Fred H. Harrington, Mount Saint Vincent University,
166 Bedford Highway, Halifax, NS B3M 2J6, Canada
In seeking explanations for behavior, biologists recognize
two different types of causation: proximate and ultimate. Proximate
causes include the mechanisms and stimuli that affect the development
and expression of the behavior in an individual during its lifetime.
In common language, proximate causes seek to answer the question:
"How does this behavior occur?" For example, changes in daylength
trigger the release of hormones that cause the midwinter peak
in reproductive behavior. Ultimate causes are the selective
factors that affect reproductive success and have consequently
been responsible for the behavior's evolution within the species.
These causes have acted over countless previous generations,
shaping not only the behavior now seen in individuals, but also
the proximate machinery that governs the development and expression
of that behavior. In common language, ultimate causes attempt
to answer the question: "Why does this behavior occur?" For
example, the midwinter peak in reproductive behavior insures
that pups are born when their chance of survival is greatest.
These two forms of causation are sometimes confused in discussions
of wolf behavior. This paper will analyze the proximate and
ultimate causes of several behaviors, including howling and
"surplus killing".
With regard to howling, for example, possible answers to the
question: "Why do wolves howl?" have included: 1) to facilitate
the reunion of pack members; 2) to advertise/defend a territory;
3) to find a mate; 4) to express an emotion (i.e., joy, loneliness,
etc.); and 5) to strengthen social bonds among packmates. All
the above may be legitimate explanations for howling, but they
represent a confounding of proximate and ultimate causation.
The first three can be reduced to two (attraction/repulsion)
and represent the most likely selective forces responsible for
the evolution of howling. The fourth represents a proximate
cause, in this case the internal hormonal and neuronal mechanisms
which stimulate a wolf to howl in an appropriate way in a specific
context. The fifth, if it occurs, likely represents a non-adaptive
consequence of the former: howling probably did not evolve to
strengthen pack bonds, although it may have that beneficial
consequence today as a byproduct of its evolved functions. Thus
wolves that howl because they "enjoy" howling (proximate cause)
may deter a neighboring pack from trespassing or may assist
a packmate in returning home (ultimate causes).
Careful analyses at both levels of causation will give us
a fuller appreciation for the multitude of causative pathways
that influence the behavior of wolves.
Beyond 2000: