Three graces, p.15
War and Peace and War, page 15
Generally speaking, the average offer amount among the 15 societies was correlated with the probability of rejecting a low offer. But there was one exception. The overwhelming proportion of offers rejected by the Au and the Gnau of New Guinea were those that were high (greater than half of the pot). Among these groups, accepting a large gift puts the receiver under a strong obligation and into a subordinate position with respect to the giver. As a result, such gifts will be frequently refused, and this cultural quirk was reflected in the way people played the ultimatum game. Another example of how culture can be reflected in the manner with which people behave in the experimental setting is the Orma. This herding people in Kenya practice wide-scale cooperation. When they decide to build a new school or a road, members of the community are asked to contribute, with the wealthier (those who have larger herds) contributing proportionally more. This system is called harambee. When the public goods game was explained to the Orma, they quickly dubbed it a harambee game. Interestingly, their contributions in the game were strongly correlated with their real-world wealth, as it would be in harambee.
THE BEHAVIORAL EXPERIMENTS USING THE public goods and the ultimatum games decisively prove that Machiavelli’s self-interest premise was wrong. It is simply not true that all people behave in entirely self-interested manner. Some people—the knaves—are like that. However, other kinds of people, whom I have called the saints and the moralists, behave in prosocial ways. Furthermore, different societies have different mixtures of self-interested and cooperative individuals. Cultural practices (for example, the harambee system) and social institutions have a strong effect on whether and how collective action can be sustained.
The experiments also point to the key role of the moralists. Kindly saints are completely ineffectual in preventing cooperation from unraveling. In the absence of effective sanctions against free-riders, opportunistic knaves waste any contributions by the saints to the common good. Self-righteous moralists are not necessarily nice people, and their motivation for the “moralistic punishment” is not necessarily prosocial in intent. They might not be trying to get everybody to cooperate. Instead, they get mad at people who violate social norms. They retaliate against the norm breakers and feel a kind of grim satisfaction from depriving them of their ill-gotten gains. It’s emotional, and it’s not pretty, but it ensures group cooperation.
A recent experiment, conducted in Zurich by Fehr and colleagues, confirms that emotions play a strong role in moralistic punishment. As in other studies, the experiment involved human subjects playing a variant of the public goods game. The new twist was that the researchers scanned brain activity of the subjects who were contemplating whether to punish a cheater. The brain scan showed that when a player was deciding on punishment, a spike of neural activity occurred in the region of brain known as the caudate nucleus. The stronger the nucleus fired, the greater was the fine imposed by the subject on the norm violator. The caudate nucleus is known to be involved in the processing of rewards. More specifically, it has an important role in integrating reward information with goal-oriented behavior. The implication of the experimental findings, therefore, is that the subjects were anticipating the satisfaction of punishing the cheaters. The individuals who derived more pleasure from revenge (whose caudate nuclei fired particularly intensely) were willing to fork out more dollars to impose a heavier fine on the cheater.
Other groups of researchers in New Jersey, Texas, and California are also busily scanning brains of experimental subjects contemplating social choices—to trust or not? To punish or not? A whole new hot discipline, called neuro-economics, was born in the last few years. The results churned out by different research groups are striking. They indicate that the capacity for trust and moralistic punishment are wired into our brains. At some level, they are as basic as our abilities of obtaining food or finding mates. It does not mean that all humans will always behave in a cooperative manner. People are different—some are knaves, others moralists. Societies differ in their ability to sustain collective action. But the capacity for cooperation (even if it is never exercised by many people) is part of what makes us human. Machiavelli was wrong.
WHAT DOES IT MEAN FOR OUR theories of social and economic dynamics? First, there is no need to flush the theory of rational choice down the drain. Science typically advances in an incremental way, by building on earlier successes, and the development of our ideas about cooperation is not an exception to this principle. Remember that the theory assumes that rational agents maximize a “utility function,” which in the general case need not be only material self-interest. It is quite straightforward to include in the utility function the prosocial inclinations of some individuals, and in fact theoretical economists have already started doing that. This research shows that in some kinds of circumstances prosocial inclinations do not matter, and we obtain results identical to the classical theory built on the self-interest assumption. Take price formation in the markets, one of the greatest successes of the classical theory, as developed by Adam Smith. Moralistic individuals may bitch and moan about not getting the “fair price,” but the invisible hand of the market will inexorably set the price at the level determined by supply and demand.
