An extract on #forestlover
A good deal of objection to The Selfish Gene stemmed from its failure to be always clear about "selection" and "replication". As pointed out in the subsection Units of selection, Dawkins says the gene is the fundamental unit of selection, and then points out that selection doesn't act directly upon the gene, but upon 'vehicles' or 'extended phenotypes'. Stephen Jay Gould took exception to calling the gene a 'unit of selection' because selection acted only upon phenotypes. Summarizing the Dawkins-Gould difference of view, Sterelny says:
"Gould thinks gene differences do not cause evolutionary changes in populations, they register those changes."
Kim Sterelny: Dawkins vs. Gould, p. 83
The word 'cause' here is a bit tricky: does a change in lottery rules (for example, inheriting a defective gene "responsible" for a disorder) 'cause' differences in outcome that might or might not occur? It certainly alters the likelihood of events, but a concatenation of contingencies decides what actually occurs. Dawkins thinks the use of 'cause' as a statistical weighting is acceptable, and it is a common though perhaps not a technical usage. Like Gould, Gabriel Dover in criticizing The Selfish Gene says:
"It is illegitimate to give 'powers' to genes, as Dawkins would have it, to control the outcome of selection...There are no genes for interactions, as such: rather, each unique set of inherited genes contributes interactively to one unique phenotype...the true determinants of selection".
Gabriel Dover: Dear Mr. Darwin, p. 56
However, from a comparison with Dawkins' discussion of this very same point, it would seem both Gould's and Dover's comments are more a critique of his sloppy usage than a difference of views. Hull suggested a resolution based upon a distinction between replicators and interactors. The term 'replicator' includes genes as the most fundamental replicators but possibly other agents, and interactor includes organisms but maybe other agents, much as do Dawkins' 'vehicles'. The distinction is as follows:
replicator: an entity that passes on its structure largely intact in successive replications.
interactor: an entity that interacts as a cohesive whole with its environment in such a way that this interaction causes replication to be differential.
selection: a process in which the differential extinction or proliferation of interactors causes the differential perpetuation of the replicators that produced them.
Hull suggests that, despite some similarities, Dawkins takes too narrow a view of these terms, engendering some of the objections to his views. According to Godfrey-Smith, this more careful vocabulary has cleared up "misunderstandings in the "units of selection" debates."