151
The following two kinds of causal reasoning are quite different from the previous three.
Concomitant Variation appeals to the related correlation of two rates of change, and concludes
that there is a causal relationship between them. Examples are easy to find:
Your grandfather gives you some vinyl records and an old LP stereo system. It’s so old, that all
the letters and numbers have been worn off the knobs. By trial and error you figure out how to
turn the stereo on, and throw on a near-mint-condition copy of Meet The Beatles. When “I Want
to Hold Your Hand” comes on, you want to raise the volume to decibels that could neuter frogs
at 100 yards. But you don’t know which knob will do the trick. You take hold of one, and at the
rate at which you turn it you hear the volume of The Beatles’ lilting refrains rise. As you turn the
knob back the other direction, the song fades in volume. You conclude that you have found the
cause of the volume level; that is, you’ve found the volume knob.
Or, you are a precocious, inquisitive, and logically-minded four-year-old child. You are sitting in
the front seat of your mother’s idling car (she’s away for the moment to deliver a Senate
document to her Justice Committee secretary). You don’t know how to drive, but are aware that
the vehicle has the capacity to go quickly. Pushing aside “Re-elect Senator Sunny Shine” flyers
from the driver’s seat, you start playing around with pedals on the floor, and note that as you
press one, the car begins to move. As you press it further, the car accelerates and moves faster.
When you press the pedal to the floor, you whisk across the parking lot, slamming into other
cars, barely missing the now-terrified ambassador from Estonia. You let up on the pedal, and the
car slows to a halt. You logically conclude that you have discovered the cause of the car’s
acceleration and rate of speed. You have just used the method of Concomitant Variation.
In both examples above, the concomitant (i.e., naturally accompanying, associated, attendant)
variation is direct, or parallel. As one thing goes up, the other thing goes up. As one thing goes
down, the other goes down. Sometimes the relationship may be inverse, as when one thing goes
down, the other thing goes up. For instance, as employment rates go up, crime goes down. This
indicates that there is a likely causal connection between crime and employment rates. It is
beyond the method of Concomitant Variation, however, to determine which is causing which. Is
it the drop in crime that is causing the rise in employment rates, or is it the rise in employment
rates that is causing the drop in crime? Common sense or further inquiry is needed to make that
determination.
More complex cases of concomitant variation occur when, for instance, one rate of change is
fairly steady, while another rate of change is different (either directly or inversely). For instance,
imagine events A are changing at the rate of 1 unit per day: 0, 1, 2, 3, 4, 5, 6…. Let’s say events
B are also ascending (or descending) but more slowly, but at a different rate: 0, 0.5, 1, 1.5, 2, 2.5,
3…. There still is concomitant variation, as when A goes up one unit, B goes up half a unit.
Since the rate of change (though different) appears to be related, we are justified in (at least
tentatively) concluding that one set of events is causally related to the other set.
But consider this. It oddly is the case in the larger of U.S. cities that as the crime rate goes up, so
too does the amount of ice cream eaten; and as ice cream is eaten less often, the crime rate drops.
Concomitant Variation indicates that ice-cream-eating is causally related to crime. We thus