In other kinds of situations, prosocial norms held by part of the population will result in an outcome completely at variance with the standard theory. Take the example with which this chapter started. From the point of view of the self-interest hypothesis, massive volunteering for the army is incomprehensible. But being familiar with the results from the public goods game and similar experiments, it is easy to understand what went on. The initial surge of volunteers must have been entirely saints and moralists. After enough moralists joined up, they put a lot of pressure on the knaves to do the same. In England during World War I, particularly effective were the female moralists who could not serve themselves but had husbands and sons in the army. Here’s an eyewitness account by William Brooks, who joined the army in 1915: “Once war broke out the situation at home became awful, because people did not like to see men or lads of army age walking about in civilian clothing, or not in uniform of some sort, especially in a military town like Woolwich. Women were the worst. They would come up to you in the street and give you a white feather, or stick it in the lapel of your coat. A white feather is the sign of cowardice, so they meant you were a coward and that you should be in the army doing your bit for king and country. It got so bad it wasn’t safe to go out. So in 1915 at the age of 17 I volunteered under the Lord Derby scheme. Now that was a thing where once you applied to join you were not called up at once, but were given a blue armband with a red crown to wear. This told people that you were waiting to be called up, and that kept you safe, or fairly safe, because if you were seen to be wearing it for too long the abuse in the street would soon start again.”
Lest the import of this quote be misinterpreted, it is worth reiterating that pure coercion is not the explanation of why the British enlisted in huge numbers in the World War I. Some undoubtedly were forced to enlist by the tactics described in the previous paragraph, but others enlisted out of sheer patriotism; yet others were motivated by a mixture of patriotism and fear of harassment. Furthermore, the harassment of shirkers by the British women is emphatically not a self-interested behavior. By inducing another man to enlist, they did not better the survival chances of their husbands or sons in any appreciable way. These women put pressure on men to enlist not out of self-interest, but because of the expectations of their society at large, or more precisely, social norms.
Cooperative inclinations played a large role in explaining mass volunteering, but it would be simplistic and wrong to say that the whole nation spontaneously and uniformly rose up to smite the enemy. Just as it would be simplistic and wrong to say that all the British enlistees were coerced to join. The explanation requires a more nuanced and dynamic understanding.
EACH SOCIETY CONTAINS A SUBSTANTIAL proportion of people who, in addition to looking out for their material interest, are also motivated, at least in part, by social norms. So how did it come about that adherence to such norms became widespread in humans? Doesn’t the theory of natural selection predict that such altruistic behaviors could never evolve? For an explanation, we need to look to some recent developments in evolutionary biology and anthropology.
Actually, Charles Darwin himself was concerned with the apparent problem altruistic behavior presented to his theory of evolution. “Selfish and contentious people will not cohere, and without coherence, nothing can be effected. A tribe possessing ... a greater number of courageous, sympathetic, and faithful members, who were always ready to warn each other of danger, to aid and defend each other ... would spread and be victorious over other tribes.” The mechanism, proposed by Darwin, is now known as “group selection”—cooperation within groups evolves as a result of competition between groups. During the twentieth century, group selection went on a dizzying roller coaster, first enjoying wide acceptance, then being completely repudiated, and now on its way back to prominence, although in a different, more mature form. The problem was that during the first stage of uncritical acceptance, a lot of very bad theory was propounded by the adherents of the concept such as the famous Austrian ethologist Konrad Lorenz in his book On Aggression.
The main problem with the initial, crude version of the group-selection theory is this. Think about two types, altruistic “saints” and self-interested “knaves.” It is true that groups that have many saints will be doing better than groups with lots of knaves. However, in addition to this between-group competition, there is a within-group competition between saints and knaves, which saints inevitably lose. This is the collective-action problem, all over again. The benefits from prosocial actions of the saints are spread evenly among all group members, including the knaves, but the costs are born entirely by the saints. As a result, the saints will suffer higher mortality and lower reproduction rate compared to the knaves. Overall, numbers of saints will change as the result of two opposing tendencies: between-group competition that causes saint numbers to increase, and within-group competition that causes their numbers to decline. It is hard to say which process will prevail without doing calculations. Unfortunately for the group-selection theory, mathematical models show that, except under quite unusual circumstances, the individual (within group) selection will almost always overwhelm group selection.
By the 1970s, this mathematical result was widely known, and it became fashionable to make fun at the expense of the “naive” proponents of group selection. The said proponents crawled into various holes with their tails between the legs (with a few exceptions, most notably David Sloan Wilson at Binghamton University, who continued to plug on in almost total isolation). The “individual-selectionist” view became the dogma in the field of evolutionary biology as seen in, for example, Richard Dawkins’s book The Selfish Gene.
Yet, although Dawkins and others pronounced that there is just one dominant unit on which natural selection operates—the individual—in Dawkins’ own book he discusses at least three distinct units of selection. These units are the gene (which is reflected in the very title), the individual, and the group of relatives (Hamilton’s kin selection). Individuals, after all, are not unitary, structureless “atoms” (despite the name—individuum—meaning “undividable”). They are made up of organs, tissues, and cells, and each cell contains many genes. It might be in the common interest for genes to cooperate to ensure the cell’s proper functioning, but there could also be incentives for selfish genes to free-ride on this collective effort. Similarly, cells usually cooperate to promote the survival and reproduction of the organism, but at times this cooperation breaks down, and a bunch of knavish cells begins to increase at the expense of the cooperative ones. We know this as cancer. To cut a long story short, things are not quite as simple as Dawkins and other adherents of individual selection imply. In recent years, Wilson and colleagues were able to mount a successful attack on the individual-selectionist dogma. It is now becoming broadly accepted that natural selection operates at all levels simultaneously—genes, cells, organisms, groups of relatives, and simply groups.
