EVALUATING SUPPORT FOR CONCLUSIONS

You have already had some practice in judging whether a conclusion follows from, or is
supported by, a given reason. This was what Exercise 3 involved, since you were asked to
pick out from three statements the one which could be a reason for the conclusion. When
trying to decide whether conclusions of arguments are established by the reasons
presented, you are essentially doing the same thing as you did for Exercise 3, but you may
have to take into account more than one reason. You may also have to assess a chain of
reasoning, which could involve judging whether an intermediate conclusion follows from
some basic reasons, and also whether it in turn supports a main conclusion.
A reason will not support a conclusion if it is not relevant to the conclusion. This may
seem very obvious, since if a reason is concerned with some topic completely unrelated
to the subject matter of the conclusion, it would be clearly mistaken to think that the
reason could support the conclusion. However, when we talk about a reason being
relevant to the conclusion, we do not simply mean that it is about the same topic. What
we mean is that the reason, if true, makes a difference to the acceptability of the conclusion.
Relevance in this sense does not necessarily mean that a relevant statement supports
a conclusion. A statement could be relevant and yet count against the conclusion. If we
look again at one of the questions from Exercise 3 on p.18, we can see an example of
this:
Conclusion: Blood donors should be paid for giving blood.
Which of the following, if true, could be a reason for the above conclusion?
(a) The Blood Donor service is expensive to administer.
(b) People who give blood usually do so because they want to help others.
(c) There is a shortage of blood donors, and payment would encourage more
people to become donors.
The correct answer to this question is (c), which supports the conclusion by showing that
if payment were offered to blood donors, this could remedy the shortage of donors. But (a)
is also relevant to the conclusion, in the sense that it has some bearing on the recommendation
to pay blood donors. If the Blood Donor service is already expensive to administer,
then this may be a reason for rejecting the recommendation. Hence (a) does not support
the conclusion; it counts against it.
You may find it useful to think about whether reasons are relevant, because if you can
quickly spot that a reason is irrelevant, then you will know that it does not support
the conclusion. However, the above example shows that the judgement that a reason is
relevant is not sufficient to tell you that the reason supports the conclusion. You will still
have to think about the way in which it has a bearing on the conclusion.
The strength of support which reasons provide for a conclusion can vary. In the argument
on p.36, for example, the reason gives the strongest possible support to the conclusion.
The argument says:
Everyone who exercises regularly in the gym has well-developed muscles. So if Mel
doesn’t have well-developed muscles, it can’t be true that she’s exercising regularly in
the gym.
In this case, if the reason is true, the conclusion must be true. Other arguments may
provide less strong support, and nevertheless be good arguments. We can have good reason
for believing that something will happen in the future based on evidence from the past, or
for believing that what is known to be true of a number of cases will be true of another
similar case. For example, we could have good reason to believe that a new car will be
reliable, based on the knowledge that most other cars of that model have been reliable. It is
not possible to be precise about degrees of strength of support, and in many cases we may
need to find out more about the context of an argument in order to assess whether the
reasons give strong, fairly strong or only weak support for the conclusion.
In addition to differences in the strength of arguments, there are also different ways in
which reasons can support their conclusions; arguments may present past experience
as evidence for their conclusion, they may use analogies (i.e. draw their conclusions on
the basis of what is true of similar cases), they may refer to statistics, or to results of
experiments in science or studies in psychology or sociology, they may base their conclusions
on general principles. In relation to all these kinds of reasons, it is useful to ask
yourself the questions in the following summary.
Summary: Evaluating support for conclusions
1 Are the reasons/evidence relevant to the conclusion?
2 If so, do the reasons/evidence provide a good basis for accepting the
conclusion?
3 If the conclusion recommends some action or policy, would it be reasonable
to act on the basis of the reasons? In order to answer this question, you will need
to consider the following points.
• Would the recommended policy or action be likely to achieve the desired aim?
• Would it have some undesirable effects?
• Are there other, possibly better, ways of achieving the aim?
4 Can I think of any other evidence, not mentioned in the argument, which
would weaken or strengthen the conclusion?


