Quick Guide to Nineteen Fallacies

1. Accident (Text 287)

The fallacy of applying a general rule to an
exception that the rule was not meant to cover.
It is wrong to deliberately hit someone in
the face. But those boxers are deliberately
hitting each other in the face. So what they
are doing is wrong.
This argument commits the fallacy of accident.
While it is (generally) true that deliberately
hitting someone in the face is wrong, this rule
is meant to cover general social behaviour and
not a competitive sport like boxing.
2. Argument Against the Person often
called Ad Hominem, which is Latin for
Against the Man. (Text 233-9)
The fallacy of attacking someones character
instead of his or her argument.
According to Bob, we should stop buying
sports shoes because poor people are
exploited in the factories that produce
them. Pooh, but what does Bob know? He
has no job and contributes nothing to the
economy. In fact, he spends all his time
drinking beer and gambling.
Bob provided a clear reason for shunning
sports shoes but his opponent just ignores the
reason and attacks Bobs character. What he
says about Bob may be true but it is not
relevant to Bobs argument.
Note that attacking someones character is not
fallacious if his or her character is relevant to
the discussion. See the text at 233-9 for many
examples of this. Also, the text distinguishes
three types of ad hominem attack:
a) Abusive
b) Circumstantial
c) You too
The example above was the abusive type but
you should know the other two types too. I can
do no better than direct you to the text for this
as it would take too long to spell them out
here. (See the text, 233-9.)
3. Appeal to Authority (Text 257-61)
Generally speaking, one makes an appeal to
authority whenever one claims that something
is true because an authority (or expert) said it

was true. Such appeals are not necessarily

fallacious: we make legitimate appeals to
authorities all the time, as when a student
contends that something is true because her
professor said it was true.
The fallacy is committed if and only if there is
something dubious about the authority in
question. For example, he/she is an authority in
a different area from the one in question:
Tiger Woods approves of Rolex watches.
So Rolex watches must be really good.
This argument is fallacious because Tiger
Woods is an authority on golf not timekeeping.
An appeal to authority is also fallacious if the
supposed authority cannot be identified:
A famous sociologist said that men are
inherently evil. So this is probably true.
The speaker cites a famous sociologist but
(let us suppose) cannot name the sociologist.
He says he cant remember who it was! So his
argument is a little dubious. A related example
is this:
Recent studies show that Vitamin C does
not prevent colds. So it must be true.
If the speaker cannot say what recent studies
are in question, or how we can check on them,
then this is equally fallacious.
An appeal to authority may also be fallacious
if there is disagreement among the authorities
on the topic in question:
J. M. Keynes says that the best way to
improve the economy is to inject lots of
money into the system. So this is what we
should do.
This looks like a legitimate appeal to authority
because Keynes is a distinguished economist.
But suppose that Milton Friedman (another
distinguished economist) disagrees strongly
with Keynes. Then appealing to Keynes and
conveniently ignoring Friedman would be
fallacious, being an instance of cherry
picking.
It is not always easy to decide whether a given
appeal to authority is legitimate or fallacious.
There is no hard and fast rule here. What
matters is to appreciate the general nature of
the fallacy since the practice of appealing to
authority is often abused.

4. Appeal to Ignorance (Text 244-7)

The fallacy of arguing that something is false
(or likely to be false) because there is no
evidence that it is true.

This is not fallacious. We can see that it is not

really a case of tugging at heartstrings. A
straightforward and relevant reason is being
given.
6. Appeal to Popularity (Text 229-33)

