Natural Selection 119of this genetic constraint.
Adaptationists tend to think that cases of heterozygote superiority are rare —
there is a good reason that most biology textbooks use the sickle-cell example, since
such cases, they argue, are unusual. Anti-adaptationists, in contrast, worry that
genetic constraints might be more common. The fittest trait can also fail to evolve
where heterozygotes areinferiorto both homozygotes. In such cases, the popula-
tion might evolve to 100% of the homozygote that is not the most fit. Unlike the
case of heterozygote superiority, there would not be a phenotypic polymorphism
as with the sickle-cell anemia case. Once the population is at equilibrium, the
evolutionary process may have destroyed the information about whether the other
homozygote was the fitter alternative.
Another kind of constraint is developmental — fruit flies have only a limited
number of eggs and can have either high fecundity early in life or late in life,
but not both. It might in some sense be optimal to have a high fecundity both
early and late in life, but it appears that this not an option because of constraints.
Many traits — e.g., nipples in males — are developmental byproducts, rather than
adaptations.
Adaptationists also worry about the extent of pleiotropy, in which one gene
leads to two different phenotypic effects. If one of these effects is harmful, the
case is known as antagonistic pleiotropy. Dawkins tends to think that the power
of natural selection will overcome such constraints. He writes “if a mutation has
one beneficial effect and one harmful one, there is no reason why selection should
not favour modifier genes that detach the two phenotypic effects, or that reduce
the harmful effect while enhancing the beneficial one” [Dawkins, 1982, 35].
The second major form of adaptationism is methodological adaptationism.
Methodological adaptationismis the view that adaptationism is an indispensable
research tool for finding out about whether a trait is an adaptation. Sober states
the position a bit more strongly “The only way to find out whether an organism
is imperfectly adapted is to describe what it would be like if it were perfectly
adapted.” [Sober, 1996, 54]. Stephens and Krebs [1986] say something similar
— the only way that one can get information about constraints is via an opti-
mality model. Amundson [2001] argues, however, that one can sometimes get
information about constraints from facts about developmental biology alone and
not by an optimality model. Still, as I discuss in more detail below, adaptationists
can claim that theirs is a position only about the power of selection acting on
phenotypes that are actually generated. Methodological adaptationism, though
somewhat contentious, is less controversial than empirical adaptationism.
Critics such as Gould and Lewontin [1979] argue that adaptationist explana-
tions are too easy to make up. Biologists have responded in part by making
theories with predictions that are harder to confirm. Biologists have two major
tools for confronting the methodological challenges associated with adaptationism
[Kitcher, 1985; Reeve and Sherman, 1993; Sterelny and Griffiths, 1999; Sober,
2000]. On the one hand, as Maynard Smith [1978] has emphasized, biologists can
and should develop optimality models that make precise, quantitative predictions.