296 Speciesindiscernible entities of physical theory.^37 So far as we can tell, biological objects are
nearly always discernible.
Without laws, biological kinds are in sharp distinction from other natural kinds,
so we shall consider shortly if there are laws in biology. Another difference lies in
the historical aspects of biological entities—an electron at t behaves as an electron at
t + n does. A radioactive atom has the same probability of decaying at one moment
as another, although in large samples of such atoms a fixed proportion will have
decayed after a set period has elapsed. However, biological kinds are highly sensitive
to prior states. Genetically individual organisms develop differently based on differ-
ences in their point of departure, such as parental antibody inheritance in placental
mammals, or inheritance of methylation patterns. Likewise, gene trajectories in a
population depend on the initial structure of the gene pool, and the evolutionary
history of a new species is tightly constrained by the states of the ancestral species,
its gene frequencies and developmental processes. Horse descendants will never^38
evolve wings, for instance. No vertebrate will evolve into a centaur. The tetrapod
body plan is too restricted by the developmental sequence. And a trait once lost
will rarely be regained in an evolutionary sequence, and almost certainly not in the
same manner, since, as the number of steps of mutation and recombination of genes
increase, the likelihood of complete reversion drops astronomically (Cope’s Rule).^39
Must we therefore deny that biology has NKs of its own, and that it has no laws,
however many law-like generalizations it may develop?^40 Must we say that theoreti-
cal biology is not actually scientific?^41 Or are all biological NKs actually the NKs of
physics or chemistry? To answer this (the last query in the affirmative) we need to
understand the role of models in biology, and their relation to contingent historical
individuals.
Species and other phylogenetic—that is to say, historical—taxa do not form NKs
just in virtue of their phylogeny, but they do form NGs. More exactly: taxa can form
NKs, but in a sense not usually intended by NK advocates: there are laws in biology,
and they do involve NKs, but as NKs are not historical entities, they are timeless
entities, sub specie aeternitatis. Grades are states that may be attained in many dif-
ferent ways and at many different times, and which need not share either the same
physical substrates or historical causal processes. The universals of grades are super-
venient universals, which, while they are de re kinds, get picked out of the world as
kinds in virtue of the models we apply to a more or less restricted domain. If nothing
(^37) A common response is to claim ecological laws as laws of biology that are not reducible to physics,
but it is my opinion that ecological laws are merely biological and complex cases of the least action
principle. As such, they are physical laws. Further, by “law” here, I simply mean a generalization or
model that covers the entire explanatory domain.
(^38) Never? Hardly ever. The probability is so low as to be unlikely to be realized within the duration
of the universe. But Hull’s Law states that there is nothing so strange that there is not at least one
example of it in biology.
(^39) One caveat: organisms may be (genetically, behaviorally, or ecologically) “identical” in the sense that
substituting one for another makes no major difference to the outcomes in a given model. This is not
unlike Templeton’s “demographic or genetic indiscernibility.”
(^40) Nagel 1961, Armstrong 1978, Dupré 1993, Rosenberg 1994, Woodward 2001, Ao 2005.
(^41) Mur ray 2001.