There is no evidence that the chemical

atrazine is toxic to humans. So we may
conclude that it is not toxic to humans.
This sort of reasoning is clearly dangerous. In
fact, the chemical atrazine has not been tested
on humans at all. So of course there is no
evidence of its toxicity, but only because no
tests have been carried out yet! So we really
know very little about atrazine and can hardly
conclude that it is not toxic to humans.
Suppose however that atrazine has been
extensively tested and no signs of its toxicity
have emerged. In that case, the argument
would not be fallacious. It would look
something like this:
Despite extensive and rigorous testing,
there is no evidence that the chemical
atrazine is toxic to humans. So we may
conclude that it is not toxic to humans.
The reasoning remains inductive, of course,
but it is stronger than the previous argument,
which was quite worthless.
So an appeal to ignorance is not necessarily
fallacious. It depends on whether attempts
were made to find out the truth of the matter.
As a rule of thumb, however, absence of
evidence is not evidence of absence.
5. Appeal to Pity (Text 227-9)
The fallacy of tugging at ones heartstrings to
persuade one of something.
Your Honour, please let me go free. I
have nine children and diabetes and my
wife will divorce me if I have a criminal
record.
This sort of plea speaks for itself. Such
reasons are generally not relevant to the
conclusion, except in exceptional cases where
the pitiful nature of the situation has a direct
bearing on the matter. For example:
Please excuse me from dance practice. I
fell down and injured my leg and I cant
even put my foot on the ground without
hurting badly.

The fallacy of assuming that something is true

just because many (or even most) people think
that it is true.
Many people believe that Vitamin C will
help cure a cold. So this is probably true.
Clearly, a popular opinion need not be true, or
even likely to be true. So anyone who invokes
popular opinion (and nothing else) in support
of their conclusion is committing the fallacy,
as in the example above.
Appealing to popular opinion is not always
fallacious though. If you are trying to find the
MRT station and many different people point
you in the same direction, its reasonable to
conclude that its the correct direction! So
whether popular opinion may be trusted
depends on the case. There is no hard and fast
rule for the fallacy: use commonsense.
The textbook (p. 232) also gives a different
kind of exception:
Most people agree that it is polite to say
thank you upon receiving a gift. So this is
really the polite thing to do.
This case is interesting because a certain
behaviour actually becomes polite if many
people think that it is polite. (That is how
politeness works.) So this argument is also not
fallacious. This case is unusual but it shows
that you should not be too mechanical when
it comes to fallacies. As a rule of thumb,
however, just because everyone thinks its
right doesnt mean its right.
7. Begging the Question also widely known
as Circular Reasoning. (Text 270-4)
The fallacy of arguing in a circle by smuggling
ones conclusion into ones premises. In other
words, the fallacy of taking for granted the
thing that one is trying to prove in ones
attempt to prove it.
The text gives three different ways of begging
the question but it helps to start with the third
sort of case (p. 273) which most clearly
illustrates the fallacy. Take this example:

God exists. How do we know? Because it

says so in the Bible. But how do we know
that the Bible can be trusted? Well, because
the Bible is the word of God.
The speaker is trying to prove that God exists
but one of his premises (The Bible is the word
of God) already presupposes that God exists!
So he has presupposed the very thing he was
trying to prove in his attempt to prove it.
This is often called reasoning in a circle, or
smuggling ones conclusion into ones
premises. Reasoning in a circle will clearly not
persuade anyone because it takes for granted
the very thing that was supposed to be proven.

presupposed in the premise, only in a rather

blatant way. Heres another example of
begging the question by restatement:
Society has an obligation to support the
needy because people who cannot provide
for themselves have a right to the resources
of the community.
Again, the premise is simply a restatement of
the conclusion and is thus not capable of
providing independent support for it. This is
why it is fallacious to beg the question.
Finally, the text has a case known as Begging
the Question by Omission of Premise. Heres
an example:

Heres another example:

Of course you and I are friends! Ive told
you this many times before and I would
never lie to a friend!
I hope its clear why this reasoning is circular.
It may be true that this person would never lie
to his friends but this presupposes that he and I
are friends, which was the very object of
contention. So he has presupposed what he
was trying to prove in his attempt to prove it.
(He has smuggled his conclusion into his
premises.)
Both of these examples are what the text calls
Begging the Question by Circular Chain. The
arguer employs a chain of reasoning, but
somewhere along the chain, a premise is
employed which presupposes the very
conclusion that was being argued for.
Heres a slightly different sort of circular
reasoning which is a little more blatant:
To get the best people you must pay the
best salary because the best people will
accept nothing less.
This is just a one-premise argument but notice
that the premise says exactly the same thing as
the conclusion, only in a different way! So this
argument is (really) no better than this one:
To get the best people you must pay the
best salary because you must pay the best
salary to get the best people.
This is sometimes called blatantly circular
reasoning. The text calls it Begging the
Question by Restatement because the premise
simply restates the conclusion in different
words. So, once again, the conclusion has been

