Meeting 2: Fitness and Adaptation (1/21)

Required Readings: (I recommend going in this order). Remember that starred items are required for grad students, but optional for undergrads; and double-starred items are optional for everyone and that 'CIEB' denotes the Sober collection Conceptual Issues in Evolutionary Biology.

• Mills & Beatty, “The Propensity Interpretation of Fitness” [CIEB §1]
• Sober, “The Two Faces of Fitness” [CIEB §2] **
• Gould & Lewontin, “The Spandrels of San Marco and the Panglossian Paradigm” [CIEB §5]
• Gould & Vrba, “Exaptation: a Missing Term in the Science of Form” [PDF]
• Dennett, “The Leibnizian Paradigm” [PDF] *
• S&D: Chapter 10: “Adaptation, Perfection, Function” **

We'll address two important issues for the interpretation of evolutionary biology this week: the status of Darwinian claims about fitness (captured in such well-known slogans as "the survival of the fittest") and the concept of adaptation and adaptationist thinking.

As I mentioned today, Karl Popper was deeply influential in the philosophy of science through much of the 20th century. His criterion of science — "falsifiability" — stipulated that in order for a theory to be judged as "scientific", a theory must be subjected to "severe tests". There must be experiments you could run or observations you could make that would show the theory to be false. That's a bad criterion of science for several reasons. Suffice it to say, that many perfectly legitimate theories have not been cast away on the basis of a surprising observation. As Kuhn put it, it's a bad carpenter who blames his tools. I might say more about this next time.

Nevertheless, the idea that science should be falsifiable continues to exert a great deal of influence. Perhaps there's some good reason for this. Unfalsifiability may bespeak other theoretical vices, like circularity. This is roughly the charge Mills and Beatty are concerned to rebut in their article on fitness. As we know, differential fitness is required for natural selection to drive evolution. But what is fitness? How is it defined? Not by "burliness" or "speed" or even being adapted to some environment. What matters is reproduction. It's tempting to understand fitness as a measure of how many descendants an individual leaves behind. But then it's no surprise that the fittest survive to reproduce: for fitness is defined as survival to reproduce! Thus, Darwin's theory is circular and not truly scientific. The basic idea behind Mills and Beatty's approach is that 'fitness' in evolutionary theory can be understood in at least two different ways: fitness of an organism and fitness of a type, and that we should construe each of these claims as claims about propensities to reproduce, rather than actual reproductive performance. (Don't worry about trying to understand the example on p. 16 of CIEB in detail — I'll run through it.)

Another source of concern about Darwin's theory is that it is merely a series of promissory notes — that it doesn't actually explain anything, but merely suggests that some evolutionary story or other is tellable. Kitcher effectively responds to this general charge in connection to the falsifiability concern in his "Darwin's Achievement" (which is certainly worth a look if you have the time). But there remains an influential strand of evolutionary thinking which Stephen Jay Gould and Richard Lewontin famously dubbed "The Adaptationist Program" and lampooned in the spirit of Voltaire: that any trait can be explained by being adaptive (by contributing to the fitness of the organisms which possess it). The ability to tell some unfalsifiable "just-so" story doesn't really explain anything! One of the interesting issues here is whether Gould and Lewontin go too far in their rejection of adaptationist thinking: whether there isn't something to adaptationist thinking.

Dennett thinks so. He offers a refreshingly frank admission of his longing to believe evolutionary stories involving avian honey guides and aquatic apes. . . . Is he just offering an optimistic counterpoint to Gould and Lewontin? Perhaps since, as Dobzhansky put it "nothing in biology makes sense except in the light of evolution", some evolutionary story or other has to be behind various adaptations we shouldn't be too worried about indulging in adaptationist reasoning. The Dennett vs. Gould/Lewontin debate is something we should talk a bit about.

Gould and Elisabeth Vrba, a few years after Gould and Lewontin wrote their influential critique of adaptationism, offered an important concept that illustrates a worry with Dennett's optimistic approach. Because some trait is adaptive does not necessarily mean that its prevalence owes to its having been adaptive in that way (this distinction will loom large in our discussion of biological function in a few weeks). Feathers may be a good example: did they arise and proliferate because they were good for flying or because they were good for other things and then get co-opted for flight? Gould and Vrba want to call this an "exaptation".

Study Questions

• You might think about these questions as plausible targets for your impromptu presentations.
• What is a propensity?
• What is the difference between fitness as actual reproductive success and fitness as a “propensity”?
• Explain how understanding fitness as a propensity relieves the circularity problem discussed in Mills and Beatty.
• Mills and Beatty distinguish between the fitness of individual organisms and the fitness of types. What is the difference? Why do they write that fitness of types cannot be a propensity?
• What is an adaptation?
• What is the adaptationist program according to Gould and Lewontin? What’s wrong with it (according to them)?
• What challenges do the adaptationists face in dividing organisms up into traits (according to Gould and Lewontin)?
• Why think that the adaptationist program is unfalsifiable?
• What is “exaptation”?
• What is a “Bauplan”? How does it fit into the debate about adaptationism?