Philosophy of Biology and Cognitive Sciences

Biologists distinguish the functions which a trait is supposed to serve from the functions it has only accidentally. For example, whilst the heart is supposed to pump the blood, but only accidentally makes a thumping sound. Those functions a trait is supposed to serve are called its proper functions. Under what conditions is some function a trait’s proper function?

There are two dominant classes of answer to this question. There is agreement between both that the notion of a proper function, at least in a biological context, is closely connected to natural selection. Where they disagree is in whether that connection concerns the trait’s history. According to backwards-looking accounts (Millikan 1986; Neander 1991a, 1991b; Godfrey-Smith 1994), the truth conditions for proper functional ascriptions concern facts about an organism’s ancestry. Thus on a fairly standard view the proper function of a trait are the effects of previous tokens of the same type, which conferred a selection advantage to the organisms that served as the bearers of those tokens. On this view, proper functions are primarily properties of trait types, and only derivatively of trait tokens. In contrast, proponents of forwards-looking accounts (Pargetter & Bigelow 1987; Nanay 2010) take the truth conditions to concern the trait’s modal or dispositional properties. Roughly, they are those dispositional or modal properties that would confer a selection advantage under suitable conditions. On this view, proper functions are primarily properties of trait tokens, not types.

 

This paper has two central aims. The first is to argue that all existing forwards- and backwards-looking accounts are inadequate. Drawing on recent work by Bence Nanay (2010), we argue that backwards-looking accounts are inescapably viciously circular. We then argue that forwards-looking accounts fail to capture the explanatory power of functional ascriptions in the biological sciences. In particular, we argue that at most forwards looking accounts show that proper functions can be explanatory, not that they are. Drawing on the debate concerning the proper function of the giraffe’s  long neck, we argue that only the latter is in keeping with actual biological practice.

 

The second is to outline and defend a novel backwards-looking account of proper functions that takes proper functions to be primarily attributable to trait tokens, and only derivatively to trait types. We call this the token etiological view. Standardly, etiological accounts that apply primarily to tokens have been thought impossible, as selection only acts on trait types. We argue, however, that whilst proper functions do have an important connection to selection, that connection can be adequately understood at the token level in terms of the dual notions of inclusive fitness and comparative similarity alone. Roughly, it is argued that a trait’s function is amongst its proper functions just in case there is a previous ancestor of the organism that bears the trait token, such that the ancestor’s most similar trait served that function, and by doing so contributed to that organism’s inclusive fitness. Since we make no appeal to types, the charge of circularity is avoided. Moreover, we argue that this conception properly accounts for the explanatory role of proper functional ascriptions in the biological sciences. More precisely, where the conditions above are satisfied, an organism’s standing in the relevant relation to ancestral traits raises the probability that the organism possesses that trait.

​

References
Bigelow, John, & Pargetter, Robert. 1987. Functions. Journal of Philosophy, 84(4), 181–196.
Godfrey-Smith, Peter. 1994. A Modern History Theory of Functions. Nous, 28(3), 344–362.
Millikan, Ruth G. 1984. Language, Thought and Other Biological Categories. MIT Press.
Nanay, Bence. 2010. A Modal Theory of Function. Journal of Philosophy, 107(8), 412–431.
Neander, Karen. 1991a. Functions as Selected Effects: The Conceptual Analyst’s Defense. Philosophy of Science, 58(2), 168–184.
Neander, Karen. 1991b. The Teleological Notion of ’Function’. Australasian Journal of Philosophy, 69(4), 454–468.