Meeting 4: Functions (2/4)

Required Readings:

• Godfrey-Smith, "Functions: Consensus without Unity" [PDF]
• Neander, "Functions as Selected Effects: The Conceptual Analyst's Defense" [PDF]
• Davies, "The Nature of Natural Norms: Why Selected Functions are Systemic Capacity Functions" [PDF]

The function of a heart is to pump blood. Some hearts are defective in this regard: they don’t do what they are supposed to do. We can explain why Smith died on the basis of this sort of malfunction. But it isn’t quite clear how malfunctions make sense in a modern conception of biology. Compare the functions of artifacts. My fridge has the function of keeping things cold (when it fails to, we say that it is malfunctioning). We make sense of this by citing the design of refrigerators: their function is to keep things cold insofar as that was the purpose for which they are built — someone had the intention of building a device that would keep things cold. Note how this explains why we do not ascribe the function of making a subtle humming noise to refrigerators: that wasn’t the purpose for which they were built.

The difficulty in biology should now be apparent. If evolutionary theory excises any need for purposes (Aristotelian Final Causes) or intentional design, how can we then attribute function in any normative sense to biological objects? Two answers (or styles of answers) have dominated the literature: etiological (or selected effect) accounts and systemic capacity accounts. The first is represented by Neander’s paper (but was first proposed by Larry Wright) and takes questions about functions as questions about the explanation of a thing’s existence. To ask what the function of the human heart is, on this account, is to ask in virtue of doing what is the heart possessed by humans. It’s in virtue of the pumping action of hearts (and not, say, the pumping noises they make) that their possessors have prospered. So natural selection, on this view, gives us the sense of normativity which we pretheoretically attribute to (proper) functions.

Not everyone is convinced by this general account. Lots of questions arise here. Is the etiological theory a general account of function (Wright’s version appears to be) or is it only applicable in the biological world? Is it meant as a conceptual analysis of the notion of function? Does it suppose that traits evolved carrying out the function that they now have — can the etiologist make sense of Gould and Vrba’s exaptations? Partly in response to difficulties like these, many philosophers have turned to a more liberal causal role (or systemic capacity) account of function which abjures reference to history (selective or otherwise) in favor of what Cummins called functional analysis. On this systemic capacity view, a trait’s function(s) are just whatever causal capacities that trait contributes to a system. On this view, the “normativity” of function wanes, as the number of distinct functions skyrockets. We may still be able to make sense of the function of the heart to pump blood and not make pumping sounds, in virtue of the fact that the former (but not the latter) activity contributes to the performance of the system in which hearts are embedded.

Davies defends this systemic capacity approach, arguing that it actually subsumes the etiological approach. For an excellent (obviously opinionated) background on the different theories and how they compare, you might consider reading his essay first and then circling back to Neander’s.

Study Questions

• Godfrey-Smith proposes that we should understand functions as effects that have led to a trait's being preserved or proliferated through recent episodes of natural selection. Does this effectively deal with the difficulties facing the etiological approach?
• Describe some prima facie differences in the explanatory aims between the systemic capacity account of functions and the selected effect account.
• Neander mentions three objections to the etiological theory of proper functions. Briefly describe them.
• Neander writes that it is “indeed the vague, unifying, everyday notion of a ‘proper function’ from which the biological notion is derived. Nonetheless, the peculiarities of natural selection impose certain constraints upon a more detailed and precise analysis of the biological notion, . . . constraints which do not apply to the everyday notion employed by artifacts” (175). How does this fit in with her take on the role of conceptual analysis in contrast to theoretical definition.
• What do you think of Neander’s response to the “intellectual arrogance” objection?
• How does Neander propose to deal with the “instant winged lions”? Do you find this a satisfactory response to the problem?
• Briefly describe Neander’s cancer objection to the causal role theory. Does Davies have a credible response?
• What is Davies’ argument that systemic capacity functions have a broader range of application than do selected functions? Does it seem sound?
• Does Davies believe that there are selected functions?
• Why does Davies object to natural norms?
• Why does Godfrey-Smith think that Kitcher is wrong to seek unity in different accounts of function?