Good Reasons for Bad Feelings

by Randolph M. Nesse

  To try to resolve the conflict between theory and observations, I joined the throng of scientists searching for evolutionary explanations for cooperation and moral emotions. Many explanations have been proposed, and most authors advocate for one over the others. This quest for simplicity causes much unnecessary controversy when, as is the case here, several different explanations are all relevant. Fair warning: the following review of possible explanations concludes that many are important, and I will, like most everyone else, suggest that one is especially so.

  First, a brief executive summary of the origins of cooperation. (1) Benefits to groups of unrelated individuals cannot explain the evolution of extreme human social abilities. (2) Benefits to kin who share the same genes explain most altruistic behavior. (3) Much apparent cooperation among nonrelatives is just individuals doing things that help themselves and that also happen to help others. (4) Extensive cooperation among nonrelatives is explained mostly by reciprocial favor trading. (5) Reciprocity systems shape costly traits for establishing a good reputation. (6) The previous five explanations explain most social behavior in most organisms, but not quite all. They represent a spectacular fundamental advance in human knowledge even though they cannot fully explain human capacities for commitment and moral behavior. Important additional explanations are offered by cultural group selection, commitment, and social selection. First, however, a bit more about group selection.

  Group Selection Redux

  Some scholars argue that group selection works after all.17 Classic group selection is the process that increases the frequency of alleles that induce behaviors that decrease an individual’s fitness but that benefit a group of unrelated individuals. Groups with more individuals who are willing to sacrifice for the group grow faster than other groups do, so group selection is possible. However, alleles for such tendencies can persist only when three special circumstances are met: groups with more cooperative individuals must grow much faster than groups with fewer cooperative individuals; individuals with alleles for helping must reproduce only a little less than others in the group who do not have such alleles; and finally, exchanges of individuals between the groups must be limited, otherwise nonhelping individuals will move into the group and their alleles will displace the ones for helping.18,19 These conditions rarely occur together. Group selection of this kind is weak and unable to explain costly traits. An essay by Steven Pinker explains why in crystal-clear detail.20 However, most people, including some evolutionary psychiatrists, find group selection intuitively correct and emotionally appealing, so I will say a bit more about its limits before describing alternative explanations of our extraordinary capacities for cooperation and morality.

  The consensus among scientists is that group selection among nonrelatives cannot explain human genetic tendencies that make morality possible, but controversies persist. In part this is because models of the evolution of cooperation can be framed equally well in terms of kin selection or group selection.21,22 Most experts find kin selection far more useful;23 however, a few famous scientists have argued that it is irrelevant.24,25 I agree with the vast majority of scientists26 who find kin selection a profoundly useful explanation.

  Despite its intuitive appeal, examples of group selection are few. Some apparent examples mostly document its weakness. For instance, chickens in cages peck one another, causing injuries that slow their growth. Breeding chickens for a few generations using eggs from the cages with the least pecking shapes more cooperative chickens that grow faster.27,28 This is real group selection. However, it is not natural selection. In fact, it illustrates that no such group selection happened to chickens previously. Except in very special circumstances, genetic tendencies that benefit the group are eliminated if they decrease the reproduction of individuals.

  Sex ratios can be influenced by natural selection, and groups with almost all females grow twice as fast as those with one-half males—only females can have offspring, after all. But most sex ratios are close to fifty-fifty. The great geneticist Ronald Fisher explained why in his 1930 classic, The Genetical Theory of Natural Selection.29 He asked what sex of offspring would maximize transmission of an individual’s genes. In a group with almost all females, the average male offspring would have many times more offspring than the average female. In a group with almost all males, the average female would do better. Having offspring of whatever sex is in short supply maximizes an individual’s genetic contributions to future generations, at a substantial cost to the rate of growth of the group.

  Fisher’s logic is illustrated by pub choices on a Saturday night. When looking for women partners, men don’t go to a sports pub; the odds are terrible there. A pub with ladies’ night is a much better bet. For women, the logic is reversed. The prevalence of fifty-fifty sex ratios demonstrates the dominance of individual selection relative to group selection.

  Then there are forests. Magnificent towering tree trunks testify to the waste that results because selection maximizes the interests of genes instead of species. All available solar energy could be harvested by clusters of leaves close to the ground. Trees that cooperated could maximize energy collection without spending huge resources to grow tall trunks. However, each tree competes to get more sunlight than other trees. They even know when to compete harder. Light reflected from adjacent green leaves shifts many saplings into desperate competition mode, putting every ounce of effort into growing tall as fast as possible, at the risk of breakage. Even trees of the same species put most of their life’s effort into competing to be higher than other trees. The exceptions are instructive. Aspen flourish in tight groves of modest height. You can guess why: they are all clones with the same genes, so there is no need to compete; they also cooperate to create dense shade in that keeps competitors in the dark.

