the discussion may even allow one to simply
make a statement and have it function as a conclusion of two implicit premises
already established earlier in the discussion. For example, in the context of
reading the residency requirement and discussing how Bob just transferred, one
could simply turn to poor Bob and say, Looks like youre outta luck, guy.
Even just a sympathetic look towards Bob would qualify as your conclusion in
this context.
As youve probably figured out by now, most of our syllogisms are
enthymemes. But then this should come as no surprise. When we were discussing the anatomy of an argument, much earlier in this etext, we saw how arguments often have implicit premises or conclusions. Syllogisms would be no
different.
We should actually be thankful that people use enthymemes so regularly. If they
didnt, their arguments would be very repetitive and much longer. We would
soon be bored and insulted by having everything explicitly stated for us.
The other virtue of using enthymemes is that it gets the audience more involved
in the argument. As they are following the argument, they must think of and fill
in the missing part[s], thereby understanding and remembering the argument
betterone hopes.
These are the beneficial uses of enthymemes. But, of course, there are abuses of
them also.
If someone has an unacceptable premise, they may want to keep it from view so
that their audience doesnt notice it. Or, if their conclusion doesnt actually
214
follow from their premises, they may try to make it appear as if it does by simply
saying, And the implication is obvious.
And thats why its a good idea to keep a critical eye on enthymemes, and test
them for validity in exactly the same way you would a full-blown syllogism.
In fact, when you test an enthymeme for validity, you must make it into a
full-blown syllogism, by making all the implicit parts explicit. This will not
only permit you to properly test it for validity, it will also help you verify that the
syllogism meets the other ARG conditions.
Lets take a look at some typical enthymemes.
1. The presidency was stressful for Clinton. Look at
how much grayer he was at the end of his two terms.
2. Of course abortion is immoral; it results in the killing
of a fetus.
3. Dr. Kevs CT software is effective. Even Karl
Kartoffelkopf passed the Critical Thinking course
using it.
Lets test the first one for validity. It leads off with the conclusion and then gives
us a reason to believe it. So a premise is left implicit: All those whose hair turns
that much grayer have experienced stressful times. Putting brackets around the
implicit parts of our enthymeme, we can then represent the entire syllogism as
follows:
CaG
[G a S]
CaS
And now its ready for diagramming on a Venn, to see if its valid. [Ill leave
that to you.]
Lets do the second one. It too leads off with the conclusion and then supplies a
premise. The implicit premise seems to be that all cases of killing the fetus are
immoral. [How well does this implicit premise do on the acceptability
condition?!]
AaK
[K a I]
AaI
215
The third example leads with the conclusion too and provides a single premise.
The implicit premise states that whatever enables Karl to pass is effective.
CTS a KP
[KP a E]
CTS a E
A special note about this one: There is a strong temptation to symbolize Even
Karl Kartoffelkopf passed the Critical Thinking course using it, as All those
who are Karl Kartoffelkopf are among those who passed the Critical Thinking
course. But look where that interpretation would leave us:
KK a PCT
________
CTS a E
We would then have four categoriesone over the limit!and we havent even
found the implicit premise yet! When this happens, its pretty good evidence that
weve misidentified the categories involved in the syllogismweve erroneously
busted up one of the categories into two. And thats exactly what we didwe
busted the predicate category of things that enable Karl Kartoffelkopf to pass,
into two categories: people who are Karl Kartoffelkopf and people able to pass
Critical Thinking. The moral of the story: Keep a birds-eye view of the whole
syllogism, to best identify its three categories.
Do you notice what all three of our examples have in common? They all have
the same formthey are all instances of Barbara, just like that famous syllogism
about Socrates being mortal. As I mentioned before, this is the most popular
and easily understood syllogism type, so its a natural to be stated as an enthymeme. [BTW: The previous sentence was self-illustrating.]
Lets do another enthymeme:
Lawyers frequently undermine the law; because anyone who
ignores the truth in order to get an acquittal undermines the law.
Once we supply the missing premise, that some lawyers ignore the truth in order
to get their clients acquitted, we can see that this syllogism has a different form
than our old friend Barbara:
IaU
[L i I]
LiU
216
Sometimes it may be a bit tricky to pinpoint the missing premise or conclusion.

In that case, it might well be that the author should not have used an enthymeme
in the first place.
But lets say they did, and now you have to figure it out. Heres a suggestion:
Use a Venn diagram to represent what you are explicitly told, and then let the
Venn help you discover what implicit premise is needed to get to the conclusion.
Or if the conclusion is what is left implicit, diagram the premises and see what
conclusion can be read off the diagram.
For example, if we were to diagram the first premise of our example above, wed
get the following:
We need to be able to read off the diagram that there exists at least one lawyer
that is undermining the law. We can then see that if there is a lawyer that ignores
the truth in order to get an acquittal [see the next Venn below], then that lawyer
would be forced by the first premise to undermine the law [see the second
Venn below].
217

Some readers are skimming this section on enthymemes; so
some readers wont be doing very well on the Enthymemes
module of CT.
Whats the missing premise?
Answer: No reader skimming this section on enthymemes is going to be doing
well on the Enthymemes module. So, the syllogism could be represented as:
RiS
[S e W]
RoW
Ill leave it to you to test the validity of this one.
Whats the missing premise in this next one?
Since all Republicans favor cutting taxes, Bush must be a
Republican.
Answer: Bush favors cutting taxes. Thats in all probability what the speaker
has in mind. But lets put the resulting syllogism into our notation and test it for
validity.
RaF
[B a F]
BaR
218
Is it valid?
Answer: No! B is not guaranteed by the premises to be an R. Just as not all
syllogisms are valid, not all enthymemes are valid.
Some logic instructors will tell you that you are permitted to ascribe missing
premises only when those premises make the syllogism valid. This policy just
wont work though. There are just too many invalid enthymemes around, and
you have to be able to make sense of them and fully appreciate why theyre
invalid. So here is the best rule to follow:
When possible, attribute the author of an argument with the
implicit premise or conclusion that makes their syllogism
valid. But, if it is pretty obvious what their implicit premise
or conclusion is, you may rightfully attribute it to the author
despite the fact that it will make their syllogism invalid.
We want to be charitable readers, helping others to construct valid syllogisms
whenever possible. But sometimes it simply isnt possible, and theres a limit to
our charity.
OK, heres a last one for you:
Everyone there enjoyed the evening very much. So there were
some people who didnt enjoy the evening.
Whats the missing premise, and is the resultant syllogism valid?
Answer: The missing premise is that some people were not there. With that as
the missing premise, the syllogism could be represented and diagrammed as
follows.
TaE
[P o T]
PoE
219
We are unable to read the conclusion off the diagramthe asterisk would have
to be placed outside of E-land by the premises; just sitting on the E fence wont
do. So heres another example of an invalid enthymeme.
Pop quiz: With the help of the Venn above, what change in the
first premise would make the syllogism valid? But would the
necessary change be at the expense of one of the other ARG
conditions?
Some more to play with:

1. The baby must not be hungry, since its not
crying.
2. In light of the fact that finishing anything
difficult is of value, completing this lesson
will be valuable.
3. Robots leave no droppings, so none of the
animals at the zoo are robots.
4. Todays exam will be hard for me, since I
didnt have time to study as much as I should
have.
One for the road:

I told my dad that everyone hates me. He said
thats impossibleI havent met everyone yet.
This is from Rodney Dangerfield. Assess this
220
enthymeme and see if Rodney gets the last laugh

on his dad, from a logical point of view.

Youre all ready to practice and verify your skills at analyzing enthymemes; so
launch the Enthymemes module of the CT software and knock yourself out!
While youre working on this module, dont forget the skills you developed in
the modules Anatomy of an Argument, Standard Categorical Form, and
Venn Diagrams.
For a general review and test on all the information weve studied concerning
categorical logic, launch Level 3 of Whad Ya Know About CT?
221
Groundedness II: Propositional Logic
9. GROUNDEDNESS II: PROPOSITIONAL LOGIC

Youve now become a pro at assessing categorical arguments. Right? OK, then
put your syllogistic skills to work on the following argument:
1. If oil prices continue to increase, investor confidence
will decrease.
2. Oil prices will continue to increase.
3. Investor confidence will decrease.
The first premise looks easy: All times at which oil prices are increasing are
times at which investor confidence will be decreasing. The second premise
looks easy too: All things that are oil prices are things that will increase. And
the conclusion is no problem: All things that are instances of investor confidence are things that will decrease. Lets put this into our notation and see
where we are:
>O a <IC
Oa>
IC a <
Whoa! We maxed out on categories; were three categories over the limit!! But
thats weird; the argument sounds so simple and so valid.
The reason we have so many categories is that we seem to have busted up the
categories from our first premise in order to make the categories needed to state
our second premise and our conclusion. But if we didnt bust them upif we
stick with the categories >O and <ICthen wed get the following way of
representing our argument:
>O a <IC
>O
<IC
Whoa! Thats not a syllogism! Now were one category under the requirement,
and only one of the three statements is a categorical statement. Can categorical
logic even handle the job of confirming the validity of this argument?
Well, yes, it can. But only by getting a big hammer and making the argument fit
the categorical format. We can use a third category that has been sort of
hovering implicitly behind the scenes all this whilethe category of things that
existE. This is a strange, huge category, but notice how it allows us to finally
put this argument into syllogistic form: Now, our second premise is the claim
that at least one of the things that exist is increasing oil prices, and the conclusion
222
is the claim that at least one of the things that exist is investors decreasing confidence.
>O a <IC
E i >O
E i <IC
Now we finally have a syllogism, and a valid one at that. [Which I leave to you
to prove by means of a Venn, if you dont believe me.]
For categorical logic to have that much difficulty dealing with such an obvious
argument reveals a serious limitation of categorical logic. There must be an
easier way to handle such a simple argument, instead of using goofy catch-all
categories such as E. And indeed there is. In this chapter, we will study a system
of logic that can easily and efficiently represent our argument in the following
way:
>O <IC
>O
<IC
This system of logic is call propositional logic, and it will provide us with a
means of assessing the validity of arguments, just as did categorical logic.
Heres another case that categorical logic would find a challenge, to say the least:
If X is necessary for Y, then Z, but only if if A then B.
Put that in standard categorical form. And, of course, you cantits three
categories over the limit. And yet many statements are like this, especially
statements expressing bureaucratic rules and regulationsjust think about those
elaborate, contorted rules in the IRS Tax Code! If you want to appeal to one of
those rules to argue for a tax deduction, you better make it a valid argument, or
there may be penalties and interest to pay.
Categorical logic cant accommodate this case above, but propositional logic can.
We could easily symbolize the statement in this way:
[(YX)Z](AB)
This may look like symbol salad right now, but by the time were done studying
propositional logic, this will be perfectly intelligible, and
If X is necessary for Y, then Z, but only if if A then B,
223
will strike you as being as simple as the statement

If A then B.
OK, now that weve seen the need to supplement our categorical logic skills with
propositional logic, lets do it.
What is propositional logic?