Abortion is wrong because the murder of

an innocent person is wrong.
This argument contains a hidden premise
(Abortion is the murder of an innocent
person) which is just as controversial as the
conclusion. This is the key feature. The
reasoning is not strictly speaking circular but
it is close to being circular because hidden
premises are supposed to be obvious things
that anyone would accept, whereas, over here,
the hidden premise is almost as strong as the
conclusion. So the meat of the conclusion is
being presupposed in one of the premises (the
hidden one).
What happens if you make the hidden premise
explicit like this?:
Abortion is wrong because abortion is the
murder of an innocent person and the
murder of an innocent person is wrong.
Then it would not be fallacious because you
are placing all your cards on the table. It is the
act of hiding the controversial premise that
makes the argument fallacious (even devious)
because it makes the argument seem
straightforward when the hidden premise is
really doing all the hard work.
The overall lesson of this fallacy is to be
careful of arguing in a circle.
8. Complex Question (Text 275-6)
A complex question (generally speaking) is a
question which presupposes a certain answer
to a prior question. For example, if I ask (a)
whether you own a car, and (b) what colour it
is, my second question presupposes that you

answered yes to my first question. So (b) is an

example of a complex question.

questionable in this particular case. (We can

just point this out and leave it at that.)

The fallacy of complex question occurs when

someone asks you a question like (b) without
bothering first to ask you the prior question (a).
He simply presupposes that the answer to
question (a) has already been given. For
example:

10. Division (Text 286)

Have you stopped beating your wife?

Notice that whether a man answers yes or no,
the presupposition is very strong that he used
to beat his wife! So whichever answer he gives
(yes or no), his opponent can conclude that
he used to beat his wife. This is just a dirty
trick, of course. The correct way to answer a
complex question is to point out that it is a
complex question.
9. Composition (Text 284-5)
The fallacy of supposing that what is true of
the parts is also true of the whole.
Each stone in this pile of stones is light.
So the entire pile of stones is light.
In this example, the whole is the pile of
stones and the parts are the individual stones
in the pile. Obviously the conclusion does not
follow from the premise: just because each
stone is light, it does not follow that the whole
pile of stones is light. What is true of the parts
need not be true of the whole.
Why would anyone fall into this fallacy? Well
probably because (like many other fallacies)
the act of composition is not always
fallacious. For example:
Each tile in this floor is white. So the
entire floor is white.
The inference from premise to conclusion is
fine in this case. So it does depend on the case.
Sometimes it is hard to tell if it is safe to
compose. For example:
Every politician in the cabinet is a born
leader. So the cabinet (as a whole) will lead
our nation very well.

The fallacy of supposing that what is true of

the whole is also true of the parts. (This is the
reverse of composition.)
This bag of sand is heavy. So each grain
of sand in the bag is heavy.
Clearly, the conclusion does not follow from
the premise: just because the entire bag of sand
is heavy, it does not follow that each grain of
sand in the bag is heavy. What is true of the
whole need not be true of the parts.
As before, division is not always fallacious.
The following inference is fine:
This whole machine is made of metal. So
each part is made of metal.
And then (just like composition) there are
troublesome cases like this one:
That mob of people is violent. So each
person in the mob is violent.
This inference is very difficult to evaluate.
While the mob (as a whole) may be violent,
each person in the mob may not be particularly
violent. It is hard to say.
Another case is this:
Sodium chloride is edible. So each of
sodium and chlorine must be edible.
If you knew little chemistry, you might be
tempted to make this inference, but it would be
quite fallacious. Sodium chloride is just normal
salt and thus safe to eat. But sodium and
chlorine are (individually) very poisonous
substances.
The point is that neither composition nor
division are safe methods of inference. It really
depends on the case and if we are unsure in a
given case if the conclusion may be drawn we
may just point it out and leave it at that.
11. Equivocation (Text 281-2)

This example illustrates the danger of

composition. What if these born leaders fail
to get along with each other? They might end
up arguing all day and the cabinet will not
function very well. So the inference is

The fallacy of using the same word or phrase

in two different ways (at different points of
ones argument) while pretending that it is
being used in the same way:

SMU has a scholarship for poor students.