  The cells in our body are cooperative for the same reason: they all start off genetically identical, thanks to a process that strips DNA down to a single strand from each parent, one in the egg, one in the sperm. The cells in our bodies are like 40 trillion identical twins. Alleles can get into the next generation only by doing things that benefit the body as a whole. The exception proves the rules. When cells replicate without regard for the good of the whole, the result is cancer.30 Natural selection has shaped powerful mechanisms to inhibit such rogue replication, including a mechanism called apoptosis that induces suicide in cells that are replicating out of control.

  Cooperation Explained (Mostly)

  As noted in chapter 3, William Hamilton’s discovery of kin selection revolutionized our understanding of social behavior. When he came up with it, he wasn’t a great biologist; he was a lonely graduate student who had spent years pondering how evolution could explain sterile honeybee workers, which die when they sting to protect the hive.31 He proposed a doctoral thesis on the topic but was told that the project was unacceptable. So he submitted a manuscript describing his idea to a scientific journal.32 The editor, John Maynard Smith, immediately recognized that Hamilton had solved a problem that had vexed biologists for decades. Maynard Smith quickly published his own article on the idea in the widely read journal Nature, christening it “kin selection.”33 This caused a lifetime of bad feelings between the two men. How ironic and sad that selfish competition for priority caused a moral lapse at the origins of scientific studies of altruism. Maynard Smith was an inspiring conversation partner, generous with his patience, as I asked him ignorant question after question about group selection. Hamilton was a distracted genius who was curious about everything, and conversations with him about mental disorders inspired me. The two were eventually able to talk, but the relationship was forever strained. Their hard feelings foreshadowed continuing intense and sometimes mean-spirited conflicts about cooperation that still continue.34,35,36

  Mutual helping is another potent explanation of social behavior. If two animals groom each other simultaneously, both benefit and there is no opportunity for cheating. If two people turn over a heavy rock,
both benefit from anything underneath. Birds that eat ticks off the backs of livestock get a meal and the animals get reduced parasites. Such mutualism is everywhere once you recognize it.37,38,39,40

  Trading favors explains most helping between nonrelatives. If helping episodes are separated by time or place, cheating is possible. If two people are turning over rocks separately to look for gold, one could secretly pocket the treasure. A person who helps someone else build a shed may or may not get help later when he builds his own shed. If you drive someone to the airport, you may or may not get a lift when you need it. Reciprocal favor trading benefits both parties if cheating can be controlled.

  The idea has a long history, but its importance for the biology of social behavior came to attention in a 1971 article by the biologist Robert Trivers.41 A game called “the prisoner’s dilemma” provides a great way to study how people do and don’t reciprocate help. It gets its name from the situation in which the police are interviewing two criminal collaborators separately. Each is told that if they confess first (defecting), they will get off easy, but if the other person confesses first, they will get a severe punishment. If neither confesses despite the risk that the other one will defect (both cooperate), both benefit by getting only a moderate punishment. This game lends itself to computer modeling and games you can play with real people. Hundreds of studies have investigated how people trade favors. My friend and colleague the political scientist Robert Axelrod analyzed much of this work in his landmark book The Evolution of Cooperation.42,43

  If the prisoner’s dilemma game is played over and over, the best strategy is called tit for tat, that is, doing what the other person did on the previous move. This maximizes benefits when paired with a cooperator (who won’t confess) but avoids exploitation if the other person defects. The game usually plays out in long sequences of cooperation, followed by stubbornly persistent bouts of mutual defection, exactly what one sees in many relationships.44,45,46 Steady cooperation maximizes the combined benefit (3 to each partner in the table below), but one player can get a payoff of 5 by defecting on a turn when the other cooperates.

  If the situations that recur in the process of trading favors have fitness consequences, they should have shaped emotions to cope with those situations. They have.47,48,49,50 After repeated experiences of cooperation, trust and friendship grow. Especially generous behavior arouses gratitude. Anticipated defection arouses suspicion; experiencing a defection arouses anger. Temptation to defect arouses anxiety and defection arouses guilt, both aversive emotions that inhibit hasty selfishness.

  If you are tempted to do something that will betray a commitment, anxiety inhibits hasty self-interested behavior. You will be late for work if you give your friend a ride to the airport, but if you owe the person a ride, you need to do it. If you don’t, guilt will motivate apologies, and reparations will be needed to reestablish trust. Alternatively, you could try to devalue what the other person provided for you. Most arguments are about who violated what expectations.

