Pathways to pluralism: Beckett Sterner on biological individuality, part 2

Aphids on dandelion

Aphids on dandelion. Public domain image from Wikimedia Commons.

Previously, I introduced Beckett Sterner’s new paper comparing and critically evaluating the views of Ellen Clarke and Peter Godfrey-Smith on biological individuality. For Clarke, individuality is recognized by the presence of ‘individuating mechanisms’: traits that increase the capacity for among-unit selection or decrease the capacity for within-unit selection. Godfrey-Smith recognizes different kinds of individuals, but at a minimum, populations of individuals must have Lewontin’s criteria of phenotypic variation, differential fitness, and heritability of fitness, i.e. be capable of adaptive change.

Sterner further classifies Clarke’s view as a purely functionalist account and Godfrey-Smith’s as a hybrid of functionalist and materialist accounts. I assume these are important differences to philosophers of biology, but we poor biologists don’t often worry about such distinctions, so what exactly do they mean? In this context, Sterner seems to be saying that Clarke’s account is agnostic as to the particular nature of the individuating mechanisms: as long as some mechanisms exist that increase among-unit selection or decrease within-unit selection, the unit in question is an individual. In contrast, Godfrey-Smith’s account is a hybrid because, in addition to the capacities for selection and complex adaptation, which can be achieved in a variety of ways, he also adds some quite specific criteria for multicellular organisms (what he calls “collective reproducers”): a single-celled bottleneck, germ-soma differentiation, and functional integration.

After summarizing Clarke’s and Godfrey-Smith’s views on individuality, Sterner shifts to consideration of how we should judge such a concept. Here I think we are in general agreement that definitions are neither right nor wrong, but rather properly judged as more or less useful (in philosophese, what “epistemic work” do they perform). For Clarke, a concept of individuality is useful largely in identifying the units to be counted for evolutionary models, what she calls “the demographer’s dilemma.” Sterner outlines four uses of individuality concepts:

to provide a coherent and justified categorization of biological entities…guiding the correct choice of evolutionary models…to support inductive generalizations…explaining evolutionary transitions in individuality

A concept of individuality is useful to the extent that it fills these roles. After considering each criterion in turn, Sterner concludes that Godfrey-Smith’s hybrid account generally does a better job of distinguishing between individuals and parts or groups. However, Sterner does not argue that Clarke’s purely functionalist account should be abandoned:

It seems likely…that functional accounts of individuality will continue to have a more principled basis in evolutionary theory but will also be more difficult to operationalize for practical use.

In other words, there may be a trade-off between the simplicity and theoretical elegance of a concept on the one hand and its utility on the other.  In addition, Clarke’s criteria are easier for biologists to evaluate than Godfrey-Smith’s:

Arguably the strongest merit of focusing on selection [as Clarke does] rather than adaptation [as in Godfrey-Smith] is its greater practicality: it is generally easier to show that a mechanism influences the degree of variation, heritability, viability, or reproductive capacity in a population than to demonstrate that adaptation can occur.

But Godfrey-Smith’s account does a better job of filling the roles (performing the epistemic work) Sterner has outlined for an individuality concept:

…the capacity for selection alone is too weak a foundation for a functionalist approach to address all the epistemic roles required, and our theoretical understanding is improved by paying attention to the outcomes of selection beyond the next generation.

As a result of these trade-offs, Sterner argues in favor of pluralism in defining individuality:

…an important motivation for pluralism arises when different contexts of use demand that a concept satisfy overlapping but distinct epistemic demands.

I’m pretty sure the discussion about biological individuality is far from over, but one thing seems clear: most recent accounts are some permutation on individuals as units of evolution. Aside from Clarke’s and Godfrey-Smith’s, Sterner mentions two others, those of Queller & Strassmann and of Folse & Roughgarden. All four are explicitly evolutionary accounts, and these are by and large the ones that philosophers of biology are focusing on these days (Thomas Pradeu’s book chapter is an exception). John Pepper and I argued for a broader pluralism:

…when biologists pose questions requiring the recognition of organisms, they should be explicit about what criteria they are using and why. This does not, however, require that we use only one operational definition for all purposes. Instead, we suggest taking a cue from systematic biology, where multiple species concepts now coexist harmoniously. As we have illustrated in the examples discussed above, the diversity of life is so great that a single organism concept cannot usefully be applied to all forms for all purposes.

In at least one sense, we were wrong. Treating individuals (or organisms; we conflated the two) as units of evolution is a strategy that can be applied across taxonomic lines. I think it does require accepting some counterintuitive results, such as that individuality comes in degrees, but, as I’ve argued before, the problem may be not with the results but with our intuitions.


  1. bsterner says

    Thanks for this, Matt! You’ve captured my approach to the problem very concisely. The only clarification I would add is that Godfrey-Smith is a pluralist in multiple dimensions: he has two functional accounts of individuality, based on the capacities for selection and adaptation respectively, plus he uses a material account of reproduction, which is where he introduces his three key physical properties (bottlenecks, germ lines, and functional integration). So there’s quite a lot going on in his 2009 book, Darwinian Populations! Clarke, by contrast, relies solely on the capacity for selection in her account.

    One thing that emerged for me in working on the paper was that it’s not clear precisely how Godfrey-Smith or Clarke’s functional accounts hook up with biologists’ concepts of fitness alignment, fitness decoupling, and modifier traits (see Michod 1999). Clarke’s individuating mechanisms can act to produce fitness alignment, for example, but the idea of alignment itself doesn’t follow from her theory. Similarly, individuating mechanisms don’t explicitly capture interactions between traits such as Michod’s modifier traits. So it seems like there’s some work to be done extending existing functional accounts to mesh properly with the theoretical concepts commonly used by biologists.

Leave a Reply