Propositional logic concerns the deductive relations among propositions.
Oh, real profound! So what are propositions, then?
Propositions are simply statements. Think of them as descriptions of some world
or other [actual or imaginary]. For example:
The Mississippi River begins in Minnesota.
Minnesota gets quite cold in the winter, and I dont like that one bit.
You get used to freezing if you freeze long enough.
These are all propositions.
Some propositions are molecular propositions, built up out of other propositions. One of the examples just mentioned illustrates this: Minnesota gets quite
cold in the winter, and I dont like that one bit, is made up of two propositions,
viz., Minnesota gets quite cold in the winter, and I dont like it one bit that
Minnesota gets quite cold in the winter.
You can only break down a molecular proposition so far, until you get to its
atomic propositionsthose elementary propositions that cant be broken down
any further and still express a coherent thought. For example, Minnesota gets
quite cold in the winter, is a complete, coherent thought; but, Minnesota or
gets quite cold in the winter is not. Lets generically symbolize an atomic
proposition with the letter p.
Molecular propositions are made up of atomic propositions and connectives.
The best way to illustrate this is by means of an analogy. You already know that:
Atoms bond to form molecules.
Molecules bond to form compounds.
Compounds blend to form mixtures.
224
Well:
Atomic propositions bond by means of logical
connectives to form molecular propositions.
Molecular propositions bond by means of logical
connectives to form even more complex molecular
propositions.
Molecular propositions are used as premises and
conclusions to form arguments.
The logical connectives

Lets now introduce our line-up of logical connectives for propositional logic.
Negation
For any proposition, p, there is the negation of that propositionnot p. So, if we
begin with the proposition, It snowed four inches last night, its negation is the
proposition It is not the case that it snowed four inches last night.
The symbol most often used to stand for negation is the tilde: ~. So, for every
proposition p there is its negation, ~p.
There are numerous ways of negating propositions in ordinary language. The
sure way was illustrated above: It is not the case that p. Here are a few more
ways, which should help you to extrapolate to any others you may encounter:
Its false that p.
Its not true that p.
Its wrong to think that p.
Its impossible that p.
Its not a fact that p.
Its other than that p.
It fails to be the case that p.
I deny that p.
Its rather strange to call the tilde a connective; it just hangs on an atomic
proposition and doesnt bind it in any way to another proposition. But, lets just
go along with the convention on this one, and classify it as a connective and
classify ~p as a molecular proposition.
225
Conjunction
Take the proposition, I went to Seattle, and I had a great time. This proposition
more intuitively lives up to its label as a molecular proposition. It has a
connective [and] that we would comfortably consider to be a paradigm example of a connective, binding together two atomic propositions.
Ill use the letters p and q to stand for our generic atomic propositions, and Ill
use the dot to symbolize andp q. BTW: The ampersand [&] is also used to
symbolize and.
The propositions making up the conjunction are called conjuncts.
There are many ways of saying and in ordinary language; here is just a sample
of them:
p but q
p although q
p even though q
p however q
p in spite of q
p despite q
p yet q
p notwithstanding that q
p whereas q
both p and q
p besides q
p as well as q
p and then q
p albeit q
p at the same time q
p but also q
p furthermore q
p moreover q
not only p but q
p regardless of q
p while q
p nevertheless q
p in as much as q
p in addition to q
Remember when we were talking about how there can be a huge loss of information when we move from ordinary language to the language of categorical logic,
e.g., when you have to settle for saying that some As are Bs, when the original
claim said that most As are Bs? Well, the same holds true with respect to our
attempts to take statements from ordinary language and represent them in
propositional logic.
Compare the statements p despite q and p nevertheless q. The first statement
would indicate that q was a bit of a hurdle to ps being true, while the second
statement would indicate just the reverse. And yet, both get rendered as only
p q when put in standard propositional form. Much information is lost.
226
Similarly with the statements p and then q and q and then p. They could
both be represented as p q or q p. The temporal order stated in the original
sentences is lost.
Once again we are reminded that no language of logic will ever approach
ordinary languages ability to convey information. But then we must also
remember that the ability of ordinary language to assess the validity of arguments
is pathetic. And thats why were studying how to translate arguments from
ordinary language into propositional logic, to assess their validity, and then
translate them back into ordinary language againfast and reliably.
Disjunction
This molecular proposition is built using the connective or, as in Quiet down
or Ill call the manager and Im going to Chicago or Milwaukee next
weekend. Here are a few ways of stating a disjunction:
either p or q
p or else q
alternatively p and q
p otherwise q
The symbol standardly used for or is the wedgep v q.
The propositions making up the disjunction are called disjuncts.
Conditional
The conditional statement has the form, If p then q: If you get your report in on
time, you will get a bonus, or If you dont get your report in on time, you will
be fired. And now you see why conditionals are so important!
The symbol we will be using to represent the conditional relation is the
horseshoe. So, If p then q, gets symbolized as p q. FYI: Sometimes, an
arrow [>] is used instead of the horseshoe, to represent the conditional.
The propositions making up the conditional get their own separate names: The
proposition after the if is the antecedent, and the proposition after the then is the
consequent.
Theres a bundle of ways to state the conditional relation in ordinary language. If
you just browse among various directions, contracts, laws, rules, and regulations,
youll run into many of them. [BTW: Many of my students have told me that
227
after studying this chapter, they were able to do their taxes much more easily.
Sorry, I still have no advice for making it easier to write the IRS checks.] Here,
then, is a sample of how diversely youll find the conditional, if p then q, stated:
pq
pq
pq
pq
pq
pq
pq
pq
pq
pq
pq
p q or q p
p q or q p
pq
pq
pq
pq
pq
pq
pq
pq
pq
pq
pq
pq
pq
pq
~qp
~qp
~q~p
~q~p
when p, q
q, if p
q, given p
q provided that p
p, only on condition that q
p, only because q
p only if q
none but those that are q are p
q is a necessary condition for p
p is contingent upon q
p is a sufficient condition for q
q, on condition that p
q is conditional upon p
all times that p are times that q
q, whenever p
in so far as p, it follows that q
q, so long as p
q follows from p
q is entailed by p
q, to the extent that p
q, in order that p
p implies q
p is q
p requires that q
from p, it follows that q
p means that q
q, assuming p
p, unless q
p, except on condition that q
not p, unless q
not p, without q
Find some exemplary conditionals youre comfortable with, e.g: If its a cat, its
a mammal. Run your examples through the list above, to get familiar with all
these ways of expressing the conditional.
One of the most important ways of characterizing the conditional is in terms of
necessary or sufficient conditions: The antecedent is a sufficient condition
for the consequent, while the consequent is a necessary condition for the
antecedent. This may sound like gibberish right now; but it soon wont, as you
228
think it through a bit and get a feel for the conditional relation: If p, then q must
be the case too, and ps being the case is all it takes for q to be the case.
Note that the statements, q is conditional upon p, and q, on condition that p,
are ambiguousone cannot tell, out of context, whether p is a sufficient
condition for q or a necessary condition for q.
A few of the phrases in the list above should ring a bell based on your knowledge
of the various ways of stating the A categorical statement in ordinary language.
In fact, the A statement and the conditional are two ways expressing the same
relation. With this in mind, the conditional should be easier to understand: The
subject of the A statement is the antecedent of the conditional; and the predicate
of the A statement is now the consequent of the conditional. And now you know
exactly which conditional statement is being expressed by means of only, only
if, only because, and none but. They all refer to the consequent, just as in
A categorical statements, when they all referred to the predicate.
Another popular way of expressing a conditional is with the word unless. Look
at the last four entries on the list above and see if you can formulate a rule about
how to put unless into standard propositional form.
The rule is: Remove unless and insert If not. Unless negates the antecedent,
so the rest of the sentence must be the consequent. [Without works the same
way in this context; e.g., I wont go without a fight is the same as saying If no
one puts up a fight, I wont go.]
Equivalence
For example, Being a leap year is equivalent to being a calendar year in which
February has 29 days, instead of 28. The equivalence relation is also called the
biconditional. The most frequent way of expressing it is p if and only if
qIts a leap year if and only if February has 29 days.
We will be using the triple bar to represent the statement that two propositions
are equivalent: p q.
FYI: Some people use the double arrow [<>] instead of the triple bar to
symbolize the biconditional. And logicians have actually coined a new word,
iff, to stand for if and only if.
Here, then, are some of the ways of stating the equivalence relation:
229
p is a necessary and sufficient condition for q

p is defined as q
p is by definition q
p means [by definition] q
p is identical to q
p is q [the is of identity]
p is equal to q
p exactly when q
p when and only when q
p only when warranted by q
p just in case q
p just in so far as q
Note again the difference between the is of identity and the is of predication.
A bachelor is an unmarried male states an equivalence, but A bachelor is a
male is merely a conditional.
The last two phrases on the list above are the only ones that arent really obvious.
[Personally, I think they should be used to mean the same as only if.] You
should be aware that when someone does use the rather technical phrase p just
in case q, chances are theyre asserting an equivalence and not merely a
conditional.
Neither p nor q
To say Neither p nor q is to say, Its not the case that either p or q. Its a way
of negating the disjunction. And, if its not the case that either p or q, then p
must not be the case and q must not be the case. Thats why neither p nor q gets
expressed in both of the following ways:
~(pvq)
~p~q
Something you just have to do

In order to be proficient at assessing the validity of propositional arguments
which is, after all, our goal here!you must be able to recognize all the various
ways of stating these basic propositions in everyday language. So please
thoroughly study the lists Ive just provided. And be on the alert for any new
ways of stating propositions.
230
Group picture
Here then is the complete line-up of our connectives for propositional logicor
prop logic, as its called for short.
~p
pq
pvq
pq
pq
Something to think about: Compare the statements possible in

categorical logic and the statements possible in prop logic.
1. What statements can be made in prop logic that cannot be
made [at least not without major effort] in categorical
logic?
2. And what statements can be made in categorical logic
that cannot be made [at least not without major effort] in
prop logic?
Answer: Categorical logic can easily express the conditional by means
of the A statement, but falters on the other connectives. Prop logic
falters on the I and O statements, which categorical logic expresses so
easily.

