2 3 5C o n c o m i t a n t V a r i a t i o ntest for a necessary condition only if we can find cases in which the feature
does not occur and then check to make sure that the target feature does not
occur either. The SCT provides a rigorous test for a sufficient condition only
when we can find cases in which the target phenomenon is absent and then
check whether the candidate sufficient condition is absent as well. In this
case, however, neither check applies, for there is always a certain amount
of acid in the atmosphere, so it is not possible to check what happens when
atmospheric acid is completely absent. Similarly, environmental damage,
which is the target phenomenon under investigation, is so widespread in
our modern industrial society that it is also hard to find a case in which it is
completely absent.
So, if there is always acid in the atmosphere, and environmental damage
always exists at least to some extent, how can we determine whether the SO 2
released into the atmosphere is significantly responsible for the environmen-
tal damage in the affected areas? Here we use what John Stuart Mill called
the Method of Concomitant Variation. We ask whether the amount of environ-
mental damage varies directly in proportion to the amount of SO 2 released
into the environment. If environmental damage increases with the amount
of SO 2 released into the environment and drops when the amount of SO 2
is lowered, this means that the level of SO 2 in the atmosphere is positively
correlated with environmental damage. We would then have good reason to
believe that lowering SO 2 emissions would lower the level of environmental
damage, at least to some extent.
Arguments relying on the method of concomitant variation are difficult
to evaluate, especially when there is no generally accepted background the-
ory that makes sense of the concomitant variation. Some such variations are
well understood. For example, most people know that the faster you drive,
the more gasoline you consume. (Gasoline consumption varies directly with
speed.) Why? There is a good theory here: It takes more energy to drive at a
high speed than at a low speed, and this energy is derived from the gasoline
consumed in the car ’s engine. Other correlations are less well understood.
Reportedly, there seems to be a correlation between how much a woman
smokes during pregnancy and how happy her children are when they reach
age thirty. The correlation here is not nearly as good as the correlation be-
tween gasoline consumption and speed, for many people are very happy
at age thirty even though their mothers smoked a lot during pregnancy,
and many others are very unhappy at age thirty even though their mothers
never smoked. Furthermore, no generally accepted background theory has
been found to explain the correlation that does exist.
This reference to background theory is important, because two sets of
phenomena can be correlated to a very high degree, even with no direct
causal relationship between them. A favorite example that appears in many
statistics texts is the discovered positive correlation in boys between foot
size and quality of handwriting. It is hard to imagine a causal relation hold-
ing in either direction. Having big feet should not make you write better97364_ch10_ptg01_215-238.indd 235 15/11/13 10:48 AM
some third party content may be suppressed from the eBook and/or eChapter(s). Editorial review has deemed that any suppressed content does not materiallyCopyright 201^3 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights,
affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.