Biology Reference
In-Depth Information
According to Sober, (1) is true but (2) is false. That is how selection-of differs
from selection-for. Here is a case of selection-of:
Consider a population in which there is selection for being green; this selection pressure
exists because being green camouflages organisms in the green environment they occupy,
thus protecting them from predators. Suppose further that there is no selection for being
small—body size is selectively irrelevant. And now imagine that all and only the green
organisms in the population are small. In this situation, the green organisms are selected,
which means that the small ones are too. However, though there is selection for being
green, there is no selection for being small. Selection-of is the concept that is tightly
connected to variation in fitness; if there is selection of green (small) organisms, then
they are on average fitter than those that are not. But the fact that small organisms are fitter
than organisms that are not small does not entail that there is selection for being small.
(Shapiro and Sober 2007 )
It is noteworthy that selection-of is a discriminate sampling process in the sense
that such a process of selection is not random. In fact, selection-of is a sampling
process which, based on differences in fitness between organisms, favors fitter
organisms. And that is how small organisms are favored and selected. On the
other hand, it is also noteworthy that, in the course of selection-of process,
differences in fitness are causally irrelevant to differences in survival and repro-
ductive success among organisms. The fact that small organisms are fitter is
causally irrelevant to that fact that small organisms have a greater success in
survival and reproduction. Small organisms are selected accidentally.
Now, let's turn to selection-for. It is beyond question that selection-for is a
discriminate sampling process. And such a sampling process, to be sure, is no less
discriminate than selection-of. Still, selection-for is distinguishable from selection-
of, and that is why Sober draws a contrast between them. In the case of selection-
for, as opposed to selection-of, differences in fitness among organisms are causally
relevant to differences in survival and reproductive success. Recall that Millstein
characterizes natural selection as “a discriminate sampling process whereby physi-
cal differences between organisms are causally relevant to differences in reproduc-
tive success.” Such a characterization, as one can see now, fits well and only well
with selection-for. As to selection-of, it is missing in Millstein's account of natural
selection.
Consider Brandon's ( 2005 ) account of natural selection. Brandon takes natural
selection to be a discriminate sampling process where discriminate means unequal
probability of being chosen in the course of sampling process. So, without any
difference in fitness among a population of organisms, there would be no discrimi-
nate sampling process. Difference in fitness among organisms is a sine qua non
condition for natural selection. In this sense, Brandon's account has the advantage
of putting natural selection-of into its place: natural selection is a discriminate
sampling process including both selection-for and selection-of processes. By con-
trast, in Millstein's account, there is a conflation between natural selection-of with
drift if (1) between the discriminate and the indiscriminate sampling processes that
she characterizes, there is no third option and if (2) by drift she means each and
every sampling process whereby physical differences between organisms are caus-
ally irrelevant to differences in reproductive success.
Search WWH ::




Custom Search