Its already time to begin practicing your skills in prop logic. Launch the
Conditionals module in the CT software and master all the ways of stating the
conditional in daily discourse.
Truth
Weve studied how to build molecular propositions, using atomic propositions
and our basic connectives. Now lets look at how the truth and falsity of those
molecular propositions are determined.
Prop logic also goes by the name of truth-functional logic. And for good
reason, as we will soon see.
231
The truth of an atomic proposition is simply a matter of accurately describing or

corresponding with the world. To use the most famous example: The statement
Snow is white is true if and only if snow is actually white.
The truth of a molecular proposition is likewise a matter of accurately describing
the world. But the truth of a molecular proposition is also determined by the
truth of its atomic propositions and the type of connectives used to build it.
The analogy to chemistry still holds here: The nature of a molecule is a function
of 1) the nature of its atoms and 2) the type of bonds holding the atoms together
to form the molecule. [Just think of how waters unique nature, to expand when
frozen, is determined by its atoms and their special hydrogen bonds.]
And so it is that the truth or falsity of a molecular proposition is a function of 1)
the truth or falsity of its atomic propositions and 2) its connectives.
The truth-functional definitions

To illustrate how the connectives determine the truth or falsity of molecular
propositions, we can use what is called a truth table. The truth table will allow
us to completely detail the behavior of the connectiveswith the truth table, we
will provide what is called a truth-functional definition for each connective.
Here is the most basic truth table possible; the one for the atomic proposition p:
p
T
F
This truth table is just representing the obvious: For any possible atomic
proposition p, p is either true [T] or it is false [F]. Thats the entire menu of
possibilities with respect to p. The truth table is like the naked Venn diagram in
this respect: It exhaustively lays out all the possible circumstances of truth and
falsity for a proposition.
Before we continue, I must introduce a term that will make this discussion much
easier. Just as color comes in many values [e.g., red, green, and blue] and
polarity comes in two values [positive and negative], so too truth comes in two
valuestrue and false. These are what are called truth values. So we can talk
about the truth or falsity of a proposition or we can talk about its truth value
being either true or false. [FYI: There are three-valued logics, in which a
proposition can be true or false or undetermined. This is the kind of logic needed
for such sophisticated studies as quantum mechanics. Lets be good to ourselves
and not worry about such multi-valued logics here.]
232
The stage is now set:

Lets see how putting a tilde on an atomic proposition affects its truth or falsity.
We begin with the possible truth values for the atomic proposition and then
record how negation makes a true proposition false and a false proposition true.
p
T
F
~p
F
T
This is simply what it is to negate [or deny] a proposition: If I have a hole in

my pocket is true, then, Its not the case that I have a hole in my pocket is
false. And, if I have a hole in my pocket is false, then, Its not the case that I
have a hole in my pocket is true. This is just what not meansnothing
earth shattering here.
Lets see how conjunctionthe dotworks:
p
T
T
F
F
q
T
F
T
F
pq
T
F
F
F
The first two columns of the truth table list all the possible truth values for
atomic propositions p and q. Because we now have two atomic propositions,
there are more possible ways for the two to be true or falsethey both could be
true, one could be true while the other is false, one false while the other is true, or
they both could be false. The third column details the truth value of p q in each
of these four cases.
Basically, the truth table just lays out what and means. Just think about it:
Under what circumstances is it true that Im in Chicago and having fun? When
Im in Chicago but having a lousy time? No. When Im in Minneapolis having
fun? No. When Im in Minneapolis having a lousy time? No! Only when Im
both in Chicago and having fun. A conjunction is true only when both of its
conjuncts are true.
Nextdisjunction.
233
p
T
T
F
F
q
T
F
T
F
pvq
T
T
T
F
This is just what or means. Think it through by means of an example: When is

it true, and when is it false, that Im going to either Minneapolis or Chicago this
summer to look at office furniture? Well, if I end up going to both cities to look
at office furniture, Ive really come through on my claim. The only time I
reneged on my claim is if I dont go to Chicago and I dont go to Minneapolis.
So, a disjunction is false if both of its disjuncts are false, otherwise its true.
Lets see how the conditional works.
p
T
T
F
F
pq
T
F
T
T
q
T
F
T
F
Whoa! That seems strangehow can a conditional be true when its atomic parts
are all false?! Usually this is chalked up to being a peculiarity of the material
conditional, as logicians call it. But that is not a very satisfactory answer. Let
me at least try to make some sense of our truth tables analysis of the conditional.
Think of the conditional as being like a promise. Lets say I promise you that
if it rains, Ill have my umbrella. Have I come through on my promise on a rainy
day that I have my umbrella with me [the first row of the truth table]? Sure.
How about the day that its raining and I dont have my umbrella [the second
row]? No; Ive broken my promisemy conditional statement is falsified. How
about those days that it is not raining? How about when I have my umbrella with
me on a sunny day [because I always have my umbrella in my pack]? I sure
havent broken my promise, have I? And, have I made a false promise when I
decide to leave my umbrella home on a perfectly sunny day? No. On those days
when it is not raining, my promise to have my umbrella when its raining goes
untested, but my promise is still alive and wellits still a true promise.
And so it is with the conditionalit is false only when its antecedent is true
and its consequent is false. The rest of the time it is true, if only by virtue of
being untested.
234
Think of the usual graduation requirements for the baccalaureatein order to

graduate one must complete the general education program: g ge. [In prop
logic too, make your symbols make sense!] The only way for this conditional
[this rule] to be made false is if there can be a person who graduates without
knocking off their gen eds. The fact that there are a lot of non-graduates running
around without their gen eds done [e.g., all the freshmen!], doesnt take anything
away from the truth of this graduation requirement.
One last connective remains: Heres how the triple bar determines truth values;
heres what equivalence means.
p
T
T
F
F
p q
T
F
F
T
q
T
F
T
F
Think about the statement, A kilogram is equivalent to 2.2 pounds. What is it

for these two weights to be identical? It means that when something is a
kilogram, its also 2.2 pounds, and when something is not a kilogram, its not 2.2
pounds. And if there ever were something that is one without being the other,
then a kilogram and 2.2 pounds wouldnt be identical.
So two propositions are equivalent when they always have the same truth
values.
Heres something else you simply must do!

Get these truth-functional definitions of the connectives down cold; know
these basic molecular propositions [when theyre true and when theyre
false] like you know how to breathe.
And when might that be true?

Truth tables do a great job of detailing how our connectives determine truth
values, i.e., under what conditions our basic molecular propositions are true or
false. But they can do this for any propositions whatsoever. So, no matter what
proposition you find, even the beefy one I threw at you at start of this chapter,
viz., [ ( Y X ) Z ] ( A B ), you can figure out what the world would have
to be like for that proposition to be true [even if you dont understand what the
proposition even means!].
235
Thats one of the many uses of the truth table. So lets try it. When is the
following proposition true and when is it false?
(p q) (q p)
Lets build a truth table for this proposition. The columns of the truth table, in a
sense, record how the proposition was constructed. This proposition began with
the atomic propositions p and q. The first molecular proposition built up was
p q. The second molecular proposition was q p. And then these two
propositions were conjoined to form the final product, (p q) (q p).
This process of construction goes on the top of our truth table:
p
pq
qp
(p q) (q p)
We next record, in the first two columns, all the possible ways that the atomic
propositions can be true or false, just as we did before when we gave the truth
tables that truth-functionally defined the connectives:
p
T
T
F
F
q
T
F
T
F
pq
qp
(p q) (q p)
And then we record how those truth values determine the truth values of our
molecular propositionsfirst the two conditionals and then the final conjunction:
p
T
T
F
F
q
T
F
T
F
pq
T
F
T
T
qp
T
T
F
T
(p q) (q p)
T
F
F
T
[That fourth column, for q p, might be faking you out: In the first row, when q
is T and p is T, q p is T, and in the second row, when q is F and p is T, q p is
T, but in the third row, when q is T and p is F, q p is finally F, and so on.]
236
And now we know under what truth conditions (p q) (q p) would be true

and under what circumstance it would be false. Does the last column look
familiar? It should; its the column for the triple bar:
p
T
T
F
F
q
T
F
T
F
pq
T
F
T
T
qp
T
T
F
T
(p q) (q p)
T
F
F
T
p q
T
F
F
T
And now we see why the equivalence relation is also called the biconditional,
and why p if and only if q is a frequent way of expressing the triple bar: p q
is equivalent to the conjunction of p q and its converse.
[Yes, indeed, you are familiar with that notion from categorical logic. Mini pop
quiz: whats the contrapositive of p q?]
Lets do another one: Under what circumstances is it true that ~(p ~q)?
First we must figure out how many rows the truth table will have. You
probably think it will have four, and youre right. Heres how to calculate
exactly how many rows your truth table will have, for any proposition youre
working with:
Rows = 2
n is the number of atomic propositions. We use 2 to the nth power, because we