Let’s put this into practice with a few examples. Consider the following argument:
You ought to take a Happitum travel sickness pill when you go on the ferry. They
are very effective against sea-sickness, and you have always been sick in the past
when you’ve travelled by sea.
In this example, it is easy to see that the reasons, if true, give fairly strong support to the
conclusion. If you have always been sick on sea crossings, then past experience suggests
that you are likely to be sick this time, unless you can prevent this, perhaps by taking some
effective drug. So it would be reasonable to act on the evidence that Happitum is effective
in preventing sea-sickness. Of course, there may be other considerations, not mentioned in
the argument, which would count against the conclusion. If, for example, Happitum had
serious side-effects, then it may be more sensible to endure sea-sickness rather than risk
ill-health from the drug. Or maybe there are techniques for combating sea-sickness (for
example, staying on deck and breathing deeply), which are likely to be effective, and which
are less unpleasant than taking a drug.
Here is another example:
New drugs have been developed which can combat the body’s tendency to reject
transplanted organs. In the past, most of the deaths which have occurred shortly
after heart transplant operations have been due to rejection. So it is likely that these
new drugs will improve the survival rate of heart transplant patients.
Are the reasons relevant to the conclusion? Yes, since if most deaths of heart transplant
patients have been caused by organ rejection then the use of drugs which counteract
rejection is likely to enable some patients to survive who would have died without the
drugs. The reasons are not only relevant to the conclusion, they give it strong support since
if some patients survive who would otherwise have died, this means that the survival rate
is higher. There may, of course, be evidence not presented here which would count against
the conclusion, for example, if the drugs were highly toxic. But on the assumption that
the drugs have been tested for toxicity, and found to be relatively safe, we can regard the
conclusion as well supported by the reasons.
Let’s look at one more example:
We could introduce a much more difficult written test for learner drivers in the UK
but, since this would not improve their driving skills, it would not result in a lower
accident rate amongst young drivers. In Portugal, learner drivers must have five
weeks’ theoretical instruction and a stiff examination before they are legally entitled
to touch the wheel, but this does not result in a low accident rate amongst young
newly qualified drivers. They soon forget about the test once they start to drive.
All it indicates is that candidates can read and write. It has no bearing on their ability
to drive.
This argument uses evidence from Portugal in order to draw a conclusion about what
would be likely to happen in the UK. Its major reasons:
In Portugal, learner drivers must have five weeks’ theoretical instruction and a stiff
examination before they are legally entitled to touch the wheel, but this does not
result in a low accident rate amongst new drivers.
and
[the test] has no bearing on their ability to drive
are offered in support of an intermediate conclusion that:
[Introducing] a much more difficult written test for learner drivers in the UK . . .
would not improve their driving skills
which in turn is offered to support the main conclusion that:
[Introducing] a much more difficult written test for learner drivers in the UK . . .
would not result in a lower accident rate amongst young drivers.
We need to ask first whether the reasons are relevant to the conclusion. Remember that we
are not questioning the truth of the reasons at this stage. We are considering whether,
assuming the reasons to be true, they support the conclusion.
So, if it’s true that the stiff written examination in Portugal does not produce a low
accident rate amongst new drivers, and that it has no bearing on driving ability, is this
relevant to the claim that such an examination in the UK would have no impact on the
accident rate amongst drivers aged 17 to 21? Well it certainly is a piece of evidence which
is worth taking into account, since it is one example of a test which has not had the result
which is perhaps hoped for in the UK. But when we consider whether the evidence gives
us sufficient basis for accepting and acting upon the conclusion, a number of further
questions come to mind. Is there any evidence from other countries besides Portugal? Are
the accidents in this age group (both in the UK and in Portugal) attributable mainly to the
driver’s lack of skill, or perhaps to the driver’s reckless attitude? Are there any cultural
differences which might give a test greater impact on attitudes amongst young drivers in
the UK than it has amongst their counterparts in Portugal? There is insufficient evidence
in this argument to give very strong support to the conclusion.