WHAT KINDS OF REASONING ARE THERE? 95
Jones and Wilson have both been sacked by the company that used to
employ them. Jones, a middle-aged male cleaner, was a poor worker with
a history of arriving late; Wilson, a young woman working as a filing clerk,
had always been judged by her boss to be a competent worker. But both
Jones and Wilson had just been elected union delegates, and thus, I would
reasonably conclude that it was their active participation in the union that
led to their dismissal.
The differences between the two workers contrast with their similar treatment
(both were sacked); so, the only other similarity between them (that they were
union delegates) seems to be the likely cause of their dismissal. We can express this
rule thus:
X is the cause of Y because it is the only relevant similarity between two
separate occurrences of Y happening.
The other key aspect to causal reasoning is to appreciate that some causal
events are necessary, and others are sufficient. If I assert that oxygen is necessary
for a fire to occur, I am not saying that oxygen causes fires. Thus the presence
of oxygen is a necessary condition, but not sufficient. However, while a lighted
match is sufficient to set fire to a pile of dry paper if dropped on it, it is not
necessary — I could also use a flamethrower or focus the heat of the sun with a
magnifying glass.
In the many complex causal situations that we encounter, it will be impossible
to isolate the only relevant difference or the only relevant similarity. We will also
struggle to determine necessary and sufficient causes because we cannot (normally)
conduct repeated experiments in which we determine the relative state of each causal
element. Rather, we will normally be confronted by a whole jumble of possible
differences and similarities. So, the main function of our investigation of causes, and
of the resulting causal arguments and explanations we write, will be to assign some
significance to each cause (was it a minor or major cause? was it significant enough
to count as a sufficient cause on its own?) or to discover the interrelationship
between causes. We also need to consider the degree to which each cause was beyond
or within human control (was the cause a direct human action, or something in the
general environment?). Further, we need to avoid assuming, simply because events
happen in close proximity to one another, that they are necessarily related as cause
and effect. Perhaps it was simply a coincidence that, for example, the two workers
were sacked. Alternatively, we might argue, in relation to the first example, that there
is a common cause: wage reductions and union exclusion are not necessarily effect
and cause but, rather, could both be the effects of some other cause—perhaps
structural changes in the political economy of Australia.
Reasoning from generalisation
Reasoning from generalisation is another common form of argument or
explanation. Yet it is very different from causation. If causation, at its simplest,