are working with only a two-valued logicpropositions are treated as either true
or false. [FYI: The number of rows in a truth table for a three-valued logic, in
n
which propositions are either true or false or indeterminate, would be 3 .]
With 2 atomic propositions in ~(p ~q), then, we will then have 2 squared rows
in our truth table.
How many columns will we have?
Well, first, one for each atomic propositionp and q. And then, one for each
molecular proposition we build on the way to constructing the final one were
doing the truth table for. So the smallest molecular proposition we build is ~q.
The next largest is the conjunction, p ~q. And the final is the negation of that
conjunction, ~(p ~q). That makes a total of five columns:
237
~q
p ~q
~(p ~q)
Now were ready to start making truth-value assignmentsviz., handing out

the Ts and the Fs. You may have noticed a pattern in how Ive done the truth
tables so far. Let me explain the method I used to list all the possible different
combinations for the atomic propositions, with no repeats.
Start with the first column, divide its rows in half, and alternate truth-value
assignments between the halves, beginning with the Ts. Then, go to the second
column and divide its halves in half and alternate truth-value assignments
between each half-of-a-half, beginning with the Ts. And so on, until youre out
of columns of atomic propositions. See?
p
T
T
F
F
q
T
F
T
F
~q
p ~q
~(p ~q)
Then, make your truth-value assignments for the other molecular propositions,
based on the truth values you have already assigned and what the connectives are
for those molecular propositions. So, when q is T, ~q is F, and when q is F, ~q is
T, etc. And for p ~q, when p is T and ~q is F, then p ~q will be F, etc. Here is
where you must know how the connectives determine truth valueswork on this
until you dont even have to think about it anymore. Your truth table will then
look as follows.
p
T
T
F
F
q
T
F
T
F
~q
F
T
F
T
p ~q
F
T
F
F
~(p ~q)
T
F
T
T
And these, then, are the circumstances under which ~(p ~q) is true and the one
circumstance under which it is false.
Where have you seen this column of truth values before? Thats right: Its
identical to the column for the conditional, p q. In a sense, what ~(p ~q) is
claiming is that it will not be the case that [the antecedent] p is true while [the
238
consequent ] q is notand that is exactly what is claimed by [the conditional]

p q.
p
T
T
F
F
q
T
F
T
F
~q
F
T
F
T
p ~q
F
T
F
F
~(p ~q)
T
F
T
T
pq
T
F
T
T
Proving equivalence
We have twice illustrated another use of the truth table: To prove that two
propositions are equivalent, i.e., just two ways of saying the same thing. If their
two columns of the truth table are identical, they are equivalent. That, after all,
was the truth-functional definition of equivalence: That the propositions have the
same truth values in all possible circumstances.
Remember how we proved two statements were equivalent in categorical logic?
We showed they had identical Venn diagrams. Well, similarly for proving equivalence in prop logic, only now our medium of proof is the truth table instead of
the Venn.
So: Identical columns; identical propositions.
I dont know what it means, but I know when its true!

Lets do one last one, and lets beef it up this time When is it true that
(p q) r, and when is it false? Wowthree atomic propositions. So how
many rows will this truth table have?
Itll have 2 to the third powerthats 8 rows. And how many columns? Well,
one each for the three atomic propositions, and then one for p q and then one
for the entire conditional enchilada. That makes a total of five columns.
Start off by making the truth-value assignments for the atomic propositions
remember our method of alternating truth values between the halves. And
then use your knowledge of how the connectives determine truth values, to make
the rest of your assignments.
Your resultant table should look like this:
239
p
T
T
T
T
F
F
F
F
q
T
T
F
F
T
T
F
F
r
T
F
T
F
T
F
T
F
pq
T
T
F
F
F
F
F
F
(p q) r
T
F
T
T
T
T
T
T
A couple observations: If someone were to ask you when (p q) r is false you

could tell them exactly under what circumstance it is falseviz., when the world
is such that p is true, q is true, but r is falsethe second row of the truth table.
(p q) r is true a lot! Thats because its antecedent, (p q), is false a lot. And
thats a shortcut you can begin to take advantage of, as you are doing your truth
tables: When the antecedent is false, the entire conditional is true! [Remember
my analogy between a conditional and an untested promise?an untested
promise is still true.]
Pop quiz: Do the truth table for (p q) r. Notice anything?
Soup or salad
You are reading a menu in a restaurant, and it says that with the entrees, You
may have soup or salad. Youve done a great job of learning the truthfunctional definitions of the connectives, and you now think to yourself, A
disjunction is false only under one case, viz., when both disjuncts are false;
otherwise its true. With this fresh in your mind, you tell the person taking your
order that youd like soup and salad [both disjuncts] with your entree. The
person is not impressed and promptly sets you straight on the real meaning of the
phrase. Which is?
You may have soup or salad, but not both. That is to say, You may have
soup or salad, but not the conjunction of soup and salad. We are now ready to
put our proposition into notation. Since the first letter of both atomic propositions is s, lets pick other letters for our symbol choice: (U v A) ~(U A).
Do the truth table for this and show how your server was correct [you were just
being a jerk]. How many rows and how many columns? [Dont look down!]
240
OK, heres the finished truth table:

U
T
T
F
F
A
T
F
T
F
UvA
T
T
T
F
UA
T
F
F
F
~(U A)
F
T
T
T
(U v A) ~(U A)
F
T
T
F
This is called the exclusive disjunctionwhen the truth of the proposition

requires that one or the other disjunction is true, but excludes the possibility of
both disjuncts being true.
Disjunction, as we defined the connective earlier, was the inclusive disjunction.
Exclusive disjunction is symbolized by putting a bar over the connective for
disjunction: p V q. So, we could introduce a new connective into our repertoire:
p
T
T
F
F
q
T
F
T
F
pVq
F
T
T
F
But this would be yet another connective, with an accompanying truth-functional

definition, to keep track of!
I think your server was right; lets not introduce another connective. Lets keep
things simple and instead just get a bit long-winded with our current batch of
connectives and use the likes of (p v q) ~(p q), when expressing an exclusive
disjunction, instead of using p V q.
So be on the lookout for how a disjunction is being used in argumentation.
Sometimes it will be obvious, e.g., when I say that Im going to either Chicago or
Minneapolis this evening. Sometimes its not, and then let the inclusive
disjunction be your default for purposes of interpretation.
Inconsistent, necessary, or contingent?

Weve gotten a lot of work out of truth tables already. Weve used them to
define the connectives, prove the equivalence of propositions, and detail when a
proposition is true and when it is false. This third function of truth tables
naturally enables a fourth: To demonstrate whether a proposition [any
proposition] is inconsistent or necessarily true or merely contingent, viz.,
sometimes true and sometimes false.
241
I mentioned before that the truth table can ironically tell you when a proposition
is true and when its false even if you dont know what the proposition means.
Well, the truth table can also tell you if the proposition is inconsistent or not even
if you dont know what it means. To be inconsistent is to be self-contradictory.
To be self-contradictory is to be logically impossible. It is to be necessarily
falsefalse in all possible cases. And how is that expressed on the truth table?
With a column of Fs. Let me illustrate with a simple example: p ~p. This is
the most blatantly inconsistent statement.
p
T
F
~p
F
T
p ~p
F
F
On the other hand, when a proposition is necessary, when is it true? Always! So

what is a necessarily true propositions column going to look like on the truth
table? All Ts. Such a proposition is called a tautology. Take a look:
p
T
F
~p
F
T
p ~p
F
F
~(p ~p)
T
T
[So the proposition that states that a contradiction doesnt happen, is necessarily
true.]
As an example of a contingent proposition, viz., a proposition that is true under
some possible circumstances and false under some, you can pick your favorite
truth-functional definition of a connective. Lets pick the conditional:
p
T
T
F
F
pq
T
F
T
T
q
T
F
T
F
The main connective

Is the following a conditional or a conjunction?
pqr
Answer: Its neither. Its not even a proposition! Its symbol salad. Its
ambiguous as it stands and is not, therefore, a wffa well-formed formula in
prop logic.
242
Disambiguating this collection of squiggles finally makes it into either a conditional or a conjunction. We use parentheses to make it either a conditional or a
conjunctionwe establish the main connective. So, we could make p q r
into a conditional(p q) ror into a conjunctionp (q r).
Note the difference between these two:
(p q) r
p (q r)
The first states that If p and q, then r; the second states that p and if q then r.
Lets try another one. Whats the main connective?
~p v q r
Wow, this is really ambiguous! This could be a conditional, or a disjunction,
or it could even be the case that the tilde is the main connective. Here are all the
ways to make propositions out of the mess above:
1. (~p v q) r
A conditional, with a disjunction for an

antecedent.
2. ~(p v q) r
A conditional with a negated disjunction for an

antecedent.
3. ~p v (q r)
A disjunction, with a conditional for a disjunct.
4. ~[(p v q) r]
The negation of a conditional.
5. ~[p v (q r)]
The negation of a disjunction.
And heres how you would say these propositionsnote their differences:
1. If either not p or else q, then r.
2. If its not the case that either p or q, then r.
3. Either not p or if q then r.
4. Its not the case that if either p or q then r.
5. Its not the case that either p or if q then r.
243
Can you tell the differences among these? If you cant, you dont have to take
my word for it that theyre different. You can tell for yourself whether Im right
or wrong, simply by doing what? Thats right, just by doing truth tables for
them.
FYI: If you run into such a complicated molecular proposition that you need yet
another means for disambiguating it, besides the parentheses and the square
brackets, you have my sympathy and you have what are called braces{ }.
Wed need them if we wanted to deny that beefy proposition I used at the
beginning of this chapter:
~{[ ( Y X ) Z ] ( A B )}
Compare this to:
~[ ( Y X ) Z ] ( A B )

