Is Natural Selection a Population-Level Causal Process? - Mechanism and Causality in Biology and Economics - page 171

It would be helpful to start with an analogy. Suppose two ethnic groups—call
them N and I , respectively—coexist in a kingdom, ruled unfortunately by a tyrant.
An apartheid regime is rigorously maintained in the kingdom. Though the two
groups currently have about the same population size, group I is known to be much
more fertile than group N . Fearing that the noble group N would be outnumbered by
the ignoble group I , the tyrant has a purge plan in mind and cannot wait to execute
it. He promulgates a law according to which the ratio of population size between
group N and I should remain x :1, where x represents the number of years after the
promulgation of the law. The imperial army is responsible for the enforcement of
the law. Since the promulgation of the law, each New Year's Eve, the imperial army
kills enough members of group I to achieve the right ratio and ensure that the law
would not be violated. Each New Year's Eve, the whole population of group I is
chased by the imperial army into a forest. And then officers of the imperial army
compete with each other to hunt as many people of group I as possible. The hunting
game will not end until the legally required ratio of the year is definitely achieved.
Let's see how such a scenario helps to shed light on the causal role played
respectively by the tyrant's law and by the brute action of the imperial army.
Consider first the tyrant's law. Apparently, without the tyrant's promulgation of
the law, the relative frequency in population size between group N and I would not
have changed in accordance with the ratio x :1. In this sense, it seems safe to say that
each annual change in the relative frequency is caused by the tyrant's law.
Now consider a member of group I , say, Ian, who unfortunately was hunted
down and fell victim to an officer of the imperial army, say, Ned. What is the cause
of Ian's death? Undoubtedly, Ian was killed by Ned. So without Ned's killing, Ian
would not have died. In this sense, Ned's killing causes Ian's death. But didn't Ian
get killed also by the tyrant's law? That is, doesn't the tyrant's promulgation of the
law also cause Ian's death? The answer is apparently yes, for without the law, Ned
would not have killed Ian. Few, if any, would deny that the tyrant's law is part of the
cause of Ian's death. Indeed, to tell a causal story about Ian's death, one needs to
invoke not only Ned's brutal behavior but also the tyrant's enactment of the law.
Otherwise, the causal story would be regretfully incomplete.
However, the fact that the two factors contribute jointly to Ian's death shouldn't
mislead us into conflating their separate effects. A definite number of people among
group I are killed due to the tyrant's law, while it is Ned's killing that makes Ian one
of them. The tyrant's law determines how many people will die, while each officer
of the imperial army determines exactly who are going to be victims. What is the
separate effect of the tyrant's brutal law? Answer: the (change in the) relative
frequency in population size between group N and I . And what is the separate effect
of Ned's killing? Answer: Ian's being among the dead. From the combination of the
two separate effects, one gets the result: Ian's death. So the fact that both the
tyrant's law and Ned's killing are part of the causal story of Ian's death does not
imply that either of them fails to have a separate effect. Indeed, the tyrant's law is
causally responsible for a definite number of people's death, while Ned's killing is
for Ian's being among one of them. To causally explain Ian's death, one needs to
pinpoint not merely Ned's brutal killing but also the tyrant's terrifying law.