Biology Reference
In-Depth Information
mechanisms or Salmon/Railton mechanisms) are a special, limiting case of
cs-mechanisms, not something altogether different. While Leuridan's thesis that
there can be no
cs-mechanisms
without some stable behavior produced by that
mechanism (cf.
2010
, p. 330) is tautologically true, Leuridan's unqualified thesis
that “there are no
mechanisms
without
stable regularities” (
2010
,
p. 318; our emphasis) is clearly false. One-off mechanisms are mechanisms
without
a macrolevel regularity. So much for the ontological claim.
...
macrolevel
...
3.1 Extrapolation of Generalizations
Things look a bit more promising if we reconstruct Leuridan's projectability thesis
as a purely
epistemic
thesis. Morgan wanted to apply what he learned about the
mechanisms of heredity by studying
Drosophila
in the lab both to flies outside the
lab and to other species. Surely the mechanist owes some kind of story about how
this is possible. The clear solution, one might think, is to recognize that there are
laws - however exception-ridden, probabilistic, and mechanistically fragile - that
license this application. And one might insist that Morgan referred to, and indeed
formulated, Mendel's second law while making a career of discovering exceptions
to independent assortment (see Allen
1978
; Darden
1991
). Scientists form
generalizations, and then they use those generalizations to say what will happen
in new cases. Of course, no mechanist denies that induction and extrapolation (or
projection) are important to science. But how are p-laws supposed to help with this
task? If p-laws are merely law
statements
, as Leuridan defines them, then they are
clearly not the kind of thing that can explain why a given regularity is stable and
strong. Law statements
express that
, but do not
explain why
, certain regularities are
stable and strong. It seems we must understand Leuridan to mean that stable
p-regularities themselves (rather than descriptions of p-regularities) are necessary
for one to extrapolate mechanistic knowledge. Here, in full, is Leuridan's discus-
sion of stability: “What are the conditions on which the regularity under study is
contingent? How spatiotemporally stable are these conditions? And what is the
relationship between the regularity and its conditions (is it deterministic, probabi-
listic, etc.?)” (
2010
, p. 325). Given that stability is defined in terms of the range of
circumstances in which a generalization holds, the epistemological thesis that
extrapolation to conditions outside of the laboratory and to conditions in other
organisms requires p-regularities, again, amounts to a tautology: if the regularities
discovered about
Drosophila
in the laboratory are to hold outside of the laboratory
and for other organisms, then there must be organisms outside of the laboratory for
which the regularity holds. If a regularity holds only in Morgan's laboratory or only
for
Drosophila
, then there is nothing outside of the laboratory or in other organisms
about which to extrapolate. But this is not an explanation of why knowledge
extrapolates beyond the laboratory; it is simply a claim that it does extrapolate
outside the laboratory. Put this point another way: by helping himself to the idea of
p-laws, which are by definition stable regularities, Leuridan does nothing to
explain
Search WWH ::
Custom Search