4.1 Inference 133
putting in false claims, and that would
explain the rise in the number of claims
generally.
These are plausible hypotheses, but they are
poorly supported by the data in the charts. The
fact that one thing would explain another if it
were true, does not not permit us to infer that
it is true. In some circumstances this can be a
powerful argument; but as you discovered in
Chapter 2.10, it can also be a dangerous one.
(Remember the Bayside fish restaurant. Just
because food poisoning would explain why a
restaurant has closed, it does not follow that
food poisoning occurred or came from the
restaurant.) In the present case, just because
cheating would explain rising claims, it does
not mean there is widespread cheating. There
are many other equally plausible reasons why
claims could be increasing in frequency, if
they are.3 Does the data in Chart 3 contradict the
data in Chart 2?Activity
Commentary
A comparison of Charts 2 and 3 is very
interesting. According to Chart 2, most people
evidently believe that there is an increased level
of dishonest claiming going on, and half of
those questioned believe that there is a big
increase. But if Chart 3 is anything to go by, very
few people say that they would so much as
exaggerate a claim. Even those who watch
daytime TV (which, we were told, carries a lot of
advertising by the so-called ‘ambulance
chasers’), or who have seen a claims
advertisement recently, say they are no more
likely to claim than the general sample of the
population. Remember, too, that being tempted
to do something and actually doing it are two
different things. Of the 13% of all adults who
said they might be tempted at all, how many
would have gone as far as making a false claim?there really are more payments being made;
or, in other words, that the widespread belief
expressed by those questioned is correct. It is
this step which is the problem: reasoning from
the evidence that most people believe
something, to the conclusion that it is true or
probable, is a classic fallacy known by the
Latin argumentum ad populum. If you prefer
more modern names there are plenty to
choose from: appeal to popular opinion,
appeal to consensus, appeal to the majority,
the authority of the many over the few. The
weakness of this argument is nicely captured
by the old joke that 40,000 lemmings can’t all
be wrong. (The joke is that every so often
whole colonies of lemmings are believed to
run to the edge of the nearest cliff and plunge
to their deaths!) C therefore is not a safe
inference.
2 Suppose the majority view represented
in Chart 1 is correct. Would it follow that
there has been a change in the number of
people making false claims?Activity
Commentary
This is another complex question. What
makes it so is that it is hypothetical. We don’t
know whether the opinions represented in
Chart 1 are true or not. The question is: If the
sample of public opinion is right and there has
been a big increase in claims, can we infer that
a significant number of people are making
false claims?
Why might this be true? Well, if it is
correct, as it is widely believed, that there are
more people getting money for injuries,
others may see this as a way of getting some
money themselves. It is a sad fact that there
are dishonest people who will seize such
opportunities. Then again, it might be the
other way round: that with the help of no-win
no-fee agreements more people have begun