It is true that among nonhuman organisms, under most conditions, group-level selection is quite weak. Empirical examples of group selection in nature are rare. Humans, on the other hand, are unique in the biological world in their capacity for thought, communication, and culture, and this makes group-level selection a very powerful force. The best current explanation of how human ultrasociality evolved is the theory of cultural group selection, advanced by the UCLA anthropologists Robert Boyd and Peter Richerson.
Probably the most important difference between humans and other organisms is the unique importance of cultural transmission of behaviors in humans. This is not to say that genes are unimportant in influencing the behavior of people. Studies of twins separated at birth have decisively shown that genetic makeup has a strong effect on behavioral traits, ranging from intelligence to political orientation. In the age-old debate on “nature versus nurture,” neither of the extreme positions is right; the truth lies squarely in between. Moreover, nature and nurture collaborate in determining behavior of people. Genes do not really tell people to vote Republican. Rather, certain genetic makeup can predispose people to conservative views, but whether they become card-carrying Republicans, or not, will depend very much on the family upbringing and accidental experiences throughout their life, such as reading a particular book or making a particular friend.
What makes cultural transmission really distinct from genetic inheritance is that people can learn from other people, not only from their parents. Young people adopt certain behaviors by imitating a particularly successful or charismatic individual in their tribe. They are also taught many things by the tribal elders, from catching fish to telling the truth. The point is that behavioral practices can spread rapidly within a group by this process of cultural transmission, much more rapidly than if the transmission process were determined solely by genes. Of course, any kinds of behaviors can spread by imitation and teaching, both beneficial and harmful for the group. That is why the competition between groups is so important—it weeds out groups that have fixated on harmful practices. For example, take the ritual consumption of the brains of deceased relatives among the Fore people of New Guinea. This turned out to be a bad practice because it allowed the transmission of a neurodegenerative disease known as Kuru. Had Shirley Lindenbaum and Daniel Gajdusek not discovered the cause of the disease, the Fore would have eventually succumbed and been replaced by other tribes that did not eat dead relatives.
Humans have large brains and highly developed cognitive abilities. Apparently, people can keep in their minds information about the history of dealings within a group of more than a hundred people, remembering those who keep their word versus the cheaters. Furthermore, evolutionary psychologists Leda Cosmides and John Tooby argue that there are specialized “cheater-detection circuits” that allow a potential cooperator to detect individuals who free-ride. In short, people are very smart when it comes to social interactions. This unique ability of humans enables us to become very efficient moralists. Remember that a moralist not only behaves according to the norms, but also detects and punishes cheaters—people who break such social rules. A “second-order” moralist also keeps track of those who shirk by not punishing cheaters, and punishes them (“Damn John Jay! Damn everyone who won’t damn John Jay!!”).
As far as we can tell, the social organization of our distant evolutionary ancestors was not too different from that of the chimps. Unlike chimps, however, who enjoy eating meat but can capture only small prey, our ancestors learned how to hunt large game in the savannas of Africa. Humans eventually learned (perhaps too well) how to kill large mammals, including the largest ones such as elephants and mammoths. When they spread out of their ancestral Africa to other continents, prehistoric humans wiped out most of the large animals living there. That is why the mammoths of Siberia or giant sloths of South America are no longer with us. Unlike African species, large animals elsewhere did not co-evolve with humans and, as a result, had no defense against their predatory ways. What made primitive humans such fearsome killers? Not their teeth or claws, obviously, but their ability to hunt cooperatively.
Taking up hunting of large game exposed early humans to intense selection at the group level. Not only coordination was key to a successful hunt for large game, but moving to regions where such prey are found exposed humans to formidable predators. Only collective vigilance and cooperative defense could protect humans against saber-tooth tigers and cave bears. Again, humans got so good at dealing with these predators that they eventually exterminated them.
Even more importantly, as humans got better at hunting large game, they also got better at killing other humans. At some point, warfare (that is, any kind of organized fighting, from several chimps waylaying and killing a member of a different band to trench warfare involving millions during World War 1) became the most important force of group selection. Several kinds of evidence show that early humans practiced extensive warfare. For example, we know that interband warfare is very common among the chimps, our closest evolutionary relatives. Warfare is also nearly ubiquitous among the small-scale societies of hunter-gatherers and farmers. The anthropologist Lawrence H. Keeley presents evidence that somewhere between 20 percent and 60 percent of males in these societies die in wars. By an argument of “interpolation,” therefore, if both chimps and modern pre-state people practiced extensive warfare, so must have our human ancestors. There is also direct evidence—cave paintings depicting lines of warriors shooting at each other, defensive walls around Mesolithic settlements, arrowheads embedded in skeletons, and mass burials of fighting-age males, many of whom were killed by a blow to the head.
