Re: [evol-psych] RE: Sober on Dawkins (and unit of selection)
- Elizabeth Lloyd:
The four different questions I define are:
The replicator question: Which entity passes on its structure directly in replication?
(usually, but not always, the gene).
The interactor question: Which entities are being directly selected at a given level such that
replication is differential?
The beneficiary question: Which entity benefits, in the long run, from the evolution-by-
The manifestor-of-adaptation question: which entity at a given level of organization
manifests an engineering adaptation?
This taxonomy also disentangles the very long-running battle between John Maynard
Smith and the American group selectionists, Michael Wade and DS Wilson. This is one
reason JMS liked it. Maynard Smith had always defined group selection as a
*combination*: group selection occurred when groups were interactors AND manifestors
of a group-level adaptation (i.e., an individual self-sacrificial trait). But this isn't a proper
definition of group selection, in that it isn't parallel to the definition of organismic
selection, and it wasn't the one that Wade and Wilson were using. Wade and Wilson were
using a *simple*, one-part definition of group selection: group selection occurred when
groups were interactors. Full stop.
I would think that put that way, it is obvious that group selection exists; groups die out differently, split up at different speed, send propagules at different speed, etc, then group selection exists necessarily. The same for species, the same for individuals. The interesting question is what bearing this has in the evolution of adaptation? Can group adaptations evolve because of group selection? All group selectionists (Wynne-Edwards, Wade, Wilson, Sober) say the answer is yes - and the most discussed example of group adaptation is self-sacrifice by a group member. Then, calling group selectionists to those that just believe that groups die and reproduce differently implies that *everybody* is a group selectionist.
Maynard Smith could never disagree that group selection exists; his disagreement was on the importance of it to evolution, and especially to the evolution of group adaptations. He doubted that it was an important force, because D. S. Wilson model needs: i) That groups have important genetic differences, and there is group heritability; ii) That these differences are strongly correlated with the survival of the groups; iii) That migration between groups is low; iv) That within group genetic variation is low.
In parasites infecting hosts, the above mentioned factors seem more likely true - hosts are "islands" more than "villages", isolation is strong. Then, I think there is a consensus that group selection (called "between host selection" in the trade) is what counts more for adaptation of parasites. Parasitic adaptation is more likely to promote the success of the whole bunch (and there is mounting evidence that it does, for example, *some* cholera bacteria promote host diarrhea through a toxin, which helps the group as a whole, even non-toxin producing bacteria). However, even in this context, the concept of group selection was attacked: Alexander & Borgia, in their 1978 paper said these cases are better understood in the realm of "superorganisms" - that is, the bunch of parasites is a superorganism, with division of labour, as in social insects, or the cells of a vertebrate, so between host selection can be usefully described as *individual selection* and not group selection. Responding to this, Sober & Wilson (1998) complained that many people are doing whatever they can to dismiss group selection!
Alexander, R.D. and Borgia, G. (1978) Group selection, altruism and the levels of organization of life. Ann. Rev. Ecology Systematics 9, 449-474
Sober, E. & Wilson, D. S. (1998) Unto Others the Evolution and Psychology of Unselfish Behavior, Cambridge MA: Harvard University Press