My professor said that I am a poor student.
So I should get the SMU scholarship.
This argument looks valid until we realize that
in the first premise, poor student means
student with little or no money, whereas in
the second premise, it means student with bad
grades. So the appearance of validity is bogus.
(To equivocate is to use ambiguous language
with the intent to deceive.)

criminals. So the use of tasers is causing

criminals to scale back on their activities.
In this case, it seems very natural to draw the
conclusion but, as our first example shows, we
should be careful. In the case of the cock, it is
really the suns rising that causes the cock to
crow (the causation goes the other way), but
other possibilities exist too. For example, in
the taser case, it may just be a coincidence that
the crime rate dropped soon after the police
started using tasers. It is really hard to tell.

Another example is this:

Pornography is of great public interest.
But the government should always promote
the public interest. So the government
should promote pornography.
This argument works only because the
phrase public interest is used in two different
ways. In the first premise, it means what the
public finds interesting, whereas in the second
premise, it means what is best for the public.
Once this is made clear, the appearance of
validity disappears.
An example with wide currency is this:
Evolution is just a theory whereas schools
should be teaching facts, not theories. So
evolution should not be taught in schools.
This argument equivocates on the word
theory. The word theory sometimes means
guess (or hunch), as when the police consider
various theories of who the murderer might be.
But, in science, the word theory means
something well-established by the evidence, as
in the theory of gravity. Evolution is a theory
in this second sense, not in the first sense, and
the argument only works because it slides
between the two meanings.

There are indeed many ways in which we

might mistakenly infer a causal relation when
none exists and I refer you to the text for a full
discussion. (See text, 261-5.)
Note that the text distinguishes three different
fallacies under this heading, namely:
a) After this, so because of this
b) With this, so because of this
c) Mistaken cause
The crowing cock and taser examples fall in
the first category and the second category is
very similar. (The umbrella term False
Cause refers to any of the three.)
The third category of Mistaken Cause is
especially interesting. It concerns cases where
it is all too easy to misidentify what is causing
what. For example:
Statistics show that people who eat lots of
fish have higher I.Q.s. So eat lots of fish if
you want to increase your I.Q.
The inference here is very suspect! It may be
that eating fish causes you to have a higher
I.Q.:
Eat Fish

Higher I.Q.

12. False Cause (Text 261-5)

The fallacy of concluding that one event was
the cause of another event just because both
events happened to coincide (or occur close
together) in time:

The arrow denotes cause and effect. But it may

also be that a third factor (like having a special
gene) causes certain people to enjoy eating fish
and (independently!) to have higher I.Q.s:
Gene

The sun rose soon after the cock crowed.

So the crow of the cock caused the sun to
rise.
This example is clearly a joke but consider
another argument of the very same form:
The crime rate dropped soon after the
police started using tasers to apprehend

Eat Fish

Higher I.Q.

If so, then if you dont already have the gene,

eating fish is not going to increase your I.Q.!
Given these two possibilities, it would require
more investigation to find out what is really
causing what here and we shouldnt jump to

conclusions. (This example is fictitious by the

way.) The fallacy of False Cause, in whichever
of its three variations, is captured in the useful
slogan, correlation does not imply causation.
13. False Dichotomy (Text 276-80)
This refers to two related fallacies which turn
on a careless use of Either or . For
example, if you argue this way:
Either James is working or hes at home.
But James is at home. So hes not
working.
then this is careless because James might work
at home! As noted in class, this form of
argument:

For example:
Either the state controls our industries or
we let them roam free of state control. If
the former, then creativity will be stifled.
But if the latter, then consumers will be
exploited, leading to a backlash. So either
creativity will be stifled or there will be a
consumer backlash.
Here, the Either or premise is again
false, since a third alternative of partial state
control has been overlooked. So this is another
case of false dichotomy by overlooking
alternatives. Some people also call this a False
Dilemma because the argument (in this case)
has the form of a dilemma.
14. Hasty Generalization