Now that weve worked with a variety of molecular propositions and discussed
the issue of determining the main connective, youre ready to practice putting
statements into propositional notation. So launch the Standard Propositional
Form module in the CT software. Start with the Easier level and eventually
shift to the Harder level, where things get interesting ;-) Your goal is to become
so proficient with our notation that you can symbolize propositions as quickly as
people blurt them out in daily argumentation.
Propositional arguments
Weve studied how to build propositions and how to find out under what
circumstances they are true or false.
Now its time to build arguments out of those propositions and find out if those
arguments are valid or invalid.
As youve probably guessed by now: We will use the truth table for this too.
[It slices, it dices, its ten utensils in one!]
To see why the truth table is such an excellent tool to use for testing validity,
lets just refresh our memories again on what validity is. Think back
244
Validity: Under no possible circumstances are the premises

true while the conclusion is false.
So this is why the truth table is so good for testing validityit can completely
lay out truth values of propositions [premises and conclusions] under all possible
circumstances. And then all we have to check is whether there are any
circumstances under which the premises are true when the conclusion is
false.
Lets begin with this simple argument and illustrate how to test its validity using
a truth table.
pq
p
q
The first thing we do is to make a truth table for all the arguments propositions:
p
T
T
F
F
pq
T
F
T
T
q
T
F
T
F
Then, we can label those propositions, as parts of the argument:

P2
p
T
T
F
F
C
q
T
F
T
F
P1
pq
T
F
T
T
Is there any row in which the premises are true while the conclusion is false?
No.
If you cant find a row on the truth table in which the premises are true
while the conclusion is false, the argument is valid.
So this is a valid argument:
pq
p
q
245
Moreover, any argument with this form will be valid too. Weve proven it by
means of the truth table.
This argument form is historically called Modus Ponens. But, to help us
remember the name of this argument form, lets rename it by what it does. [Like
giving a nickname to someone. For example, Ill never forget who Booger was,
and what he did.] What would be a good name for this argument?
pq
p
q
Well, its second premise is doing what to the first premise? Its affirming the
antecedent of that first premise. [You were about to say that, right?] This name
tells you exactly what the argument looks like: Since the only proposition with
an antecedent is a conditional, the first premise must be a conditional, and the
second premise is the antecedent of that conditional, and the conclusion is whats
left over.
Now that we have proven that Affirming the Antecedent is a valid argument
form, you can prove that an argument is valid simply by correctly citing it as
an instance of Affirming the Antecedent.
This is handy: You can now use the argument form as a template for spotting
one type of valid argument. In fact, Affirming the Antecedent is probably the
most frequently used propositional argument. In fact, the argument we used to
kick off this chapter was an instance of Affirming the Antecedent:
1. If oil prices continue to increase, investor confidence
will decrease.
2. Oil prices will continue to increase.
3. Investor confidence will decrease.
>O <IC
>O
<IC
Lets try testing the validity of another argument, using the truth table:
pq
q
p
This should be pretty easy, since this argument is made up of the same propositions, so we can use the same truth tablewe only need to relabel its columns:
246
C
p
T
T
F
F
P2
q
T
F
T
F
P1
pq
T
F
T
T
And now we can ask the magic question: Is there any row in which the
premises are true while the conclusion is false? Yes there isthe third row of
truth-value assignments details exactly that case.
Such a row is called a counterexample. It proves that this argument is invalid.
Lets name this argument form, again by what it does. Its Affirming the
Consequent. And any token argument that has this form will be invalid, so you
can now use this argument form as a template.
Some cases of Affirming the Consequent are quite obvious, and youd never be
fooled by them:
If an animal is a dog, then its a mammal. Bobs pet is a
mammal (it just gave live birth). Therefore Bobs pet is a dog.
Bob could very well have, for a example, a cat for a pet. A cat, then, is a
counterexample that illustrates how these premises fail to guarantee the truth of
this conclusion.
But sometimes, things are not so obvious:
The sign said Gold permit parking only. And I have a Gold
permit. So I should not have been ticketed for parking in the lot.
But there may be other necessary conditions, besides having a Gold permit, that
must be met, e.g., that the parking not be overnight. The sign merely states that
if one may park then one must have a Gold permit. And thats why this person
who was ticketed for overnight parking, has only a fallacious leg to stand on.
Lets name yet another argument and test it for validity:
pq
~p
~q
247
Rather than affirming the antecedent, this one is Denying the Antecedent. Here
is the truth table we need to test itnote the added columns for molecular
propositions ~p and ~q.
p
T
T
F
F
q
T
F
T
F
P2
~p
F
F
T
T
C
~q
F
T
F
T
P1
pq
T
F
T
T
Are there any circumstances [rows] in which the premises are true and the
conclusion is false? Yesthe third row. So Denying the Antecedent is an
invalid argument form wherever you might find it. Heres an example:
Whenever Kevin gets bogged down in mindless paper work for
the University, hes bummed out. But lately, he hasnt been
plagued with paper work. So he must not be bummed.
I get depressed for a gazillion reasons, mindless paper work being only one of
them. [Another being that a certain word-processing application is unable to
reliably deal with a fair number of tables in a modest-length document. Sorry,
but I needed to vent a bit.]
pq
~q
~p
The name should spring to mind: Denying the Consequent. Historically, this
goes by the name of Modus Tollens. It too is a very frequently used argument,
so its crucial to be able to quickly spot it in daily discourse and know whether
its valid or not. Look at its truth table and ask the magic question of validity:
p
T
T
F
F
q
T
F
T
F
C
~p
F
F
T
T
P2
~q
F
T
F
T
P1
pq
T
F
T
T
Denying the Consequent is validthere is no circumstance [row] in which the

premises are true but the conclusion false. This argument form is very important
248
for inferring what follows when one is out of step with rules and regulations, for
instance:
In order to qualify for the stock options program, one must have
been employed full-time for at least one year. While you have
been here for one year, it was not as a full-time employee; so I
regret to inform you that you are not eligible to participate in the
Company stock options program.
Argument forms such as Denying the Consequent [and others!] can be hidden
from view sometimes. Here is an example:
If Peter showed up at the party, Paul did not have a good time.
But in fact Paul had a great time; so Peter did not go to the
party.
We would symbolize this argument as follows:
p ~g
g
~p
This is indeed an instance of Denying the Consequentthe second premise is
equivalent to saying that it is not the case that Paul did not have a good time,
thereby denying the consequent:
p ~g
~~g
~p
Lets test some arguments that have some other connectives. How about this
one?
pvq
~p
q
Lets call this argument form Denying the Disjunct. Is it valid?
p
T
T
F
F
C
q
T
F
T
F
P2
~p
F
F
T
T
~q
F
T
F
T
249
P1
pvq
T
T
T
F
Its valid. The reason I included that unused column for ~q, is to let you see that
it wouldnt matter which disjunct is denied, the other disjunct would validly
follow.
Heres an example of this argument form:
Ill be there for Thanksgiving or the Holidays. Oh, but now I see
by my calendar that Im already committed to be elsewhere for
Thanksgiving; so Ill be there for the Holidays.
BTW: I offer these examples to help you get a real feel for these argument
forms. If they dont work and you find better examples, please use them instead,
because theres nothing like a good example to help you learn these argument
forms and how to apply them.
How about this argument form?
pvq
p
~q
Call this Affirming the Disjunct. Is it valid?
P2
p
T
T
F
F
q
T
F
T
F
C
~q
F
T
F
T
~p
F
F
T
T
P1
pvq
T
T
T
F
We see, by the truth table, that Affirming the Disjunct is invalid. The first row
of the truth table tells us exactly the circumstances in which there is a
counterexample. [Mini-pop-quiz: Which row is the counterexample when the
other disjunct is affirmed?]
And heres an example:
Either Ill go to Minneapolis or Chicago to look at office
furniture. Since I looked at furniture in Minneapolis, I wont be
going to Chicago.
By looking at this example, you can see even more clearly why Affirming the
Disjunct is invalid. Its because its disjunction is an inclusive disjunction, which
can be true by having both its disjuncts true.
250
Here is another argument form for you to investigate:

pq
qr
pr
This is called the Hypothetical Syllogism, appropriately enough
hypothetical because all its propositions are conditionals, syllogism because
if we were to translate these propositions into categorical statements they would
form a syllogism. [Do you recognize which one?]
Lets test it for validity. How many rows and columns will its truth table have?
n
Remember, the number of rows = 2 , where n = the number of atomic
propositions. So there will be 8 rows. And there will be 6 columnsfor 3
atomic propositions and 3 molecular propositions.
And remember how to make all those initial truth-value assignments to the
atomic propositions? Halve the columns and alternate truth-value assignments;
then halve the halves, until you run out of columns.
Then, just make your truth-value assignments to the molecular propositions as
usual. The only difference here is that you are making a few more of them.
Your knowledge of when a conditional is true and when a conditional is false
should get a real work out here, and should begin to be automatic [when p is T
and q is T, p q is T; etc., etc.] .
p
T
T
T
T
F
F
F
F
q
T
T
F
F
T
T
F
F
r
T
F
T
F
T
F
T
F
P1
pq
T
T
F
F
T
T
T
T
P2
qr
T
F
T
T
T
F
T
T
C
pr
T
F
T
F
T
T
T
T
Are there any rows in which the two premises are true while the conclusion is
false? No, so this is a valid argument form.
In fact, this is our old friend Barbara, just dressed in propositional clothing.
Your familiarity with this argument form in categorical form, should make it
easy for you to spot it in the field, so Ill spare you an example.
251
Sometimes, however, the Hypothetical Syllogism is disguised a bit, as they were

when we studied categorical logic. The most frequent disguise is the
contrapositive. An example will make this point obvious:
~q~p
qr
pr
If you just take the contrapositive of the first premise, youve got our old friend
Barbara back again.
The short truth table