  EMOTIONS SHAPED TO COPE WITH THE SITUATIONS THAT ARISE IN EXCHANGE RELATIONSHIPS51

  EMOTIONS AROUSED BY SITUATIONS IN RELATIONSHIPS

  OTHER COOPERATES

  OTHER DEFECTS

  YOU COOPERATE

  (3 for each)

  Friendship

  Trust

  (0 for you, 5 for other)

  Suspicion (before)

  Anger (after)

  YOU DEFECT

  (5 for you, 0 for other)

  Anxiety (before)

  Guilt (after)

  (1 for each)

  Disgust

  Avoidance

  Although the reality of social life is far more complex, this simple table provides a useful guide to the origins and utility of social emotions.52 Anger signals that a betrayal has been recognized and apologies and reparations will be required to keep the relationship going and to avoid spiteful revenge.53 People who feel they can’t leave a relationship are reluctant to express anger, leading to passive-aggressive behavior or sullen withdrawal that limits cooperation and fuels chronic conflict. Such situations are at the core of much neurosis and many marital problems. The psychologists Timothy Ketelaar and Martie Haselton have advanced this construct much further,54,55,56 but clinical applications remain to be developed.

  Something Is Missing

  The explanation of social behavior by kin selection, mutual benefits, and trading favors is one of the great scientific advances of our time. Together, they explain almost all cooperation.57,58,59,60,61,62,63 But not quite all. They have trouble accounting for people who lie awake nights feeling guilty because of a minor misstep no one else knows about. They don’t provide a full explanation for the huge sacrifices made in committed relationships. They don’t explain fighting to defend the group despite knowing that death is inevitable. And they can’t explain why, for every sociopath, there are ten people who worry constantly about how to avoid offending others. Humans have extreme prosocial tendencies that need additional explanations. Providing them is a major academic endeavor that is making progress.64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80 The key to a general solution comes from recognizing that altruists who associate selectively with other altruists reap advantages compared to others who merely trade favors with random others.

  Simple geographical proximity is the simplest mechanism; the offspring of altruists will likely live close to other altruists. This is relevant even for bacteria. Because they divide fast, bacteria are usually surrounded by close relatives, so bacteria that contribute to the common good, for instance by allocating resources to manufacture a substance that digests host cells, derive advantages for their own genes.81,82

  Humans have many ways to find and stay close to good partners. Avoiding jerks results in more time spent with generous types. Leaving suboptimal partnerships results in selective association with altruists.83 Gossip provides invaluable information about whom you can trust.84 Hiring committees spend hours checking references for good reason. Such models of selective association among altruists are sometimes called group selection, but this creates confusion. As the biologist Stuart West said incisively, “An alternative is to state as simply as possible what they are—models of nonrandom assortment of altruistic genes.”85

  Descriptions of cultural group selection developed by anthropologists Robert Boyd and Peter Richerson provide a major explanation for deep cooperation and intense altruism.86 Groups with cultural norms to sacrifice for the group grow faster than other groups. Groups can give advantages to individuals who obey the norms, creating selection for the tendency to do what is good for the group. Individuals can benefit the group by punishing cheaters, but rewarding cooperators is usually more effective and less dangerous. A recent article by Richerson and colleagues reviewed extensive evidence for the power of cultural group selection.87 The article is convincing, but it argues that altruistic behaviors that cannot be explained by group selection, mutualism, kin selection, or reciprocity must be therefore explained by cultural group selection. However, there are at least two other ways that selection can shape capacities for cooperation: commitment and social selection.88

  Commitment

  Commitment is not just keeping promises to a spouse. In game theory, commitment can explain altruistic acts that have no guaranteed or even expected return benefits.89,90,91,92 The core idea is paradoxical: convincing others that you are committed to some future actions that won’t be in your interest can influence their behavior powerfully. Promising to stay with someone in sickness or in health can cement a relationship with a better partner than would otherwise be possible and, hopefully, also get you help when you are sick. Threatening a nuclear response to
an attack is far more irrational, but it is a powerful influence if others believe you would do it. Mutually assured destruction has prevented war so far, but commitment strategies are unstable, so civilization as we know it could end at any time.

  Relationships based on commitment are more valuable than those based on reciprocity. Evolutionary psychologists John Tooby and Leda Cosmides have written profoundly about “the banker’s paradox.”93 Banks operate on reciprocity only: they are glad to lend you money when you have collateral, but when you have nothing and really need financing, they won’t even talk to you.

  Relationships based on commitment provide help when people need it most: when they have little to offer in return. The challenge is to convince others that you will, in some future situation, do something that will not be in your own interest. The related challenge is to convince yourself that others will help you when there is no way to enforce the obligation. The solution is to go ahead with non-self-interested actions that demonstrate your commitment. Skip the big game to stay home and help her nurse her cold. Cancel the big presentation to go on the planned vacation. Before you know it, what may have started as a manipulation turns into an enduring commitment.