Six Years Old Today
Retro as Genre, Retro as Art

Why Co-operation? (2) Group Selection Strikes Back

Last week, the rise of kin selection as an explanation for co-operative behaviour, and the dismissal of group selection as significant. And now, the conclusion…

Williams George C. Williams, an evolutionary biologist specializing in fish (pictured here with his always awesome beard), was a particularly influential voice in the case mounted against group selection. His 1966 book Adaptation and Natural Selection laid out the basis for the gene centred view of evolution (and was, in fact, a major source of the ideas Dawkins presented in The Selfish Gene ten years later). Williams examined a variety of possible cases of group selection (mostly in fish species) and determined in each than an explanation in terms of individual selection was always available. He thus concluded that selection between groups was “impotent in a world dominated by genic selection and random evolutionary processes”. He did not dismiss the possibility of group selection completely, but asserted that biologists should “postulate adaptation at no higher a level than is necessitated by the facts.” Williams was mindful of the inherently changeable world of scientific models, however, admitting that there was still much to be learned: “I am sure that by the standards of a generation hence, our current picture of evolutionary adaptation is, at best, oversimplified and naïve.”

Sober and Wilson recently returned to Williams’ seminal text in order to reassess the relative merits of the kin selection and group selection perspectives. Praising much of the foundational work Williams pursued, they draw particular attention to the idea that “adaptation at a level requires selection at that level”, calling this Williams’ principle. This adage sets the requirement that must be met for group selection to be considered valid: the fact that some trait provides a recognizable benefit to groups isn’t enough for that trait to be considered a group adaptation – it must have developed precisely because it was beneficial to the group. Using this as the basis for their discussions, they criticize a number of biologists for violating Williams’ principle, and thus continuing to defend a kin selection-inspired perspective against group selection.

The biologists in question are typical examples of the exaltation of genes this chapter addresses, maintaining (Sober and Wilson assert) “that the individual is always a unit of adaptation no matter what the mix is of group and individual selection.” Sober and Wilson point out that this position is being held on the basis that individuals are predicted to maximize their inclusive fitness, and this claim holds true irrespective of the relative strengths of individual or group selection. But this violates Williams’ principle. Kin selection arguments can’t rule out group selection from happening, since inclusive fitness only entails claims about the rates of gene survival – it has nothing whatsoever to say about the scale at which adaptations occur. In fact, as philosopher of science Samir Okasha has noted, rather than claiming individual selection always occurs, there are many situations in which the question of whether or not there is individual selection taking place is not defined.

There are now a vanguard of scientists and philosophers who recognize that it is not the case that group selection is a force too weak to have any influence in evolution, and champion what is called multi-level selection – the idea that selection and adaptation can occur at the level of the individual, the group, or even at the level of an ecology. Note, however, that in all these cases, the accumulation of changes still happens via alterations in gene frequencies (although it is important to remember that even among animals there are also persistent cultural and environmental effects that play an important role in the story of life). The breakthrough realization has been that the metaphorical ‘selection’ that takes place occurs within a specific context. It is always possible to render these effects solely in individualistic terms, but transformable perspectives are not always equivalent in explanatory value. If group selection played a major role in the occurrence of a particular trait, is it really helpful to talk about that trait as an individual adaptation?

This is not just, as Dawkins has accused, merely “semantic doubletalk”. Empirical results validate the multi-level selection perspective, and show how focusing solely on genes can only give an incomplete perspective on evolution. For instance, William Muir has demonstrated that artificial group selection of hens results in decreased mortality of the birds, and increased egg production, with the bulk of the improvement occurring within just three generations. As another example, William Swenson, David Sloan Wilson and Robert Elias showed that artificial group selection of microbial ecosystems in soils lead to improved growth for plants grown in that soil. Far from group selection being too weak a force to have any serious influence, it seems it is strong enough to have dramatic and demonstrable effects.

However, if Kuhn taught us anything about the history of science it is that research communities do not change their beliefs without a fight – often, you might say, to the death. Hamilton, however, who did much of the heavy lifting on kin selection and had been originally opposed to the viability of group selection, changed his view almost completely as a result of a fascinating equation by George R. Price that, alas, is slightly too complex to explain here. As Wilson, notes, Hamilton was happy to change his mind once he recognized that “inclusive fitness theory is not an alternative to group selection after all; the role of group selection was merely obscured by the way it was formulated.”