These truth tables are getting a bit long! And things can only get worse. Just
do the math and see how things can quickly get out of control: Weve seen how
an argument with 3 atomic propositions needs an 8-row truth table to test it for
validity. Well, 4 atomic propositions need 16 rows. Five atomic propositions
need 32 rows. And 6 atomic propositions need 64 rows. And it doesnt take that
complicated an argument to get you into 6 atomic propositions! You want to
churn out a 64-row truth table? I dont!
Luckily, theres a better way.
When someone claims that their argument is valid, they are daring youtheyre
saying, in essence, You will never find a case in which my premises are true and
my conclusion is false. Well, what you are doing with a truth table is taking that
dare, looking for any case in which the premises are true and the conclusion is
false.
The problem is that the truth table hunts for such a case so inefficiently, by
completely laying out all possible cases, including those in which the premises
are false and those in which the conclusion is true. It would be much more
efficient to strategically hunt for the case we are being dared to find, and ignore
the restwe should just hunt directly for truth-value assignments in which the
premises are true when the conclusion is false.
If we look again at the truth table for the Hypothetical Syllogism, we see that
only 2 of the 8 rows were the least bit interesting in this respectthey were the
rows in which the conclusion was false. Ive shaded them in the truth table
below.
252
p
T
T
T
T
F
F
F
F
q
T
T
F
F
T
T
F
F
r
T
F
T
F
T
F
T
F
P1
pq
T
T
F
F
T
T
T
T
P2
qr
T
F
T
T
T
F
T
T
C
pr
T
F
T
F
T
T
T
T
But even though these two rows were interesting candidates for counterexamples,
they did not pan out, because they were not rows in which the premises were
both true. And thats why the argument is valid.
I propose we do some serious trimming on the truth table and adopt the short
truth table method for testing validity.
We cut right to the interesting rows of the truth table by assigning truth
values to the conclusion so as to make it false, and then see if, under those
truth values, we can make the premises all true.
Here, then, are the steps in applying the short truth table method for testing the
validity of prop arguments:
Step one: To apply the short truth table method, you must lay out your
argument horizontally, instead of vertically. [Stacking the argument vertically
does help to reveal its argument form, though.] A slash is used to indicate where
the premises end and the conclusion begins [some people indicate this with a
little pyramid of dots [] and others use both [/]. Ill just use the slash:
pq qr / pr
Step two: Assign the truth values necessary to make the conclusion false. To
indicate that we are making p r false, we put an F under the horseshoe. To
make p r false, we have to make p true and r false; so, we put a T under the p
and an F under the r.
pq qr / pr
TFF
Step three: Carry over your truth-value assignments. The truth values for p
and for r have now been determined for the rest of the test, so make those same
253
truth-values assignments to p and r wherever you find them in the premises. [Not
making the same assignments would be just plain inconsistent.]
pq qr / pr
T
F TFF
Step four: Assign the truth values necessary to make the premises true. Or at
least try to. Lets begin with the first premisein order to make p q true,
given that p is already true, we must make q true. So lets make those truth-value
assignments below.
pq qr / pr
TTT
F TFF
But when we carry that truth-valued assignment over to the second premise and
make q true in q r, we see that that would make q r false, in light of the fact
that r is already false:
pq qr / pr
TTT TFF TFF
Step five: The final verdict. If the premises c a n all be true while the
conclusion is false, we have thereby proved the argument invalid and have
provided a counterexample. Or, if the premises cannot all be true while the
conclusion is false, we have thereby proved the argument valid [as we just did
above].
Lets apply the short truth table method to the following argument:
pq
rs
pvr
qvs
The first order of business is to lay it out horizontally:
pq rs pvr / qvs
And now we do a kind of contra-Picard on the conclusion. Rather than Make
it so! we Make it not so! What does it take to make the conclusion, which is a
disjunction, false? Thats right, make both of its disjuncts false.
Important note: To apply the short truth table method, you must know
the truth-functional definitions of the connectivesknowing under what
254
conditions our basic molecular propositions are true and when they are
false should be effortless by now.
OK, heres where we are so far:
pq rs pvr / qvs
F
F
FFF
Now our project turns to that of making the premises all true, under these truthvalue assignments. Where should we begin?
Strategy suggestion: Begin where there has already been some work
done for youmake it easier on yourself. [Notice how much work it
would be to begin with the third premisep v r. There are three ways to
make that proposition true. So beginning there would almost be like
doing a regular truth table on the argument. Yuck!]
Lets begin with our first premise and then do the second premise. To get p q
true, p must be false. And to get r s true, r must be false. Lets enter those
truth-value assignments on our short truth table:
pq rs pvr / qvs
FTF FTF
FFF
But now, when we bring our truth values over to p v r, we see that it must be
false, since we made p false and r false.
pq rs pvr / qvs
FTF FTF FFF FFF
If we tried to make p v r true, for example, by making p true, look what happens:
pq rs pvr / qvs
TFF FTF TTF FFF
We see that we simply cant make all the premises true while the conclusion is
falsethe falsity of at least one premise is like a wrinkle in the carpetwhen we
stomp it out in the third premise, it just popped up in the first. Thats your proof
that the argument is valid!
The argument we just proved valid is a very popular one; its called the
Constructive Dilemma.
255
pq
rs
pvr
qvs
Be on the lookout for this important argument form. Sometimes the horns of
the dilemma are stated first, viz., the argument leads off with the disjunction.
Either I go to school full-time or I work full-time. But if I work
full time, Ill never get a better job. And if I go to school fulltime, Ill run up a lot of loans. So it looks as if Im screwed with
either a dead-end job or a bundle of debts.
Sometimes the horns of the dilemma can get extended by Hypothetical
Syllogisms. This is something newargument forms can join together to
build even more complex arguments. Heres an example:
pvr
pq
qt
rs
su
tvu
Arguments like this happens all the timeat least every Friday night for some of
us:
We can either make dinner or go out. If we go out, well just be
disappointed with the food, and that will spoil our evening. If
we make dinner, itll be better than restaurant swill, but well be
too pooped to enjoy it, and that will spoil our evening. So we are
doomed to have a lousy evening!
Heres yet another argument form for your listthis will be the last one, I
promise:
pq
rs
~q v ~s
~p v ~r
Let me illustrate its test for validity by means of the short truth tablesee if you
can describe the play-by-play of how I made my truth-value assignments:
256
pq rs ~qv~s / ~pv~r
TTT TTT FTFFT
FTFFT
This is called the Destructive Dilemma, and its a very popular valid argument
form.
If I take a winter vacation, I will have to skip preparing for
Spring Semester. But, if I take a summer vacation, Ill have to
skip prepping for Fall Semester. But Im going to have to prep
for at least one semester, so Im going to stay home either during
winter break or the summer.
Heres a little tip on how to remember the structures of these two dilemmas:
pq
rs
pvr
qvs
pq
rs
~q v ~s
~p v ~r
Constructive Dilemma
The disjunction of 2 cases of Affirming
the Antecedent
Destructive Dilemma
The disjunction of 2 cases of Denying
the Consequent
Often both horns of a dilemma can lead you to the same consequent, so that
youre damned if you do and damned if you dont. Heres an example:
If I dont drive into town tonight and return the video, Im out
three bucks. And if I dont stay home and send in my threedollars-off rebate coupon before it expires, Ill be out 3 bucks.
So.
FYI: When the disjunctive premise of a dilemma is known to be

false, the dilemma will not be cogent, even if its valid. We have
a popular phrase for pointing out such a caseslipping between
the horns of a dilemma.
257
Argument forms round up

pq
p
q
Affirming the Antecedent
Valid
pq
q
p
Affirming the Consequent
Invalid
pq
~p
~q
Denying the Antecedent
Invalid
pq
~q
~p
Denying the Consequent
Valid
pvq
p
~q
Affirming the Disjunct
Invalid
pvq
~p
q
Denying the Disjunct
Valid
pq
qr
pr
Hypothetical Syllogism
Valid
p v r
pq
rs
q v s
Constructive Dilemma
Valid
pq
r s
~q v ~s
~p v ~r
Destructive Dilemma
Valid
Here are all the argument forms weve introduced. You may want to make a
copy of this list, to help you get a good feel for the shapes of these arguments.
258
There are other argument forms, of course, but theyre either too trivial [e.g.,
Simplification: p q / p] or too rare [so just do a short truth table for them].
Testing for validity

You now have two methods at your disposal for testing the validity of
propositional arguments: Argument forms and the short truth table. Once you
get proficient at identifying argument forms, this is a very efficient method, but it
doesnt apply to all instances. So, if you cant analyze an argument by means of
argument forms, you always have the short truth table to fall back onit applies
to all propositional arguments.
Working with argument forms