Either p or q
p
> Not q
is invalid if an inclusive disjunction is at
issue, which is the case here. So the argument
is just invalid. But we can call this a case of
False Dichotomy by Overlapping Alternatives,
as the text does. (The alternatives of being at
home and at work overlap: could both be true.)
On the other hand, if you argue this way:
Either you like her or you hate her. You
clearly dont hate her, so you must like
her.
then this is careless in a different way. This
time, the form of argument is valid, being a
case of disjunctive syllogism:

The fallacy of generalizing from a small (or

unrepresentative) sample to the population at
large.
My whole family loves seafood. So
everyone else must love seafood too!
Clearly, just because her whole family loves
seafood, it doesnt follow that everyone else
loves seafood. It doesnt even follow that many
people love seafood!
Note that a hasty generalization must involve a
generalization. In the example given, someone
is generalizing from her family members to
everyone as a whole. (If no generalization is
going on, no hasty generalization can be going
on either.)
Another example is this:

Either p or q
Not q
>p
The problem (rather) is that the Either or
premise is false because it does not cover
all possible alternatives. A third alternative has
been overlooked of your being affectionately
neutral towards her. So this argument just has
a false premise. But we can call this a case of
False Dichotomy by Overlooking Alternatives,
along with the text.
Note that this second type of error also
commonly occurs as part of a dilemma:
Either p or q
If p then r
If q then s
> Either r or s

I vacationed in Indonesia once and had a

horrible experience. The people were rude
and the service was really bad. Indonesia is
just a terrible place.
Clearly one cant reasonably generalize from
this one experience to the stated conclusion.
Indonesia is very large for one thing and this
person only vacationed there once. The
generalization is obviously hasty.
Not all generalizations are bad or hasty, of
course. A generalization is legitimate if the
sample is representative of the population. The
fallacy of hasty generalization is committed if
and only if the sample is unsatisfactory in
some way, e.g., the sample is too small or not
selected at random. This generalization, for
example, is quite legitimate:

800 SMU students were selected at

random and most said they were enjoying
their university experience. So SMU
students in general enjoy their university
experience.
Admittedly though, some knowledge of
statistics may be required to determine if a
sample is really good enough e.g., is 800 a
sufficiently large sample? but we will rely on
commonsense for our purposes. If you are
unsure whether the sample is good enough to
fairly represent the population, just say so.
This fallacy is covered in some detail in the
text under the headings, Biased Sample and
Small Sample. (See text, 251-6.)
15. Slippery Slope Causal (Text 265-6)
The arguer claims that a chain of events will
occur, as though out of necessity. But the sheer
number of links in the chain makes it unlikely
that the entire chain will occur. Also, the
weakness of some of the links may easily
prevent the entire chain from occurring:
We should not allow our son to get a
tattoo. If he gets a tattoo he will soon start
smoking. If he smokes, hell start hanging
out with bad hats. And if he starts hanging
out with bad hats, hell soon graduate to
crime. And then he will start committing
murders, and then he will end up being
hung on death row.
This example is characteristically dramatic, but
note that the validity of the argument is not at
issue, since the following form of argument is
perfectly valid (its just an extended version of
hypothetical syllogism):
If p then q
If q then r
If r then s
If s then t
> If p then t
The problem (rather) is that, very typically,
some of the premises are not clearly true. On
top of that, even if each premise is individually
plausible, the sheer number of premises makes
it unlikely that all of them are true. So this sort
of argument has little chance of being sound.
16. Slippery Slope Semantic (Text 287-8)
An arguer claims that we cannot pick out a
specific point at which a thing changes from

