The Meaning of Human Existence (6 page)

At this point, prior to developing the theories further, it will be instructive to take a specific example in the evolution of social behavior and see how it is treated respectively by each approach.

The life cycle of ants has always been a favorite of inclusive fitness theorists as offering proof of the role of kinship and the validity of inclusive fitness. Many ant species have the following life cycle: their colonies reproduce by releasing virgin queens and males from the nest. After mating, the queens do not return home, but disperse to establish new colonies on their own. The males die within hours. The virgin queens are much larger than the males, and colonies invest a correspondingly larger fraction of their resources to their production.

The inclusive fitness explanation of the size difference between the sexes, introduced in the 1970s by the biologist Robert Trivers, is as follows. The means of sex determination in ants is peculiar, such that sisters are more closely related to one another than they are to their brothers (providing the queens mate with only one male). Because the workers raise the young, Trivers continued, and because they favor sisters over brothers, they invest more in virgin queens than in males. The colony, with workers in control, accomplish this end by making the queens individually much larger in size. This process deduced with inclusive fitness theory is called indirect natural selection.

The standard population genetics model, in contrast, posits direct natural selection and tests it with direct observation in the field and laboratory. The larger size of
the virgin queen is necessary, as all entomologists know, because of the way she starts a new colony. She digs a nest, seals herself in, and raises the first brood of workers on her large bodily reserves of fat and metabolized wing muscles. The male is small because its only function is to mate. After achieving insemination, it dies. (Queens live on in a few species, incidentally, for more than twenty years.) The roundabout inclusive fitness explanation for investments according to gender is therefore wrong.

The assumption of inclusive fitness theory that workers control the colony’s allocation, a crucial point in this reasoning, is also wrong. Using the valve on her spermatheca, the bag-like organ in which the sperm are stored, the queen determines the sex of the offspring born. If a sperm is released to fertilize an egg in the queen’s ovary, a female is born. If no sperm is released, the egg is not fertilized, and from the unfertilized egg a male is born. Thereafter, a complex of factors, only some of which are under worker control, determine which female eggs and larvae will become queens.

For half a century, while data were still relatively scarce, the theory of inclusive fitness was the prevailing explanation of the origin of advanced social behavior. It began in 1955 with a simple mathematical model by the British geneticist J. B. S. Haldane. His argument was in the following form (which I’ve altered here a bit to
make it intuitively easier). Imagine that you are a childless bachelor standing on a riverbank. Looking out over the water, you see that your brother has fallen in and is drowning. The river that day is raging, and you’re a poor swimmer, so you know that if you jump in and save him, you yourself will probably drown. So the rescue requires altruism on your part. But (Haldane said) it does not also require altruism on the part of your genes, including those responsible for making you altruistic. The reason is the following. Because the man is your brother, half of his genes are identical to yours. So you jump in, save him, and sure enough, you drown. Now you’re gone, but half of your genes are saved. All your brother has to do in order to make up the loss in genes is to have two additional children. The genes are the unit of selection; the genes are what count in evolution by natural selection.

In 1964, another British geneticist, William D. Hamilton, expressed Haldane’s concept in a general formula, which came to be known in later years as the Hamilton inequality. It said that a gene prescribing altruism, such as that of the heroic brother, will increase if the benefit in number of offspring to the recipient exceeds the cost in offspring to the altruist. However, this advantage to the altruist will be effective only if the recipient and the altruist are closely related. The degree of kinship is the fraction of genes that are shared by the altruist
and recipient due to their common descent: one-half between siblings, one-eighth between first cousins, and so on in a rapidly declining rate as the degree of kinship becomes more distant. The process later came to be called kin selection. It seemed, at least from this line of reasoning, that close kinship is the key to the biological origin of altruism and cooperation. Hence close kinship is a primary factor of advanced social evolution.

On the surface, kin selection seemed at first to be a reasonable explanation for the origin of organized societies. Consider any group of individuals that have come together in one manner or another but remain unorganized—a fish school, for example, a flock of birds, or a local population of ground squirrels. The group members, let us say, are able to distinguish not just their own offspring, leading to evolution of parental care by standard (Darwinian) natural selection. Suppose they also recognize collateral relatives related by common descent such as siblings and cousins. Allow further that mutations occur that induce individuals to favor close collateral relatives over distant relatives or nonrelatives. An extreme case would be Haldane’s heroism biased toward a brother. The result would be nepotism, resulting in a Darwinian advantage over others in the group. But where does that lead an evolving population? As the collateral-favoring genes spread, the group would
change into an ensemble not of competing individuals and their offspring, but of an ensemble of parallel competing extended families. To achieve group-wide altruism, cooperation, and division of labor, in other words organized societies, requires a different level of natural selection. That level is group selection.

Also in 1964, Hamilton took the kinship principle one step further by introducing the concept of inclusive fitness. The social individual lives in a group, and it interacts with other members of the group. The individual participates in kin selection with each of the other group members with which it interacts. The added effect it has on its own genes passed into the next generation is its inclusive fitness: the sum of all the benefits and costs, discounted by the degree of kinship with each other group member. With inclusive fitness the unit of selection had passed subtly from the gene to the individual.

