Biology Reference
In-Depth Information
among others, William Wimsatt (1974, 1976), Lindley Darden & Nancy Maull
(1977), and William Bechtel & Robert Richardson (1993). Robert McCauley
recognizes this sort of multilevel explanation requires what he calls 'explana-
tory pluralism' (1996). The pluralism is crucial. Insofar as pluralism has any
substance, it must at least acknowledge that higher levels of explanation are
not eliminable. Systems level explanations matter. Systems matter. Carl Craver
accurately characterizes mechanistic explanations as multilevel causal explana-
tions that 'explain by showing how an event fits into a causal nexus' (2001,
p. 68). There is some ambiguity in Craver's endorsement, but it leaves open
the interesting thought that multilevel explanations are constitutive and need not
refer exclusively to proximate causes.
For the introduction of our own model of mechanistic explanation, it will
be useful to draw attention also to a more conventional account of reductionist
ambitions. In Kenneth Schaffner's
Discovery and Explanation in Biology and
Medicine
(1993), we are offered a comprehensive account of theory develop-
ment in the biological sciences. He does not take himself to be developing an
account of mechanistic explanation as we understand it, though all his models
are certainly mechanistic. Most important for our purposes, Schaffner offers a
detailed and comprehensive account of what he calls the 'general reduction-
replacement model', aimed at understanding the relations between theories or
models pitched at different levels of organization.
3
It was originally inspired
by the classic model of Ernest Nagel (1961). Nagel originally distinguished the
reduced and reducing theories on a number of grounds, including the obser-
vation that a reducing theory will typically have greater scope and precision
than the reduced theory. Often the thought was that the reducing theory could
displace the reduced theory, at least for explanatory purposes. Unlike some other
models, such as that from Kemeny & Oppenheim (1956), the reducing theory
was supposed to capture the explanatory principles of the reduced theory (see
Richardson, 2006). Schaffner recognized that there could be a range of dif-
ferent cases. At one extreme, the phenomena explained by the reduced theory
are captured by the reducing theory, though without the explanatory principles
characteristic of the reduced theory. These seem paradigm cases of explanatory
replacement. At the other extreme, the reduced theory captures the explanatory
principles of the reducing theory, though perhaps with some modification. These
seem paradigm cases of explanatory reduction. Schaffner's model is premised
3
We do not want to suggest that theories and models are to be identified. One may describe a phenomenon
without explaining it; a model is at most a partial explanation of a phenomenon; and theories are at best
collaborative models. Within systems biology, the ambition is sometimes to elaborate models in such a way
that they become comprehensive theories. We are skeptical that the ideal can be realized. In any case, a
mechanistic model is a model geared toward explaining some behavior (or phenomenon) in terms of the parts
and processes within the system, given specified (and realistic) boundary conditions.
