Even if the story is apocryphal, the outcome of the experiment is not. Had
Galileo conducted the experiment he would have discovered that the balls did
fall at the same rate and that their rate of fall was independent of their mass.
Now having done this, what conclusion can be drawn?
From the point of view of science, as distinct from scientific method, the ideal
conclusion is a universal law of physics. It would take the following form: “All
bodies fall to earth at the same rate”. It is inherent in this proposition that the
rate at which a body falls to earth is independent of its mass.
This universal law, however, is not totally supported by the one experiment.
So far it has been shown to apply on just one occasion. In principle this might
be overcome by repeating the experiment at different places, from different
heights, with objects of different shapes materials and with different weights.
Assume that these all showed the same result, as they surely would, does this
prove the law? It does not prove the law absolutely but, to state the minimum
positions, each time the experiment is repeated with the same outcome it
makes it more probable that the supposed law is true.
While repeated trials with the same outcome may not conclusively prove the
existence of the causal law, it does provide inductive support for the law.
Essentially the best inductive reasoning says that many observations of the
same outcome from a process provide strong even if not conclusive support
for a law. The more times a supposed law seems to work the more likely it is
that the law exists.
There is also another approach. Where science is dealing with inert matter
such as the falling of metallic balls it is a reasonable assumption, based on
centuries of experience (strong induction), that inert matter exhibits constant
properties. If this reasonable assumption is accepted it strengthens the
inductive argument to the point where it is almost deductive.
By contrast to experiments with inert matter, experiments with living things,
such as humans, have limitations that do not apply to inert matter. (i) Scientists
cannot draw on an assumption that all people are the same. People, who are
the subject of behavioural science, are not as unyieldingly homogeneous as
physical matter. Obviously people share vast similarities but there are also
variations. This is one reason that any generalisation from one experiment must
always be cautious. (ii) Behavioural scientists generally cannot control and
measure the variables in their experiments as easily as physical scientists can.
(iii) The typical experiment involves not the whole population of people with a
certain characteristic that is of scientific interest such as a common disease, a
common traumatic experience or a common occupation. Instead it relies on a
sample. Thus any result from the experiment involves deriving a conclusion
 |
 |
 |