Let me spend bit more time illustrating how to use argument forms. The key is
simply practice. Seeing argument forms is a very gestalt sort of thing.
Remember those pictures in your General Psych class? For example, that one of
all the black blotches that all of a sudden turned into a picture of a Dalmatian in
the street. The dog seemed to jump into view; and then, once you saw it, you
couldnt see it as just a random collection of blotches ever again. Well, thats
what will begin to happen more and more reliably with argument formstheyll
start jumping off the page at you, the more you practice looking for them.
If animals have sentience, they have some degree of moral
status. And animals certainly are sentient. Its obvious, then,
that they have moral status.
Here it is in standard prop form:
s ms
s
ms
This is an instance of Affirming the Antecedent, which is valid.
In order for animals to have sufficient moral status that would
require that we make significant sacrifices of liberty, they would
have to have rights. But it simply makes no sense to think, for
example, that guppies have rights on a moral par with yours or
mine. We are not, then, morally obligated to make significant
sacrifices for the benefit of animals.
259
sms r
~r
~sms
This is an instance of Denying the Consequent, which is valid too.
There are only two ferries per day to the island. And, since he
didnt arrive on the second, he must have taken the first.
fvs
~s
f
This is a case of Denying the Disjunct, which is also valid.
To pass this course, one needs to practice diligently. To keep my
job at the hotel, I need to increase my afternoon hours. But I
dont have time to both practice sufficiently and work enough
afternoon hours at the hotel. So somethings got to give!
pc pd
kj >h
~pd v ~>h
~pc v ~kj
This is a Destructive Dilemmavalid.
The rule doesnt hold true for us, because we made too much
money that year. It says, If your income was under $20,000 for
the year 2000, then if you incurred any moving expenses, they
are completely deductible.
<$20K (me d)
~<$20K
~(me d)
This is Denying the Antecedent. Its invalid.
There was a little something new in this last one: Molecular propositions can just
as well be used as the building blocks of argument forms.
Since a receipt is required for the rebate, he must have received
the rebate, because I was the checkout teller that night and I
remember handing him his receipt.
260
reb rec
rec
reb
This is Affirming the Consequent, which is invalid.
I know he didnt major in Philosophy, in view of the fact that
people in the Society & Law Program major in either Philosophy
or History, and I know he chose History.
pvh
h
~p
This is Affirming the Disjunctinvalid.
If someone wins arguments only by using informal fallacies, that
is intellectually dishonest. And that is sufficient reason to not
trust what such a person claims. Rush Limbaughs stock in trade
is winning arguments with informal fallacies. And thats why
his claims cant be trusted.
wif id
id ~t
wif
~t
Here we use two argument forms to get to the conclusion: Hypothetical
Syllogism and Affirming the Antecedent. First we use a Hypothetical Syllogism
to distill the first and the second premises to the subconclusion, wif ~t. And
then we add the remaining premise to draw the conclusion by means of
Affirming the Antecedent. Since both of these argument forms are valid, our
overall march to the conclusion is valid. If, however, any step we use en route
to the conclusion is invalid, the entire argument is invalid too.
Production at this plant will continue, but only if costs can be
cut. And that will be accomplished only with the cooperation of
the union. But the union has just promised to strike rather than
cooperate with the wage freeze. The plant closing is inevitable.
p cc
cc u
~u
~p
261
There are two ways of analyzing this argument by means of argument forms. We
could combine the first and the second premises and infer that p u, by means
of Hypothetical Syllogism. And then we would combine that with the remaining
premise, to infer our conclusion, by means of Denying the Consequent.
p cc
cc u
[p u]
~u
~p
HS
DC
Or, we could combine the third and the second premises and infer that ~cc, by
means of Denying the Consequent. And then combine that with the first premise,
to infer our conclusion, again by means of Denying the Consequent.
cc u
~u
[~cc]
p cc
~p
DC
DC
Both analyses prove that this argument is valid, since both of these sets of
argument forms are valid. But which analysis is most likely to be how the author
thinks that their premises act as reasons to believe the conclusion? In all likelihood, it is our first analysis that best captures how someone would infer the
conclusion on the basis of these premises. It would be easier to progress towards
the conclusion rather than work sort of backwards to it.
This is an extra aspect of analyzing an argument by means of argument
forms: You are not only trying to assess the arguments validity, but also
trying to capture as best you can the authors inference processtheir
mental steps to their conclusion.
Heres a worst case scenario for you:
He must be hiding in the basement. Hes either in the attic or the
bottom floors. But we checked the attic already and didnt find
him, so that leaves the bottom floors. If so, hed pick a level
where you couldnt see him through the windows. And that
would be the basement.
Here we have an argument consisting of three argument forms, all of which are
valid; so, the overall argument is valid. Heres how it goes:
262
a v bf
~a
bf
bf ~s
~s b
[bf b]
b
Denying the Disjunct
Hypothetical Syllogism
Affirming the Antecedent
OK, after that one, you should be able to handle pretty much anything life will
throw you by way of argument forms.
Working with short truth tables

Lets work through a few more examples of testing arguments by means of the
short truth table; there are still a few quirky things that crop up and some
strategies that you should be on the alert for.
Im not retiring this year, nor am I planning on retiring next year.
By not retiring, I continue to chair the department. And by not
planning to retire next year, I will be applying for sabbatical. So,
I am chairing, while at the same time submitting my sabbatical
proposal.
We could analyze this as a couple of instances of Affirming the Antecedent
just break the conjunction into its conjuncts, apply the argument forms, and
combine the entailed conjuncts back into a conjunction.
~r ~p
~r c
~p s
cs
But if you arent confident enough with you ability to apply the argument forms
in this slightly more adventurous fashion, just do a short truth table. Here is how
we start:
~r~p ~rc ~ps / cs
FFF
FFT
TFF
The first order of business is to make the conclusion false. But with a
conjunction, there are three ways of making it false! List all of them appropriately.
263
Next, we turn to the project of trying to make all the premises true, which would
prove the argument invalid. To find any such counterexample, we must investigate all three ways of making the conclusion false, if thats what it takes. All we
need to find in order to prove the argument invalid is one such counterexample.
Lets investigate the first way first:
~r~p ~rc ~ps /cs
FTFFT FTTF FTTF FFF
We couldnt make the first premise true, under this way of making the conclusion
false. Lets try the second way of making the conclusion false:
~r~p ~rc ~ps /cs
FTF
FTTF
T FFT
Again, we couldnt make the first premise truewe didnt even have to give ~p a
truth-value assignment to know this.
Lets investigate the third way of making the conclusion false:
~r~p ~rc ~ps /cs
FFT
T FTTF TFF
Once again we cant make the first premise true, and we didnt even have to give
~r a truth-value assignment to know this. We have now done an exhaustive
search and can find no way to make the premises all true while the conclusion is
falseand that, after all, is the claim of validity. So this proves that the
argument is valid.
Always remember, as you test arguments for validity using the short truth
table method: Your failure to make the premises all true is the arguments
success at validity; your success is the arguments failure.
Pop quiz:
1. In the middle of the last problem, we were investigating if we
could make the premises true when the conclusion was false in
this way:
~r~p ~rc ~ps /cs
F
T FFT
264
We began trying to make our premises true by making the

second premise true:
~r~p ~rc ~ps /cs
FTF
FTTF
T FFT
Question: Why not begin by making the third premise true?
Answer: There are two possible ways of making ~ p s true,
given that its consequent is already true. Wed be juggling two
possibilities then. Whereas, with the second premise, there is
only one way to make it true, given that its consequent is already
false. Moral of the story: Work where the work is easier, as you
assign truth values.
2. If you did a short truth table on an argument with

( p q r ) as its conclusion, how many possible ways are there
to make it false?
Answer: Seven. How do you arrive at this answer? So you
might as well use what, to do your proof?
3. Rank our connectives in descending order on the number of

ways to make them false. So, which propositions do you not
want to see in the conclusions of arguments youre testing, and
which do you not want to see in the premises?
Lets do a few more for the road:

You can indeed fly those dates, but only if you take a connecting flight
out of Detroit and you fly out of Midway. Since you have no problem
flying out of Midway, as opposed to OHare, it looks like those tickets
are yours!
f(cm) m / f
FT F F T T F
Weve shown that this argument is invalidwe were able to make the premises
both true after making the conclusion false. This is the first short truth table
weve done with a biconditional. Since f was already false, we needed to make c
false, in order to get ( c m ) false, to ultimately get f ( c m ) true. [Cripe,
truth tables take longer to describe than to do!]
265
When George goes Sherry goes too. But Sherry will go if and only if
Larry doesnt. And if Rose wont go, Larry will. To make matters
worse, either George is going or Rose isnt. So George is going, but just
in case Larry isnt.
GS S~L ~RL Gv~R / G~L
FFTF
TFFT
There are two ways of making the conclusion false; lets check the first way first:
FTT TTTF FTTF FFFT FFTF
Three out of four isnt good enough. Lets try the second possible way of
making the conclusion false:
TFF FTFT
T T
TFFT
And we see right away that we cant make the first two premises true, so we have
demonstrated that this is a valid argument.
Heres a last one, the likes of which you might find as you are trying to think
through the terms of a service agreement:
If the seal has not been broken and routine servicing has been performed,
the warranty is still in effect. The owner is responsible, however, for the
damage only if routine service was not performed or the seal was broken.
Therefore, even if routine service was not performed, the warranty is in
effect.
( ~ B R ) W D ( ~ R v B) / ~ R W
TFFF T F
TF
TFFF
FTFF T F
TF
TFFF
Notice how there are two ways of making the first premise true, given the single
way to make the conclusion false.
Now lets bring those possible truth values for B over to the second premise and
see what happens:
266
(~BR)W
TFFF T F
TFFF T F
FTFF T F
FTFF T F
D ( ~ R v B) / ~ R W
T T TFTF
TFFF
F T TFTF
TFFF
T T TFTT
TFFF
F T TFTT
TFFF
We find that under each of the two ways of making the first premise true, there
are two ways of making the second premise true. This makes for a grand total of
four counterexamples, thus demonstrating the invalidity of this argument. We
only had to find one of these, to prove our case against this argument. But by
finding all four, we have done a real service to the person using the argument,
thereby showing them the extent of circumstances in which these premises fail to
guarantee the truth of the conclusion. This is particularly important when youre
the user of the argument, wondering under which circumstances they can rely on
the warranty referred to in the conclusion and when they cant!
Pop quiz: See how many counterexamples can you find when
we change the conclusion to: Therefore, the warranty is in
effect, unless the owner is responsible for the damage.
( ~ B R ) W D ( ~ R v B) / ~ D W

For an overall review of the fundamental concepts of propositional logic, launch
Whad Ya Know, Level 4.
Youre also ready to hone and access your skills on Argument Forms and Short
Truth Tables.
Warning: Argument Forms is tough!
If youre having trouble identifying the argument forms in the passages, be sure
you are translating the arguments correctly from ordinary language to standard
propositional form.
You may need to go back to Conditionals and Standard Propositional Form to
brush up on those skills.
Use scratch paper, if it helps you keep track of the arguments or your truth-value
assignments.
267
Where do you go from here?

