MORE EFFECTIVE REASONING II: BETTER LINKS 85groups would need to be covered. By contrast, in a report on the legal aspects of
European-Aboriginal relations, the audience's expectations would be narrower: the
context of the report ('legal aspects') would exclude other reasons, which if we were
to introduce them, might actually weaken our arguments because they would be
irrelevant to the particular issue being reported on.
In summary, not only do we need to understand issues well, but we must also
understand our audience and other contextual factors so that we can judge what
should or should not be included in any argument or explanation.
Exercise 6.6
Take any argument or explanation that you are writing at the moment or have
recently written. Begin by establishing clearly in your own mind the context for
your work, including its audience, and the sorts of constraints or requirements
that the context places on you. Step by step, apply to it all of the issues discussed
in this section, with the aim of improving it.
Coherence in scope and certainty
Finally, we must consider the relationship between what we are claiming as our
conclusion and the evidence used to support it, as expressed through the scope and
certainty aspects of the claims. If the premises and conclusion are coherent in this
respect, then our reasoning is more effective. Coherence of scope, while always
important, is particularly significant in reasoning from specific cases. Here is an
example:
John has met a few Aborigines who are alcoholics, and therefore he
concludes that all Aboriginal people are alcoholics.
The error John makes here is that the scope of his premise ('a few') is not
coherent with the scope of the conclusion ('all'). Hence he has overgeneralised in
his conclusion. Similarly, if John was to visit one Aboriginal community in
which, say, a third of its members were alcoholics, he would also be wrong to
conclude that A third of all Aboriginal people are alcoholics'. The scope of his
premises (just one community) is not coherent with the conclusion about all
Aborigines, since that community is most unlikely to be a representative sample
of the entire Aboriginal population. However, if John were to continue his
investigations and discover that, say, 70 per cent of Aboriginal people in outback
areas suffer from poor health, he would be equally in error to conclude that 'Poor
health is, thus, a small problem for outback Aboriginals'. Such a conclusion
understates the extent of the situation and again reflects a lack of coherence
between premise and conclusion. General conclusions are not, of themselves, the
problem: we could not think and know without reasonable generalisations.
Rather, we must always be sure that the generalisations are properly grounded in
the specific cases on which they rely.