Other Games in Other Towns: Pluralism about Biological Function and Representation

Abstract

At the heart of this dissertation is an argument about the differences in some of the explanatory practices of ecology and evolutionary biology. While evolutionary biology is often concerned with the causes of traits – their evolutionary history – ecology often studies traits as causes. I argue for this claim in the introduction, and the rest of the monograph concerns some of the consequences of this inversion of explanandum and explanans. Treating traits as causes rather than as effects has profound consequences for our understanding of functional explanation in biology, as well as in cognitive science, especially for teleosemantics. In general, treating traits only as effects leads to problems when we try to explain how organisms and populations respond to novelty in their environments. Treating traits as causes resolves this difficulty. In the first chapter, I identify and develop a notion of function – realized function – implicitly used in some ecological niche models from conservation biology. A trait’s realized function is its contribution to a population’s ability to occupy its realized niche. I show why ecological niche models require this notion of function as well as how they make use of it via a case study: the use of an ecological niche model to predict the changes in the range of Sub-Saharan amphibian species due to anthropogenic climate change (Garcia et al. 2014). In the second chapter, I re-examine the case of Fodor’s frogs, showing how the notion of a realized function can help resolve the problem of indeterminacy for teleosemantics, according to which a representation’s biological function is unable to provide a determinate content for that representation. I argue that a backward-looking function that did fix the content wouldn’t be sufficient to resolve the problem this indeterminacy poses in ecology, but an appeal to the realized function of the frog’s visual system can. As a result, teleosemanticists would be best served by adopting pluralism about the biological functions that can give contents to representations, becoming pluralists about representation as well. In the third chapter, I show how this same pluralism about biological function can provide us with an answer to the challenge posed by Swampman. We do not need to deny that he would have any representations at all. We can instead say that he has realized representations but lacks selected-effects ones. I use this discussion as a springboard towards the more general problem of novelty, according to which a backward-looking teleosemantics is unable to assign evolutionarily novel contents to representations. I consider the best-developed attempt to resolve this problem and argue that it fails for the same reasons a backward-looking teleosemantics generally fails to handle ecological questions. Again, pluralism about biological function and representation is our best bet to resolve the issue. Finally, I conclude with some general remarks about the pluralism I’ve advocated for. There is no simple way to resolve the various notions of function I’ve used throughout into a single notion, and there is no “best game in town.” Instead, there are many games in many towns. To do good work in biology and cognitive science, we must admit – at least for now – that there is more than one notion of function at play in these sciences. These various notions are more and less applicable in different explanatory and predictive contexts, but are also sometimes jointly required in order to better understand a system of interest. As a result, I advocate for integrative pluralism about biological function.