Page 170 - MODES of EXPLANATION

Basic HTML Version

with something less than universal, exceptionless, necessary generalizations.
Others, such as Elliott Sober (1997) and Ken Waters (1998), have argued that biology
does have laws, on the standard account of laws, but that there are not very many of them.
These laws may be
ceteris paribus
laws or they may be very abstract; maybe they are
mathematical truths. Finally, I have defended the view that biology does have laws, but that
we have to revise the standard account of a law.
This is the kind of conceptual space that the debate about laws and their failure to
apply to biology has occupied. It is a kind of menu. You can choose “no to laws in biology;
yes to standard view,” “yes to laws in biology; yes to standard view” – although these end up
being very odd-looking laws, like Hardy-Weinberg Equilibrium; or “yes to laws in biology,
no to standard view.” I would suggest that the third one is the appropriate way. The
motivation for accepting this view also follows from what we are learning about complex
evolved systems and their impact on our philosophical conception of explanation by appeal to
laws. Again, contemporary science is demonstrating that there are modes of explanation – in
this case, appealing to contingent laws – that need to be characterized philosophically in
order to keep up with the scientific discoveries and understandings that are being developed.
What do I mean by contingency? It is clear that all natural truths are logically
contingent. There is an argument that aims to organize everything into two distinct boxes: the
logically necessary box and the logically contingent box. Suppose that there is a domain that
only contains As and Bs. Then further suppose that in this domain all of the As are Bs. One
might say, “All As are Bs.” This is either a logical truth with logical necessity or it is a
contingent truth that depends on what the As and Bs are. “All squares are four sided” is
logically necessary; “all animals with hearts are animals with kidneys” is a contingent truth.
This is something that was discovered by looking at animals with hearts and animals with
kidneys.
That language of logical necessity, which has a dichotomous structure, has been
carried over to natural necessity. Some truths about the natural world are naturally necessary:
for example, there are no spheres of uranium larger than 100 meters in diameter because such
a sphere becomes unstable and would implode. Some truths about the natural world are
accidentally true generalizations: for instance, there is no sphere of gold that has a diameter
more than 100 meters. The standard view is that accidental generalizations – that is,
generalizations that are merely true of our universe – are not laws. For a true generalization to
be a law, the generalization has to display natural necessity, which is modeled after logical
necessity. However, all laws of science and laws of nature are true of our world but not true
in all logically possible worlds, so they are all logically contingent.
On this kind of view, there are only two options regarding a logically contingent true
generalization. Everything is either in the one box, naturally necessary, or the other box,
contingent. As it turns out, almost nothing is in the naturally necessary, universal,
exceptionless box. We are very hard pressed to find any scientific laws of fundamental
physics, of non-fundamental physics, of chemistry, of biology, of anything that has the
features that are supposed to be the features of a law. However, on the standard view, a law is
needed in order to generate explanations, so we are in trouble here.
6