Battle of the Titans. Column in ComputerreOnline #70

Richard Dawkins and Edward Wilson. Don’t you think they look alike? Yet their temperaments differ…
I would like to discuss the conflict between these scientists with the utmost respect for both. I must say that the texts of both Dawkins (whom I have read considerably more) and Wilson do not receive my full agreement, but it is clear that these people belong to the thin layer of the intellectual elite that drives the growth of human understanding. After Wilson announced the need to revise established scientific concepts that he helped create, Dawkins pounced on him with harsh criticism. In Dawkins’s view, Wilson’s latest book and his current ideas should simply be discarded. Wilson had not entered the debate for long, but he did not miss the chance to strike back at Dawkins as hard as possible. He pointed to the low qualification of his opponent (a “good‑ish” person, writes popular books, but does not publish in peer‑reviewed journals and does not truly understand the problem…).
The main problem of the science founded by Wilson is the study of the mechanisms underlying the origin of genuine sociality (and its characteristic altruistic behavior) in various animals—from insects to humans. Since we are dealing with the evolution of organisms whose behavior is largely genetically programmed (in the case of humans this claim had to be proved to Wilson, but he has no contradictions with Dawkins on this point), the problem can be described in genetic terms. What mechanisms have promoted the evolution of a genetic predisposition to altruism?
To answer this question, we should consider what parameter is maximized during evolution. The traditional answer is fitness. Unfortunately, this concept is very complex; its excellent analysis is presented by Dawkins in Chapter 10 of “The Extended Phenotype”. There, five different interpretations of the term “fitness” are compared, and the chapter is titled “A Headache in Five Manifestations” (in the translation published by the “Astel” publishing house). For our discussion two manifestations are important, i.e., interpretations: classical fitness (No. 3 in Dawkins) and inclusive fitness (No. 4).
Classical fitness (developed from the ideas of Spencer, Darwin, and Wallace) is the expected or actual number of offspring of an individual that survive to a certain age. Inclusive fitness (proposed by Hamilton) is the number of an individual’s own offspring plus the effect of that individual on the number of offspring of its relatives (taking relatedness into account).
Let us illustrate with a hypothetical example of humans. Individual M had two children and persuaded his sister to have an abortion when she became pregnant by a scoundrel. Individual N was infertile, but he extinguished a burning house belonging to his brother, thereby saving his brother’s three children.
Calculate classical fitness. M has two children, each of which is related to him by ½ (that is the proportion of shared genes). Thus, M’s fitness = 1. N has no children, so his fitness = 0.
Inclusive fitness is more complicated. The relatedness between M and his sister is ½, between the sister and her unborn child is also ½, therefore the relatedness between M and the child he prevented from being born is ¼. M’s inclusive fitness = ½ + ½ − ¼ = ¾. N, however, ensured the survival of three children, each of whom he is related to by ¼, so his inclusive fitness is also ¾. If N had saved four children (or if M had succeeded in sending his sister to a monastery and she completely gave up childbearing), it would turn out that the childless N contributed more to the next generation than the father of two children M!
Which approach is more correct? Within their own logic both are correct. But note: over time, under certain conditions, the proportion of individuals related to N‑type individuals may increase in the population, rather than those related to M‑type individuals. In that case a measure emphasizing the fitness (potential evolutionary success) of individual N would be more appropriate.
Classical fitness is a measure of the efficiency of individual selection, discovered by Darwin and Wallace. Inclusive fitness reflects the action of kin selection, Hamilton’s brainchild.
Now (drum roll!) the trump argument in favor of kin selection. Hamilton explained why many hymenopterans—bees, wasps, and ants—often display true sociality and altruistic behavior. They have a specific sex‑determination mechanism. A female can lay unfertilized (haploid) eggs that develop into males, and fertilized (diploid) eggs that develop into females. The relatedness between a mother and her daughters is ½, while the relatedness among sisters is ¾ (the father has a single set of chromosomes and all his daughters receive that set completely). This means that a female will ensure better survival of her own genes if she cares for her sisters rather than for her daughters!
