The Units of Selection Debate

This post will be the first is a short series dedicated to the units of selection debate in evolutionary biology.  In this post I will recount the history of the debate including the major trend setters and the arguments which they presented for their own ideas concerning the level at which natural selection acts.  This will set the stage for later posts in which I will present my own positions on this as well as other related matters in biology and evolution.

Central to Charles Darwin’s Origin of Species is the idea that natural selection operates at the level of the individual organism.  It is individual organisms which are selected for breeding in artificial selection and since his version of natural selection was largely and argument from analogy with artificial selection it too was an account of individual selection.  More specifically, due to the geometric increase of individual organisms combined with the inevitably limited supply of resources, some individual organisms would tend to be more fit due, presumably, to the traits which they had inherited from their parents and who therefore have more access to resources.  Darwin did, however, find some exceptions to this rule of individual selection primarily in the colonies of the social insects.  He reconciled this discrepancy by suggesting that the entire colony acts as if it were one single organism.

This suggestion, however, paved the way for other forms of group selection.  Notice that it is due to the “altruistic” nature of insect colonies, in which workers are working not for the reproduction of their sterile selves but for the “good of the colony” that forced Darwin to posit a form of group selection.  This rationale simply opened the doors to more accounts of group selection as more and more account of natural altruism came pouring in.  Whenever there was an individual sacrificing itself in some way for the good of the group, there group selection was supposed to be happening.

In 1966 George Williams published his Adaptation and Natural Selection, in which he argued that parsimony dictates against explanations in terms of group selection.  As I have recently argued, since individual selection is always at work due to the limitation in resources, one can never have group selection without individual selection as well.  Thus, if a phenomenon can be explained completely and solely in terms of individual selection, then adding group selection to the account contributes no explanatory power whatsoever.  Thus, in his book Williams argued that individual or gene selection (he vacillates between the two through out his book) should be the primary explanation and that group selection should only be appealed to as a method of last resort.

A decade later, in his book The Selfish Gene, Richard Dawkins took Williams’ account even further.  In contrast to Dawkins, Williams wavered between individual selection and gene selection while allowing for the rare occurrence of group selection as well.  Dawkins, on the other hand, argued that selection only happens at the level of the gene and seeing things from the gene’s-eye-view can account for all forms of both individual selection as well as group selection.  Genes, and not organisms or groups, were the only biological entities which had the longevity, fidelity and fecundity of replication which was essential to the selective process.

Throughout the 1980’s, however, Sober and Lewontin argued strongly and repeatedly against the views of Williams and Dawkins.  Viewing natural selection from the perspective of the individual genes does not take into account the important roles which the gene-pairs play in nature.  Consider the well known case of the sickle cell anemia gene in which having one set of the gene in question, which is recessive, gives the carrier a protection from malaria but having both pairs gives the carrier sickle cell anemia.  What matters here is not the fitness of the individual gene, but rather the fitness of the gene-complex or gene-pair.

During the 1990’s group selection also saw a bit of a comeback primarily in the work of David Sloan Wilson who developed a number of statistical models in which altruism in individual organisms could arise under some circumstances due to group selection.  His primary argument was that individual selection has only beat out group selection by mere definition by the methods which biologists adopt in their account.  When the biologist considers fitness across the entire population rather than across individual groups within the population, individual selection wins by fiat.  Such an approach does not even allow for the possibility of group selection for there are no groups under such a view.  Thus, to say that selection operates on the individual rather than the group in such a context is not to say anything of interest at all.

In the 21^st Century other efforts have been developed to overthrow the gene’s-eye-view account of natural selection.  Jablonka and Lamb, for example, argue that gene selectionism all but ignores the importance of developmental biology.  They claim that the latter demonstrates that what matters are not individual genes or even individual gene-complexes but rather the traits which such things partially contribute to in the developmental stages of the organism.  Gene development, they argue, is a largely canalized process which is influenced by numerous gene-complexes which are scattered through out the genome, epigenetics and other environmental factors; the isomorphic relation which Dawkins’ model seems to presuppose simply does not exist.  Since traits can change without any corresponding change in the genome and the genome can change without any corresponding change in the traits, the gene’s-eye-view is inadequate.

In the next post, I will consider the claim of C. Kenneth Waters that each of these accounts is, or at least could be just as true as the others.  Each account is simply another way of carving up the same world into replicators and an external environment.  I will save a more detailed description for then.