Mechanisms and Laws: Clarifying the Debate - Mechanism and Causality in Biology and Economics - page 122

Leuridan believes they cannot. In this chapter, we assess his arguments. Though
his arguments, as we show below, leave one with no compelling reason to maintain
the traditional view, his discussion demonstrates the need for greater clarity about
the place of laws in mechanistic sciences.
But first the ground rule: All parties to this discussion, as Leuridan points out,
agree that the traditional notion of a “strict law,” the universally quantified material
conditional with unrestricted scope and a good deal besides, has little application in
biology and other special sciences. Mechanists have openly embraced a number of
arguments for this conclusion, most notably John Beatty's ( 1995 ) suggestion that
the laws of biology are evolutionarily contingent and Stuart Glennan's ( 1996 , 2002 )
idea that the generalizations of biology are mechanistically fragile and so probabi-
listic and prone to breakdown. Other mechanists emphasize that theoretical claims
in biology are typically limited in scope, applying only to some species and strains
(cf. Hull 1978 ), and that the scope of such generalizations is restricted to life on
earth in a particular epoch (cf. Smart 1963 ). Whatever the reason, mechanists have
been happy to echo these criticisms as evidence of the limited applicability of the
traditional law-based view to the philosophy of biology. But contra Leuridan's
suggestion, it should also be noted that the mechanist's general opposition to strict
laws does not entail opposition to the idea that biologists and other scientists of the
middle range seek to learn about and describe general facts. None of these
arguments showing that the idea of a strict law distorts crucial features of biology
shows that there are no general facts about biology or that generalizations play no
important role in biological research practice. And no mechanist has ever made
such claims.
Though we proceed by criticizing Leuridan's arguments, we have a larger
purpose, namely, to illustrate how thinking about mechanisms enriches and
transforms the philosophical debate about the role of laws in biology. In our
view, the debate over whether or not there are laws in biology has outlived its
usefulness. Nobody anymore denies that there are stable regularities that afford
prediction, explanation, and control of biological phenomena. Whether such stable
regularities count as laws depends on what one requires of laws, but it is undeniable
that generalizations of this sort do many kinds of work in biology. What remains is
the admittedly difficult work of showing how this is possible. If one takes the
biological sciences to be largely dedicated to the search for mechanisms, in
contrast, one can begin to ask in relatively precise ways how generalization
contributes to the search for mechanisms and, conversely, what the idea of mecha-
nism brings to long-standing questions about how generalizations afford prediction,
explanation, and control.
We begin by clarifying Leuridan's thesis and his central ontological and episte-
mological arguments (Sect. 2 ). In Sect. 3 we consider Leuridan's ontological claims
and argue that Leuridan fails to show that mechanisms must involve regularities
(Sects. 3.1 and 3.2 ) or that there must be fundamental laws without underlying
mechanisms (Sect. 3.3 ). Despite the emphasis Leuridan places on the notion
of projection (i.e., extrapolation), he fails to explain why the generalizations
of biology are stable and why certain facts can be extrapolated while others