Recall the example of gentlemen M and N. If it turned out that nephews are genetically closer to us than our own children (as is the case in bees), then, from the standpoint of inclusive fitness, it would be more advantageous to care for them rather than for one’s own offspring!
I will now temper the just‑presented argument. Not all hymenopterans with such a haplo‑diploid sex‑determination system are social. Moreover, not all eusocial species possess this system. For example, sterile individuals caring for their sisters and brothers are also found in termites, some shrimp, beetles, and even mammals (naked mole‑rats). By the way, although human societies are not maintained by sterile individuals, our level of sociality (and altruism) is also extremely high and requires explanation. Yet, with these caveats, one can hypothesize that haplo‑diploid sex determination promotes eusociality, supporting the plausibility of kin selection and inclusive fitness.
Let us take another important step. What happens during reproduction? New individuals appear—somewhat, not always exact, copies of the previous ones. They carry genes—somewhat, not always exact, copies of parental genes. Population reproduction can be viewed both as the reproduction of individuals and as the reproduction of genes! Are these approaches equivalent?
At the level of gene reproduction, both individual and kin selection can be successfully described. At the level of individual reproduction—only individual selection. A sterile individual that gave life to its nephews did not reproduce. Only its genes were reproduced (their copies increased).
From this arose the concept of the selfish gene. It was not Dawkins who invented kin selection and inclusive fitness, but he was the first to loudly proclaim that evolution is not the improvement of individuals, but a competition of genes in self‑replication!
Can the emergence of altruistic behavior be explained without reference to selfish genes? Such attempts have been made repeatedly. I have already written about the idea of group selection, originally proposed by Darwin. Some arguments of its supporters (for example, Vero Corner Vinn‑Edwards, who tried to revive it in 1962) were refuted, some remained valid. Even if altruists lose within groups, the preferential growth of groups that contain them can promote the spread of altruistic behavior via Simpson’s paradox. Dawkins and proponents of the selfish gene spent much effort debating Stephen Gould (1941‑2002), who argued that selection can operate at multiple levels. Dawkins and Gould were contemporaries, but Gould had a shorter lifespan. And when the battlefield was left to Dawkins, unexpected blows to kin selection and the selfish gene were delivered by the classic sociobiologist Wilson.
What does he do? He insists that the emergence of eusociality can be explained without kin selection. In 2007 Edward Wilson, together with his namesake David Sloan Wilson, published an article supporting the idea of group selection (the article is discussed in Elena Kleshchenko’s piece for “Chemistry and Life”, posted on “Elements”). In 2010, together with mathematicians Martin Nowak and Carina Tarnuta, he wrote a paper for *Nature* demonstrating that the origin of eusociality does not require references to kin selection. That paper was the subject of a news item on the same “Elements”, written by Alexey Merkuryevich Gilyarov. Finally, in 2012 Wilson released the book *The Social Conquest of Earth*, where he continues to persist in this heresy…
I will try to briefly characterize Wilson’s logic as I understand it. Before kin selection can support the spread of altruism genes, those genes with their characteristic effects must first appear. They will appear only in those species for which the formation of groups of tightly interacting individuals is typical. Such groups most often initially consist of more or less close relatives. Groups in which altruists appear gain an advantage and grow faster. Moreover, this increases the number of genes characteristic of altruists, because an individual’s genes do not appear out of nowhere; they are part of the population’s gene pool. Thus, group selection in a kin‑related group simultaneously works as kin selection. If a group is not kin‑related, we will be dealing only with group selection.
Are there situations where only kin selection works and group selection does not? Yes. They can be imagined or deliberately constructed. But, according to Wilson, such situations are not characteristic of the stages through which eusociality and altruism develop…
Could Dawkins have refrained from attacking Wilson? Could Wilson have yielded?
As the epigraph makes clear, I think that, to some extent, both opponents are right. How the correctness of one aligns with the correctness of the other is a matter for another time.