being an F to being a G. So he concludes that

it never becomes a G.
A man whose height is 130 cm is short.
But if you increase a short mans height by
one cm, he will still be short. So a man
whose height is 131 cm is also short.
Likewise, a man whose height is 132 cm is
also short. And so on, without end. So all
men are short, really. There is no such
thing as a tall man.
This argument looks comical and clearly does
not work but people engage in such fallacious
reasoning all the time. In the debate on
abortion, for example, some people invariably
reason like this:
A baby is clearly human 180 days into a
mothers pregnancy and you obviously
cant abort it then. But one day cant make
a difference: the baby doesnt turn human
overnight! So you cant abort a baby 179
days into the pregnancy either: the baby is
still human then. And so on, without end.
So the baby is really human even on the
very first day of conception.
All such arguments have the same discernible
structure and they are generally unreliable, no
matter what the subject matter. They can look
very convincing though!
When people speak of a slippery slope
argument, they usually have this sort of
argument in mind. The qualification
semantic refers to the fact that the key term
(short, human) does not have a sharp
meaning. You cant specify the exact height
when a short man ceases to be short: there is
no such thing. Nor can you specify the exact
day when the foetus becomes a human being:
there is no such day. (Its a slippery slope.) But
it doesnt follow that all men are short, or that
the foetus is human right from the start.
If someone unleashes a slippery slope
argument on you, the correct response is to
point out that slippery slope arguments are
fallacious and to show them the first example
shown above.
17. Straw Man (Text 240-1)
The fallacy of pretending to refute a persons
argument by refuting an argument that is a
distortion (usually a caricature) of her
argument. Note that two people are always
involved in this fallacy:

Jones: It would be really nice to have a

submarine defence for our navy but the
costs of procuring these submarines are just
too high. Given our budget, we must find
other ways to secure our shores.

missing the point. However, in this course,

the term missing the point should be
reserved for the fallacy of irrelevant
conclusion, as illustrated above.
19. Red Herring (Text 242-3)

Smith: So you are saying that we should

leave our waters defenseless just like that.
Im sorry but I cannot accept that kind of
argument.
Here, Smith has distorted Joness argument
into a form that is especially vulnerable to
attack. According to Smith, Joness plan is to
leave our waters defenseless, which would
naturally be a terrible plan. But nothing in
what Jones said carried that implication and
Smith surely knows this. So Smith is just
attacking a straw man. (A man made of straw:
very easy to knock down.)
See the other examples in the text, p. 240.

The fallacy of promising to argue for a certain

conclusion but then wandering off to establish
some other (typically irrelevant) conclusion.
There is talk these days of cutting down
on pesticides used in growing foods and
vegetables. This is not right and let me
explain why. Look, we know that these
foods are essential to our health. Fruits and
vegetables are sources of vitamins and
minerals and we cannot do without them.
Of course, these items must be fresh, clean
and green and therefore we are making it a
priority to monitor supermarkets all over
the country to ensure that healthy stocks of
these produce are available at all times.

18. Missing the Point (Not in the text)

The fallacy of drawing an irrelevant
conclusion from a given body of premises
when most people can see what the correct
conclusion should be.
Crime is increasing and the public is
frightened. The police have done little or
nothing about this. So we must bring back
the death penalty for criminals.
Clearly, the conclusion is unwarranted given
the facts of the premises. A more natural
conclusion would be to beef up the police
presence or to fund a study to understand why
crime is increasing.
Heres another example:
The tuition situation is getting out of
hand. Students are forced to attend tuition
classes because they are failing to learn in
school. Many teachers also moonlight as
tuition teachers to make extra income. So
we should create a body to regulate tutors
to ensure that the quality of tuition is high.

The arguer starts off promising to explain why

pesticides are not a problem but wanders off
quickly and never addresses the issue.
A red herring is a certain sort of dried fish
with a strong smell that is used to throw
hunting dogs off the correct trail. Someone
with no good argument for their position will
often try to shake an audience off the real issue
by diverting them gradually into another one.
Any such maneouvre may be called a red
herring.

See also the Quick Summary of the fallacies in

the text at 289-96. Many more things may be
said about these fallacies but the general idea
is to increase ones appreciation of common
pitfalls in reasoning. All these fallacies are
here because they are commonly committed by
the man in the street.

The conclusion drawn at the end (We should

create a body to regulate tutors ) clearly
misses the point. The point was that we should
find a way to improve the education in schools
so that tuition becomes ultimately unnecessary.
Note that the term missing the point has a
broad use on which even Straw Man (above)
and Red Herring (below) count as fallacies of