It can be no coincidence that the first philosopher mentioned in Ernst Mayr's What Makes Biology Unique? is Immanuel Kant. Mayr is negatively impressed with what he takes the state of philosophy of biology to be, and he can't understand why philosophers are engaged in interminable debates that are caused by taking the wrong approach: “most philosophers of science believe the problems of philosophy of science can be solved by logic.” Like Kant, Mayr claims the territory he surveys is unexplored, and like Kant, Mayr means to change both the method and the conclusions.

Despite what he says in his preface about the emphasis on logic, the main methodological problem Mayr identifies with philosophy of biology is that its practitioners have unreflectively viewed biology through a conceptual lens that was developed for physics. Philosophers, he claims, have not realized that sciences are not all the same, and have imposed on biology a philosophy that does not fit. As a consequence, the features that are unique to biological systems—degree of complexity, biopopulations, and dual causation, for instance—have been ignored, often in favor of ‘physicalist’ notions. David Hull, Elliot Sober, Alexander Rosenberg, Phillip Kitcher, and Michael Ruse are among the offenders, and they are often mentioned in the same breath.

Mayr's understanding of the state of philosophy of biology is seriously flawed. The question of whether there is or ought to be a one-size-fits-all philosophy of science has been addressed directly by some of the philosophers he mentions. One example should suffice to show how badly Mayr has erred. David Hull begins his Philosophy of Biological Science (Reference Hull1974), with a section entitled “One Philosophy of Science or Many?” and then goes on to argue explicitly that it is a mistake to think about science “solely in the context of the physical sciences.”

Mischaracterization is not the end of Mayr's troubles. Most of his claims about what makes biology unique are suspect for reasons that begin with his distinction between ‘historical’ or evolutionary biology and ‘functional’ biology. The latter studies physiology, cellular processes, and genetics, and can “ultimately can be explained purely mechanistically by chemistry and physics.” This distinction causes trouble for Mayr's thesis about autonomy partly because it turns out that it isn't biology as a whole that is autonomous: the cut is not between physics and biology, but between evolutionary biology and everything else.

With so much autonomy given up already, one wonders about the strength of Mayr's claims about evolutionary biology. Why stop at a single, biology-specific distinction? Why should only part of biology be different in kind and thus deserving of its own method while all of physics is of a piece? There are after all, historically-oriented subdisciplines in the physical sciences. Cosmology comes immediately to mind, as does planetary physics. Is there a historical/functional distinction to be drawn in physics as well, or do degree of complexity, biopopulations, and dual causation mark biology off from the physical sciences?

There is not space here to take up this question in detail, but a partial look at an answer is enough to show that there are strong arguments against Mayr's approach. Mayr singles out the concept of the biopopulation as “probably the most fundamental difference between the inanimate and the living world.” The idea is that the world of physics is made up of classes with members that are rightly regarded as identical by physicists. The biological world, by contrast, is made up of unique individuals, and the biologist regards them as identical at his or her peril.

Standard predator-prey models, however, are a case in which the thinking is unquestionably typological: there are predators and there are prey, and their interactions are often modeled using sets of differential equations in which variation among the individuals is not even represented. So too with the kind of evolutionary game theory favored by John Maynard Smith. In the well-known Hawk-Dove game, the differences between Hawks, Doves, and Mutants or Defectors are in kind, not degree. Mayr may say that neither of these pursuits is evolutionary in the right way, but a response that further narrows the autonomous part of biology weakens his thesis to the point of collapse.

Mayr's claims about causation and biology's degree of complexity are problematic as well, and for the same sort of reason. The objections that can be raised against them are parallel to the one I voiced with respect to biopopulations—namely that the features to which Mayr points are not discipline relative. Biology is not unique because it is complex or causally messy, since some physical and chemical systems are complex and messy as well. Mayr's mistake is not in arguing that evolutionary biology diverges from logical empiricist notions of science, but in not noticing that parts of physics do as well.

Drawing lines the way he does either limits the autonomous part of the biological sciences to a very small part of evolutionary biology, or it opens the door to finer-grained comparative analyses of both the biological and physical sciences.

What Makes Biology Unique? consists of a dozen chapters that have appeared in other, or the same, form elsewhere in Mayr's corpus. Topics range from those that are directly relevant to the title question to others that relate tangentially at best. In addition to chapters that argue for a separate philosophy of biology, there are chapters on Darwin's influence, on the lack of Kuhnian revolutions in biology, on the origin of humans and our place in the cosmos, on natural selection, and on difficulties with species definitions. The book may not be a prolegomenon to a new method in philosophy of science, but there is much in what Mayr says about the autonomy and uniqueness of biology that should wake us from our dogmatic slumber. Far from being unexplored territory, I think Mayr's conclusions about what makes biology unique are taken by many to be obviously correct. I also think that these conclusions are on very shaky ground.