Natural Selection and Causal Productivity - Mechanism and Causality in Biology and Economics - page 149

Although Salmon distinguishes between causal propagation and causal
production, it seems to me that causal propagation can be construed as a type of
causal production, or, at least, it can be construed as causal production in Glennan's
sense. Recall that, according to Glennan, causal production is (1) a relation between
events, (2) local, and (3) transitive. Propagation very clearly satisfies all three of
these criteria. The events in question are the ball at one point in space-time and the
ball at a subsequent point in space-time; these events are contiguous in time and
space and, given a third event in space-time, transitive. I will refer to my claim
that causal propagation is a type of causal production later in the chapter.
Now suppose that, like a baseball, a population were capable of transmitting a
mark; it would then be considered a causal process on Salmon's account, capable of
propagating causal influence through space and time. If so, the population could
interact with other causal processes, producing a change in the characteristics of
those processes at the same time that the other processes produced a change in the
characteristics of the population. Then it would seem as though a population could
be causally productive of its own changes without citing the activities of the
organisms that compose it. But for this to be the case, a population would need to
be an object (categorically, and not just in certain situations), so let us turn to that
question.
4.2 Populations as Individuals
Elsewhere (Millstein 2009 , 2010 ), I argue that populations are individuals
(“objects”), using the Ghiselin-Hull individuality thesis as my inspiration (Ghiselin
1974 , 1997 ; Hull 1976 , 1978 , 1980 ). Briefly, my argument is that populations are
composed of individual organisms, just as organisms are composed of individual
cells; a population is a particular thing - not a class, since it exists in space and time,
and not merely a set, since it is integrated via the survival and reproductive
interactions of its constituent members with members having a shared fate (albeit
less so than organisms); a population has a beginning in time (e.g., migration
of organisms away from a population) and an ending in time (e.g., death of the
last organism in a population); a population does change over time, but so do
organisms; and a population is continuous in time via the causal interactions that
occur over time. 10
10 Here one might worry about circularity if individuals (“objects”) are characterized in terms of
interactions, if causal processes are objects persisting and changing through space-time, and if
interactions are intersections of causal processes. However, Salmon ( 1994 ) clarifies that
interactions are not to be defined in terms of causal processes, only in terms of processes more
generally, where “[a] process is something that displays consistency of characteristics” ( 1994 ,p.
299). Causal processes are then characterized by their ability to transmit marks, where a mark is a
type of interaction - “an alteration to a characteristic that occurs in a single local intersection”
(Salmon 1994 , p. 299). An object persisting or changing through space-time is one example of a
causal process; however, a carrier wave is another.