By now, youve acquired quite a few critical thinking skills. Its time to put them
all to good use by applying them to problems and issues. For a discussion of that
project, begin reading Self-Defense: A Student Guide to Writing Position
Papers. A complimentary copy came with this etext. The Self-Defense manual
will further enhance your abilities to select and argue for a position and critically
review competing positions and their arguments.
Because learning is life-long, here are some further suggestions:
We just finished learning how to assess the validity of deductive arguments using
both categorical logic and propositional logic. Armed with both of these formats,
you are prepared for almost any deductive argument life will throw at you. The
next most powerful means of analyzing deductive arguments is called predicate
logic or quantifier logic. It has notation and methods of assessment that
subsume and surpass categorical and prop logic. If youre interested in this sort
of thing, please consider studying symbolic logic.
We examined the nature of inductive arguments, but we only had time to study a
few specific aspects of assessing them for cogency. In the next edition of this
etext, I will remedy this by including numerous chapters dedicated to the ARG
conditions as they are applied to inductive arguments, especially scientific
arguments concerning theoretical hypotheses, statistical hypotheses, and causal
hypotheses. Until then, however, I highly recommend Ronald Gieres
Understanding Scientific Reasoning, as an excellent text on this area of critical
thinking.
Well, this is all for me for now. I hope Ive been of some help in building your
critical thinking skills, and I hope you find those skills both useful and personally
rewarding.
BTW: If youve got any suggestions, constructive suggestions, that is ;-) Id
love to hear from you. Send them, along with any examples of fallacies or other
topics weve discussed, to kpossin@winona.edu
Thanks!
268
Index
INDEX
Categorical Statements 153-66
Categories 153
Cogency
conditions of See ARG
conditions
definition of 36
Coherence Theory 91-2
Conceptual Analysis 55-9
Conclusion
definition of 6
implicit 27-8
indicators 9-10
Conditional 227-9
antecedent of 227
consequent of 227
necessary conditions of 228
sufficient conditions of 228
Conditional Argument 108
Conjunction 226
Conjuncts 226
Constructive Dilemma 254-7
Contradictories 171
Contrapositive 182-8
Contraries 189-92
Converse 174-80
Critical Thinking
definition of 3
elements of 3-4
goal of 2
study of 1-267 ;-)
Criticism, types of 48-50
A
Ad Hominem 131-37
Acceptability 32-3, 88-117
coherence theory of 91-2
foundationalist theory of 89-91
relativity of, 92-3
Accepting Premises 108-17
Affirming the Antecedent 246-7
Affirming the Consequent 247
Affirming the Disjunct 250
Ambiguity 70-7
Analogy 120-6
Appeal to Authority 94-8
Appeal to False Authority 96-8
Appeal to Force 148-9
Appeal to Ignorance 141-6
Appeal to Popularity 146-8
Appeal to Past Practices 148
Appeal to Precedent 148
Appeal to Tradition 147-8
ARG Conditions 37-8
See acceptability, relevance, and
groundedness
Argument
anatomy of 26-31
definition of 6
diagrams 29-30
explanation, differs from 19-22
identification of 6-25
subargument 26-7
test for 22-3
Argument Forms 245-63
Assertion 10-11
D
Deductive Argument 152-267
definition of 39
inductive argument, differs from
39-44
relevance condition 119-20
Definition
circular 60
counterexamples of 55
definition of 53
B
Bar, the 163
Barbara 213
Begging the Question 106-7
Biconditional 229, 237
C
Categorical Logic 152-221
269
Index
definiendum of 60
definiens of 60
necessary conditions of 55
persuasive 67-8
question begging 60
stipulative 64-7
sufficient conditions of 55
too broad 56
too narrow 56
virtues of 59-60
Denying the Antecedent 248
Denying the Consequent 248-9
Denying the Disjunct 249-50
Description 16
Destructive Dilemma 256-7
Dictionary Definitions 59-60, 63-4
Disjunction 227
exclusive and inclusive 240-1
Disjuncts 227
Dot, the 226
Double Negation 188
Grice, Paul 105

Groundedness
definition of 33, 152
See Categorical Logic and
Propositional Logic
Guilt by Association 137-40
H
Horseshoe, the 227
Hypothetical Syllogism 251-2
I
Immediate Inference 174-89
Implication, logical vs.
conversational 104-5
Impossibility, types of 34-5
Inductive Argument
analogy 125-6
deductive argument, differs
from 39-44
generalization 124-5
Informal Fallacies 72
abuse of language, involving
72-87
acceptability, involving 96-107
relevance, involving 128-49
list 150
Internal Negation 192
Is of identity 156
Is of predication 156
E
Enthymeme 214-21
Equivocation 72-5
Equivalent Categorical Statements
173-89
Equivalence 229-30
Euphemistic Language 69-70
Explanation
argument, differs from 19-22
types of 17-9
Existential Commitment 159-61,
204-7
External Negation 192
J
Jabberwocky 51
L
Loaded Language 68-9
Logical Connectives 224-31
Logical Entailment 34-5
mere intimation, differs from
104-5
Logical Impossibility 34-5
F
False Analogy 123-4
False Dichotomy 190
Foundationalism 89-91
G
Genetic Fallacy 133
Giere, Ronald 173, 268
Govier, Trudy 38, 44
M
Main Connective, the 242-4
Mere Intimation 104-5
270
Index
Modus Ponens 246

Modus Tollens 248
Standard Propositional Form 22531, 242-4

Stipulative Definition 64-6
Straw Man 128-31
Subcontraries 192
Subject 154
Subjective 12-3
Subsumption Argument 126-8
Syllogism
definition of 193
validity of 194-213
T
Tautology 242
Tilda 225
Triple Bar, the 229
Truth 231
Truth-functional Logic 231
Truth-functional Definitions 232-5
Truth Table
counterexamples 247, 254
define connectives 232-5, 241
number of rows in 237
prove contingent 241-2
prove equivalence 239-40
prove inconsistent 241-2
prove necessary 241-2
prove validity 244-52
Truth Values 232
Tu Quoque 133-5
N
Non sequitur 118
O
Objective 12-3
Obverse 180-2
P
Persuasive Definition 67-8
Poisoning the Well 133
Predicate 154
Premise
definition of 6
false 99
false assumption, involving
102-3
implicit 27-8
inconsistent 99-102
indicators 9
Preston, Ivan 104
Propositional Logic 222-67
Propositional Arguments 244-68
Proposition
atomic 224
definition of 224
molecular 224
R
Redefinist Fallacy 64
Red Herring 119
Reductio ad Absurdum 108-17
Relativism 12-5
Relevance 37-8, 118-51
Reverse the Argument 106
U
Use-Mention Distinction 61-2
V
Vagueness 75-87
Validity
definition of 35-6, 42-7
See Venn Diagram, Truth Table,
and Short Truth Table
Venn Diagram 166-221
S
Short Truth Table 252-7, 263-7
Slippery Slope Fallacy 79-87
Soundness 36-7
Square of Opposition 171
Standard Categorical Form 155-65
W
Wedge, the 227
Well-formed formula 242
271
C r i t i c a l Th i n k i n g & C T So f t w a r e
A completely computer-assisted, step-by-step guide to developing
your critical-thinking and logical-reasoning skills.
Whats covered in Critical Thinking?
Argument identification
Anatomy of an argument
Conditions for good arguments
Use and abuse of definitions
Use and abuse of language
Rhetorical tricks
Categorical arguments
Propositional arguments
Whats special about Critical Thinking?

Conversational, informal style
Lots of humor
Hands-on directions
Lots of examples
Its fun
Whats special about the CT Software?

Easy to use
Real life exercises
For practice and exams
Over 3,600 exercises
Attractive interface
Immediate feedback
For PCs and Macs
Installs in under a minute
No need to be online
Reliable
Kevin Possin is Professor of Philosophy and Director of the Critical

Thinking Lab at Winona State University.

Critical-Thinking-Lab.com

Critical Thinking

Uploaded by

Copyright:

Available Formats

Critical Thinking

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Critical Thinking

Uploaded by

Copyright:

Available Formats

What is the purpose of the document?

What is the purpose of the document?

What topics are covered in the document?

What topics are covered in the document?

CRITICAL THINKING

The Critical Thinking Lab

Library of Congress Control Number: NA

The Critical Thinking Lab

Thanks to Fred Dretske, for making me the roving teaching

Much that is taught in college classes grows soon out of date,

Our best guide to truth is free and rational inquiry; we should

. . . arguments, like men, are often pretenders.

4. What Makes an Argument Good? -

5. The Use and Abuse of Language

8. Groundedness I: Categorical Logic

9. Groundedness II: Propositional Logic

Times have changedluckily!

The CT software has thousands of exercises. So youll be demonstrating your

The goal of critical thinking

Something to think about: Why have both of these goals?

The elements of critical thinking

Identify Arguments: Identify arguments by their function.

Dissect Arguments: Identify and distinguish the parts of

Assess Arguments: Identify and assess the goodness, or

acceptability of their premises and 2) the relevance of their

Identify Fallacies: Identify types of fallacious arguments

Taxonomize Arguments: Identify and distinguish between

Critically Review Definitions: Analyze terms and concepts

Critically Defend a Position: Critically select and defend a

This isnt going to be a logic course, is it?

This isnt going to turn me into a machine, is it?

This isnt going to affect my social life, is it?

So, whats an argument?

A set of claims or statements {P}, called premises, that are

A set of statements {P} that is intended to be a reason to

A set of statements {P} that is intended to be why one ought

A set of statements {P} the truth of which is claimed to

Something to think about: What do the intentions of the

A set of claims or statements {P}, called premises, that

A set of statements {P} that actually is a reason to

By this time, you probably see what all-too-popular kind of

According to our analysis, then, an argument is an argument by virtue of the fact

Something to think about: Heres a popular definition of what

Can you think of A linguistic means of persuading

Can you think of an argument that is not A linguistic

Discussion: As examples of A linguistic means of persuading

unnecessary one, thats all. But it persuaded no one, because its

Something to think about: Whats wrong with the following

Consider, for example, Abortion is perfectly moral, or Abortion is murder.

But isnt everything just a mere assertion?

Something to think about: But isnt it true that everyone has a

Something else to think about: If relativism is as lame as

just as intolerant, if its their personal opinion that they need

A few other speech actsdescriptions and explanations

Pop quiz: If the author were to make a derivative point by

An explanation is a set of conditions, events, or states of affairs that (one claims)

Different types of explanations

Pop quiz: Answer: Our appendix and disco.

Distinguishing between explanations and arguments

Arguments and explanations are especially hard to distinguish, because they