Massimo Pigliucci tries to dissolve the species problem in a recent essay in Philosophy Now. He suggests that we use Wittgenstein and regard “species“ as a cluster concept, one that is characterized by a loose network of interrelated concepts, rather than by a single defining feature. While attractive in some ways (it mirrors the kind of pluralism I championed in a Philosophy of Science paper I published in 1984) there are other ways in which the solution is unsatisfactory. Or rather, the approach Pigliucci suggests is promising, but it is also incomplete.
Pigliucci makes the excellent point that “species” are used for diffeent purposes by diffeent groups of biologists. The “species“ of an evolutionary geneticist studying the evolution of post-zygotic reproductive isolation may or may not correspond to the “species” of a cladisitc systematist looking for the least comprehensive diagnosable unit. “Species” is a theoretical term, so its meaning derives from the theories in which it is employed. To the extent that biologists use “species” in diffeent theories, the instantiation of “species” will depend on which theory is being used. In some, perhaps many, cases the instantiations determined by different theories will roughly correspond, but in others the instantiations may conflict.
In cases of conflict, we must either reconcile ourselves to using “species” with two (or more) different meanings or we must give one of the theories priority. A reasonable case can be made that evolutionary theory should be given primacy -- I made the argument in my 1986 paper in Philosophy of Science --, but even if we could agree on that, it may not get us very far. Some people argue that horizontal gene transfer is so prevalent in prokaryotes that there are no organismal phylogenies. Even in those organisms in which there is something resembling a purely branching phylogeny, it's not obvious that equivalent “chunks” of the phylogeny can be recognized in all organisms. In what sense does a “chunk” of phylogeny corresponding roughly to what we call a species in mammals correspond with any chunk of diversity in a plant, like several species of wild oats, wild wheat, or wile barley, that reproduce almost entirely by self-fertilization. The tokogeny-phylogeny boundary (some) cladists depend so heavily arises at the boundary between individuals (or small families) in the grasses and may involve hundreds, thousands, or tens of thousands of individuals in a mammal, so a purely cladistic or diagnosability criterion doesn't seem likely to help.
If the “chunks” of diversity are different in different groups of organisms, it's reasonable to think that the processes producing those chunks may also differ. There isn't a process of speciation. There are several, or many, processes of speication. What they have in common is a similar product -- a supra-organismal “chunk” of biological diversity corresponding to some part of the tree of life. So we're back to Wittgensstein's idea of a cluster concept. The cluster may be smaller than Pigliucci suggests, but it's still a cluster.
Leave a comment