Natural Selection and Causal Productivity - Mechanism and Causality in Biology and Economics - page 155

productive, too, both through causal propagation and causal interactions. I was
initially motivated to respond to Glennan because it seemed to me that, if correct,
his claims would imply that the merely causally relevant causes occurring at the
population level were somehow “lesser” than the more robust causally productive
causes at the level of individual organisms. These thoughts probably have more to
do with my views about causation than Glennan's, although I am apparently not
alone in this way of thinking; as Jaegwon Kim suggests, “causal production, which
respects the locality/contiguity condition,” involves “ real connectedness between
cause and effect” ( 2007 , p. 236; emphasis in original). Furthermore, Glennan
claims that full understanding of the causal basis of an event requires both the
causally productive causes and the causally relevant causes even though he believes
that at the population level of natural selection, there can be causal relevance
without causal production. This seems to leave bare causal relevance at the popula-
tion level a bit free-floating and weird. Finally, although in this chapter I have not
sought to question the claim that there are two kinds of causes, I find it somewhat
troubling. For all of these reasons, it seemed to me that he was mounting a serious
challenge to my claim that natural selection is a population-level causal process
(Millstein 2006 ): that those population-level causes were “lesser” or “free floating
and weird” or part of a distinction that was not fully coherent and thus perhaps
ephemeral. So, responding to Glennan here is, in part, a defense of my earlier work.
However, I hope to have made some other, more general points along the way.
One is that while I find the new mechanists' approach appealing for many areas of
biology (such as molecular biology and neuroscience), I do not think it illuminates
all cases. This echoes a claim of Skipper and Millstein ( 2005 ), but here I go beyond
that negative claim to show how Salmon's Mark Transmission account can be more
helpful in understanding other sorts of biological phenomena, such as natural
selection. Salmon eventually abandoned his Mark Transmission account because
he felt it relied too much on counterfactuals; however, for people like me who do
not find counterfactuals ontologically objectionable (and anyone who defends a
causal dependence view of causality cannot find counterfactuals ontologically
objectionable), there is much insight to be gained by analyzing cases in terms of
Salmon's account. In part, this is because (as I argued above) phenomena such as
natural selection are better suited to non-decompositional, etiological accounts,
rather than the constitutive decompositional accounts that the new mechanists
emphasize. Sometimes, all we need is to cite causation “at” a level. However, I
also think that concepts such as “causal processes,” “causal propagation,” and
“causal interaction” are rich and powerful tools. I recommend Salmon's Mark
Transmission account as an alternative to the new mechanists' approach - again,
not as a replacement but as a supplement. I expect that other areas of biology and
science more generally might be fruitfully examined through the lens of Mark
Transmission. Whether Salmon's account should itself be considered a type of
mechanist approach is a matter for another time, and I do not think anything I have
said here turns on that question.
Finally, I think it is important that we understand what sort of entities can enter
into causal relations and in what ways. I think we have certain human-centered