Biology Reference
In-Depth Information
So, to return to the point at hand, recall my claim that, contra Glennan, there
seems to be at least prima facie reason to think that populations of water bugs as a
whole often do interact with other entities as a whole. If it were the case that a
population could interact with other entities as a whole to produce changes in the
very same population, then a pure “at a level” etiological explanation would better
illuminate this phenomenon than a constitutive explanation. Thus, Salmon's
account, which emphasizes etiological explanations over decompositional ones -
an account that Glennan accepts as providing an account of causal production, as
mentioned earlier - is a more promising strategy for characterizing natural selection
than the new mechanist accounts, which emphasize decompositional explanation
over etiological explanation. 7
To that end, let me briefly review Salmon's views. Salmon's ( 1984 ) account 8
describes both causal propagation and causal production . Salmon suggests that a
baseball at rest or in motion is a causal process because it is capable of transmitting
(or propagating ) a mark through time without further interactions. For example, if
one makes a scuff on a baseball, the scuff simply persists on the baseball; the
baseball, with its mark, propagates through time. On the other hand, changes in
causal processes are produced by causal interactions, that is, intersections of
processes where changes in the characteristics of the processes occur at and persist
beyond the space-time point of intersection. For example, the interaction of a
moving baseball (a causal process) and a window (another causal process) can
produce a change in both the window and the baseball, namely, the breaking of the
widow and a change in the trajectory of the baseball. Note that there is no
decomposition here; neither the baseball nor the window needs to be broken
down into parts in order to explain the interaction between the two causal processes
or the production of change. Indeed, the mass of the entire ball is one factor
(aside from velocity, wind resistance, etc.) in the window's breaking exactly the
way it did. 9
7 Skipper and Millstein ( 2005 ) offer additional reasons for thinking that the new mechanistic
philosophy does not, in its current form, adequately characterize natural selection. I have focused
on the issue of decomposition here in order to address the decompositional assumption behind
Glennan's claim that population-level properties do not produce change because the population is
not a part of the mechanism that produces changes in genotype and phenotype frequencies. I thus
seek to highlight the way in which Salmon's account can provide a non-decompositional picture of
causal production in natural selection.
8
I focus on Salmon's Mark Transmission account rather than his later Conserved Quantity account
because I believe that it is more broadly applicable to causation outside the domain of physics.
Indeed, Salmon explicitly states that his 1984 account of scientific explanation is intended to cover
many different disciplines, such as the behavioral sciences, the physical sciences, and the biomed-
ical sciences ( 1984 , p. 267).
9 Similarly, Salmon notes that when two moving pool balls intersect in space-time, energy and
momentum are transferred, altering the states of motion of both balls; thus, the intersection is a
causal interaction in which the change in each process can be said to be produced by the other
process ( 1984 , pp. 169-170).
Search WWH ::




Custom Search