Thinking Skills: Critical Thinking and Problem Solving

2.10 Flaws and fallacies 75

[7] I knew the Bayside was bad news. I’ve

never liked the food there, and certainly

never eaten the fish. Now we hear that

four people who went there have all

reported sick, and the next day the

restaurant is closed. So, it’s pretty clear

that their food is to blame. My

suspicions were correct all along.

A classic fallacy

A fallacy, you will recall, is a flawed

argument. It is also the word we use for the

flaw itself. We can say that [7] is a fallacy,

because it is a flawed argument. But we can

also say that it commits a fallacy, or has a

fallacy in it. Some fallacies appear over and

over again in different arguments. The

best-known examples were discovered and

classified centuries ago, and many have Latin

names. They are often referred to as the

classic fallacies, for that reason.

There is a classic fallacy lurking in [7], and

in the three inferences from [6] that we

discussed. It is known as the post hoc fallacy, or

in full: post hoc ergo propter hoc, meaning

literally: ‘after this, therefore because of this’.

The fallacy is in assuming that when one thing

happens and then another, that the first must

be the cause of, or reason for, the second. The

absurdity of this assumption can be illustrated

if we imagine someone opening an umbrella

just before it starts to rain, and arguing that

opening the umbrella made it rain! Of course,

there are many situations in which one act or

event does cause another. If a tree falls into the

road and a driver swerves to miss it, it is

perfectly reasonable to infer that the falling

tree caused the driver to swerve. The fallacy is

not that there is never a causal connection

between two events, but that a causal

explanation cannot and should not be

assumed, even when it looks quite plausible.

Indeed, it is when a causal explanation looks

quite plausible that the fallacy is most

dangerous, because it is then that people are

the Bayside or somewhere else. The facts that
we have, even if true, do not support the
inference.
The same goes for inference B, that fish
caused the sickness. We are told that all four
of those who ate at the Bayside and reported
sick had eaten fish. But we are not told what
else they may have eaten or drunk. Even if the
Bayside was the source of the illness, which is
not certain, it need not have been the fish
that caused it. The cause may have been a side
dish, or a sauce, or contaminated water, or a
general lack of hygiene in the kitchen.
Nor is it safe to infer C, that the inspectors
closed the restaurant. The statement in the
newspaper that restaurants found responsible
for food-related sickness have to close is
actually irrelevant: it does not mean that
because a restaurant closes it is responsible for
the sickness. Many restaurants close on one or
more days of the week. Today may be the chef’s
day off. Many explanations for the closure are
possible besides the seemingly obvious one, that
it was closed because of food poisoning.

Jumping to conclusions
Often when people read of incidents like this
they infer too much, given what they know –
or rather, despite what they don’t know.
Without more than the information in the
report, it would be jumping to a conclusion to
draw any of the three proposed inferences
about the restaurant, its food, or the reasons
for its closure.
It is particularly tempting to jump to a
conclusion if you carry some prejudice in the
matter. Suppose, for example, you had eaten a
couple of times at the Bayside and had not
enjoyed the experience. Perhaps one of the
waiters had been rude, or the service had been
slow; or you just don’t like fish. In other words,
you had reasons to be critical of the restaurant,
but ‘reasons’ in the sense of motives rather
than reasons for a sound argument. With that
motivation, you argue as follows: