Biology Reference
In-Depth Information
2. THE BASIC CONCEPTION OF MECHANISM
I begin with a basic characterization of mechanisms that captures many of the
features that have figured in recent philosophical accounts of mechanism. I will
then elaborate it into a framework for mechanistic explanation. A mechanism
'is a structure performing a function in virtue of its components parts, component
operations, and their organization. The orchestrated functioning of the mechanism
is responsible for one or more phenomena'.
(Bechtel & Abrahamsen, 2005)
The first thing to note about this characterization of mechanism is that a mecha-
nism is responsible for a phenomenon (Bogen & Woodward, 1988) that is here
characterized as the function of the mechanism.
6
The identity conditions for a
mechanism are provided by the phenomenon such that what count as parts, oper-
ations, and organization are determined by the phenomenon (Kauffman, 1971).
By characterizing a mechanism as a structure I mean to emphasize that it consists
of an arrangement of parts and has at least some enduring identity. Sequences
of causal operations not organized into an enduring system are not mechanisms
on this account.
Just as mechanisms themselves are identified in terms of phenomena, mecha-
nistic explanation starts with a characterization of the phenomenon to be
explained and seeks to characterize the responsible mechanism. Researchers do
not simply hunt for mechanisms, but seek them to explain an already identified
phenomenon. Part of identifying a phenomenon involves empirical research that
identifies environmental conditions under which the phenomenon will appear.
For example, Pasteur determined that yeast perform fermentation when in an
oxygen-free environment. This does not mean that the characterization of the
phenomenon remains fixed; on the contrary, investigating the responsible mecha-
nism may lead researchers to revise their conception of the phenomenon (in
Bechtel & Richardson, 1993, we characterized this as 'reconstituting the phe-
nomenon'). For example, research on metabolic mechanisms in organisms began
by construing them as responsible for the generation of heat (Mendelsohn, 1964).
Only after Karl Lohmann's (1929) discovery of adenosine triphosphate was it
on the interactive processes that make it up, that is, on the dynamic organization in which biomolecules (or,
rather, their precursors) actually get integrated'.
6
There are two important features of this characterization. First, in characterizing the resulting phenomenon
as the function of the mechanism, I am not committing myself to an evolutionary analysis of function in the
manner of Wright (1972). Indeed, as I note below, the construal of biological systems as autonomous systems
provides the basis for a very different characterization of function. Second, while there is a significant amount
of flexibility available to the scientist in demarcating the phenomenon and determining whether the researcher
is dealing with one or more phenomena, if the researcher demarcates multiple phenomena, then the researcher
will offer multiple, potentially overlapping, mechanisms to account for them.