At first I found the theory of inclusive fitness, winnowed down to a few cases of kin selection that might be studied in nature, enchanting. In 1965, a year after Hamilton’s article, I defended the theory at a meeting of the Royal Entomological Society of London. Hamilton himself was at my side that evening. In my two books formulating the new discipline of sociobiology,
The Insect Societies
(1971) and
Sociobiology: The New Synthesis
(1975), I promoted kin selection as a key part of the
genetic explanation of advanced social behavior, treating it as equal in importance to caste, communication, and the other principal subjects that make up sociobiology. In 1976 the eloquent science journalist Richard Dawkins explained the idea to the general public in his best-selling book
The Selfish Gene
. Soon kin selection and some version of inclusive fitness were installed in textbooks and popular articles on social evolution. During the following three decades a large volume of general and abstract extensions of the theory of kin selection was tested, especially in ants and other social insects, and purportedly found proof in studies on rank orders, conflict, and gender investment.

By 2000 the central role of kin selection and its extensive inclusive fitness had approached the stature of dogma. It was a common practice for writers of technical papers to acknowledge the truth of the theory, even if the content of the data to be presented were only distantly relevant to it. Academic careers had been built upon it by then, and international prizes awarded.

Yet the theory of inclusive fitness was not just wrong, but fundamentally wrong. Looking back today, it is apparent that by the 1990s two seismic flaws had already appeared and begun to widen. Extensions of the theory itself were growing increasingly abstract, hence remote from the empirical work that continued to flourish elsewhere
in sociobiology. At the same time the empirical research devoted to the theory remained limited to a small number of measurable phenomena. Writings on the theory mostly in the social insects were repetitive. They offered more and more about proportionately fewer topics. The grand patterns of ecology, phylogeny, division of labor, neurobiology, communication, and social physiology remained virtually untouched by the asseverations of the inclusive theorists. Much of the popular writing devoted to it was not new but affirmative in tone, declaring how great the theory was yet to become.

Inclusive fitness theory, fondly called IF theory for short by its defenders, was showing increasing signs of senescence. By 2005 questions about its soundness were being openly expressed, especially among leading experts on the details of the biology of the ants, termites, and other eusocial insects, as well as a few theoreticians bold enough to seek alternative explanations of the origin and evolution of eusociality. The researchers most committed to IF theory either ignored these deviations or summarily dismissed them. By 2005 they had gained enough representation in the anonymous peer review system to hinder publication of contrary evidence and opinions in leading journals. For example, a keystone early support of inclusive fitness theory, cited in textbooks, was the prediction of overrepresentation of the
Hymenoptera (bees, wasps, ants) among eusocial animal species. When after a time one investigator pointed out that new discoveries had nullified the prediction, he was told, in effect, “We already knew that.” They did know that, but hadn’t done more than just drop the subject. The “Hymenoptera hypothesis” was not wrong; it had just become “irrelevant.” When a senior investigator used field and laboratory studies to show that primitive termite colonies compete with one another and grow in part by the fusion of unrelated workers, the data were rejected on grounds that the conclusion did not adequately take into account inclusive fitness theory.

Why did an outwardly arcane topic of theoretical biology excite such fierce partisanship? Because the problem it addresses is of fundamental importance, and the stakes in trying to solve it had become exceptionally high. Furthermore, inclusive fitness was beginning to resemble a house of cards. To pull even one out risked collapsing the whole. Pulling cards, however, was worth the price to reputation. There existed in the air the promise of a paradigm shift, a rare event in evolutionary biology.

In 2010, the dominance of inclusive fitness theory was finally broken. After struggling as a member of the small but still muted contrarian school for a decade, I joined two Harvard mathematicians and theoretical biologists, Martin Nowak and Corina Tarnita, for a top-to-bottom
analysis of inclusive fitness. Nowak and Tarnita had independently discovered that the foundational assumptions of inclusive fitness theory were unsound, while I had demonstrated that the field data used to support the theory could be explained equally well, or better, with direct natural selection—as in the sex-allocation case of ants just described.

Our joint report was published on August 26, 2010, as the cover article of the prestigious journal
Nature
. Knowing the controversy involved, the
Nature
editors had proceeded with unusual caution. One of them familiar with the subject and the mode of mathematical analysis came from London to Harvard to hold a special meeting with Nowak, Tarnita, and myself. He approved, and the manuscript was next examined by three anonymous experts. Its appearance, as we expected, caused a Vesuvian explosion of protest—the kind cherished by journalists. No fewer than 137 biologists committed to inclusive fitness theory in their research or teaching signed a protest in a
Nature
article published the following year. When I repeated part of my argument as a chapter in the 2012 book
The Social Conquest of Earth
, Richard Dawkins responded with the indignant fervor of a true believer. In his review for the British magazine
Prospect
, he urged others not to read what I had written, but instead to cast the entire book away, “with great force,” no less.

Yet no one since that time has refuted the mathematical analysis by Nowak and Tarnita, or my argument favoring the standard theory over inclusive fitness theory in the interpretation of field data.

In 2013 Nowak and I were joined by another mathematical biologist, Benjamin Allen, in a still deeper expansion of the ongoing analysis. (Tarnita had moved to Princeton, where she was busy adding field research to her mathematical modeling.) In late 2013 we published the first in a planned series of refereed articles. Because of the need for exactitude, and the material that these articles contain that may be relevant to the history and philosophy of the subject, I’ve taken the step of providing a simplified summary of the first one in the appendix of this book.

Now at last we can return to a key question in a more open spirit of inquiry: What was the driving force in the origin of human social behavior? The prehumans of Africa approached the threshold of advanced social organization in a manner parallel to that in the lower animals but attained it in a very different manner. As brain size more than doubled, the bands used intelligence based on vastly improved memory. Where primitively social insects evolved division of labor with narrow instincts that play upon categories of social organization in each group, such as larvae and adults, nurses and foragers, the
earliest humans operated with variable instinct-driven behavior that made use of detailed knowledge of each group member by all the others.