What really put the cat among the pigeons in recent years was a paper by Martin Nowak, Corina Tarnita and Edward O. Wilson that denies the validity of using kin selection to explain the origins of the behaviour of social insects such as ants. Edward O. Wilson is particularly famous in evolutionary circles as the founder of sociobiology, not to mention an expert on insect species of all kinds, and his apparent change of heart in respect to the validity of group selection raised both eyebrows and tempers. In fact, he has been moving in this direction for some time, and published several earlier papers dealing with the topic, all stating the same thing: kin selection wasn’t a significant factor in the evolution of ant social behaviour, while selection between colonies (i.e. multi-level selection at the level of the colony) appears to have played a highly significant role.

Ants and wasps had generally been considered a great example of the merits of the inclusive fitness perspective advanced in kin selection, owing to a rather unique system of inheritance. Whereas most animals pass on half of their genes to their offspring (hence children share half their genes with each of their parents), many social insect species have females with genes from both parents and males with genes only from the mother (the queen). Since the queen passes on all the genes to the males, this means that whichever male the queen mates with, her daughters will have three quarters of their genes in common with their mother queen. In terms of Haldane’s jest, this means that four ant sisters will sacrifice themselves for three queens – a pretty good rate of exchange! These numbers were sometimes claimed to be a validation of inclusive fitness, serving to explain why ants and bees had such tight social structures.

Trouble is, this explanation never actually stacked up in practice. For a start, remember that the formula for kin selection, Hamilton’s rule, is only a description of optimal conditions – it isn’t actually an explanatory mechanism. As Christopher Jensen has noted, it’s rather remarkable that no-one has incorporated Hamilton’s ideas into a predictive scientific model that demonstrates how these insect species reach the optimal state predicted. In the absence of such a model, as Nowak and colleagues suggest, the high relatedness might be better understood as a consequence of the highly social living arrangements, not as a cause.

What’s more, as Raghavendra Gadagkar reports, the actual level of relatedness in insect colonies is considerably lower than the ideal values suggest, not to mention there are a wide variety of social insects that don’t have the unusual inheritance pattern described above. Gadagkar suggests that an exclusive focus on relatedness isn’t good enough for proving Hamilton’s rule; the benefits and costs have to be measured in some way, not simply brushed under the carpet. His own research group has attempted to find ways to attend to this deficit, and when they have done so they have concluded that “ecological, demographic and physiological factors” have a bigger role in bringing about the unique social arrangements of communal insects.

Both Jensen and Gadagkar, commenting on the controversial Nowack, Tarnita, and Wilson paper, reach similar conclusions: kin selection and group selection aren’t really in competition with each other, they are simply different tools that need to be applied in different ways. Jensen notes in particular that there are situations in which Hamilton’s rule can be useful – it just shouldn’t be taken as the default explanation. Furthermore, even if group selection is a better explanatory device in respect of social insects, “we need to acknowledge that different theoretical constructs will be useful to explain different evolutionary phenomena.”

Regarding the intensity of the backlash against the paper, Jensen recognises that there are evolutionists who are “wholly committed to the idea of gene-level selection” and who “insist that all arguments rest on the tenet that the gene is the sole unit of selection”. He comments that the paper has thus “exposed some very disturbing things about the sociology of evolutionary biology”:

This publication has really angered a great number of people, and this anger has revealed the kind of tribal orthodoxy that exists in the mainstream of evolutionary biology. A great number of evolutionary biologists will become enraged whenever someone in the field espouses a hypothesis that contradicts the gene-centred approach to explaining evolutionary processes... the anger often seems excessive in relation to the affront. After all, if an idea is clearly dumb and can easily be dispatched by theoretical or empirical demonstration, why get upset about it? We need more publication of relevant results, not more rhetoric.

It is worth reiterating that the gene’s eye view is a perfectly workable scientific metaphor. It only creates problems when it develops into a monomaniacal fixation with genes. Yes, genes are absolutely crucial to understanding the evolution of life, because without changes in the abundance of different gene variants much of what we consider natural selection simply has too little to work with. But when it comes to the evolution of co-operation, the kin selection perspective has radically less to offer than multi-level selection theory.

Co-operation allows groups of animals – even unrelated animals – to gain a significant edge over rival groups that don’t co-operate, or who co-operate less well. This situation can lead to the evolution of group adaptations for co-operation, as appears to have happened with social insects. And of course, even without group selection for collaboration, co-operation can still be the best strategy. Rather than the myth that co-operation is constrained to occur only between relatives, we need a new myth that expresses the inherent advantages of working together.

Extracted from the draft manuscript of Myths of Evolution, due from Zero Books in 2012.


Feed You can follow this conversation by subscribing to the comment feed for this post.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Your Information

(Name is required. Email address will not be displayed with the comment.)