A quarter-century ago stephen Emlen set out to investigate tales of an African utopia. He had heard of villages in Kenya where as many as 300 individuals lived in apparent harmony, despite being packed together in clustered homes. Even when food grew scarce, everyone shared what was available. Males and females took turns caring for children. Couples were monogamous, and to all appearances faithful. Grown children, rather than leaving home and starting families of their own, often stayed on to help their parents tend to younger brothers and sisters. Sometimes they helped neighbors. Altruism, in short, prevailed over self-interest.
The residents of this peaceable kingdom were not people--they were birds, of a species called the white-fronted bee-eater. But to Emlen, a Cornell biologist who studies animal behavior, that trivial fact did nothing to lessen the mystery of the phenomenon. Evolution should create struggle between bee-eaters as individuals try to maximize their own and their offspring’s chances of survival--not a purely selfless society. When Emlen set out for Kenya, he thought he was about to discover an exception to nature’s laws. But after years of research, he realized that he had instead found the exception that proves the rule: bee-eaters in fact do not altruistically help strangers; rather, through the complex dynamics of their large, multigenerational families, they help themselves. The myriad ways in which bee-eater parents, children, aunts, uncles, and grandparents behave toward one another exquisitely match what you’d predict from the principles of natural selection. Hidden under the utopian veneer is a swirling soap opera of love, deceit, harassment, divorce, and adultery.
Was Emlen disappointed with his hard-won findings? Not in the least. They led him to thinking about families. Why, he began to ask himself, do some species form families while others cut their ties to their children as fast as possible? Why do individuals behave differently in different kinds of family structures, and is that behavior predictable? In pursuit of answers to such questions, Emlen has embarked on a quest for a full-blown, mathematically rigorous model of the family.
It’s time to lay this out as a unified theory, he says--a theory of not just bee-eater families, but all families, humans included.
As Emlen walks into his office at Cornell he brushes past Clint Eastwood. The cardboard cutout of the movie star squints back in utter disgust, as if itching to gun him down. Ten feet away, Indiana Jones lurks in a doorway. Students set up the cutouts--they regard Emlen as something of a scientific swashbuckler, but he tries to play down the reputation. It is true that he has walked barefoot at night through Kenyan grasslands (so as not to disturb sleeping chicks), despite the deadly resident puff adders. And when in Panama, he has worried a bit about the alligatorlike caimans--especially when he gets excited, stands up in a canoe, spills a laptop or a pair of binoculars, and then has to dive into the water to wrench those valuables from the ooze. But no caiman has ever really bothered him; they’re interested only in birds and eggs.
In Emlen’s experience, the biggest threats to tropical biologists are akin to the ones that might fell a weekend gardener. In 1991, for example, Emlen hurt his back in Panama while pulling 30-foot-long poles from river-bottom muck (the poles held up bird-catching nets). He ruptured two disks and he’s had two operations; right now he can barely sit. In the office he works mostly standing up. Soft rubber under the carpet cushions his fidgety pacing; a lectern allows him to read while standing, and he has raised his telephone, computer, and microscope chest high. In the field, Emlen says, he’s been able to modify a canoe seat so he can work in quasi comfort.
In light of these hardships the man seems remarkably chipper. He greets everyone with equal gusto--students, colleagues, the handyman refitting his office. Walking around campus--which he seems eager to do to help relieve the stress in his back--he quickly becomes excited, at times exuberant, when discussing his work. If he lost another laptop, a few more birds to predators, or even another utopia or two, they would simply be grist for his mill.
Emlen traces the origins of his current fascination with families to the early 1970s, when evolutionary biologists were hot on finding a scientific explanation for altruism. How does selfless behavior--helping others at one’s own expense--make sense? It seemed to contradict Darwinian thinking. In those days the notion of the selfish gene began to reign supreme. According to this idea, bodies are essentially fancy machines that genes invented to help themselves replicate. When an animal has genes that confer some advantage--when a predator gains better eyesight, or a prey animal better camouflage--that individual becomes more likely to survive long enough to mate and multiply its genes. It’s as if the genes themselves are forever seeking better strategies to build as many bodies as possible, while pouring copies of their tiny life codes into them all. The fittest genes--the ones that produce the most copies of themselves--dominate the population.
Behavior fits into this theory if genes can somehow control it. Animals that can become more or less aggressive, for example, or more or less cooperative, might be more likely to survive in certain situations than others. Altruism, however, presents a challenge. Self-sacrifice simply doesn’t seem to offer any beneficial return, especially if it means sacrificing an opportunity to breed. Genes that tell bodies not to breed should not survive, and yet such behavior clearly exists. In birds, the most extreme cases of apparent altruism are nicknamed helpers, says Emlen. Individuals forgo breeding for large portions of their lives and instead help someone else raise kids. The apparent cost is high: you’re forfeiting your own reproduction; you’re increasing the reproduction of somebody else. So it’s a classic case of the paradox.
A theorist named William Hamilton solved at least part of the puzzle. Starting in 1964, he produced a sheaf of mathematical work showing how your genes can benefit when you help other people’s children survive-- if those kids happen to be your kin. When you raise your own children, your genes are nurturing copies of themselves. If your genes can program you to feel tender emotions toward those kids, it works to your genes’ selfish advantage. But Hamilton realized that other relatives carry copies of at least some of your genes, and he expanded the concept of fitness to include the survival of more distant kin. It’s as if your genes get a 50 percent payoff by helping perpetuate your own kids’ genes (with whom you share 50 percent identical genes), or 25 percent by helping your sister’s kids, your grandchildren, and so forth. The more distant the relation, the lower the payoff.
With arithmetic like this, it becomes clear that in plentiful times young adults may get the best genetic payoff by leaving home and starting their own families. But in times of scarce food supplies, the odds of those kids surviving might be low. The genetic payoff will then be higher if those young adults cooperate with other adults to raise a close relative’s offspring. By Hamilton’s logic, a spinster aunt who helps bring up her nieces and nephews can sometimes be genetically fitter than a recklessly prolific breeder.
It was an elegant model of family behavior--but the white-fronted bee-eater seemed to fly in its face. The reason I went after bee-eaters, Emlen says, is they were seen to have complex, nonfamily helping. Humans give that kind of help, but anthropologists often explain it in terms of reciprocity. Perhaps, he thought, the bee-eaters were also somehow returning help they got from outside their own families. Today his evaluation of that speculation is simple. I was wrong, he says with a laugh.
Science has always been a family business for Emlen. His father, renowned biologist John Emlen, took him along when he did fieldwork in Wyoming and Michigan’s Upper Peninsula. When Stephen first headed to Africa in 1973 to study bee-eaters, he took with him his wife, Natalie Demong; on later trips they brought their children. (One son, George, is now an evolutionary biologist also, carrying the family business into a third generation.) It was a family of humans observing families of birds. Usually a student or two would take a semester off to accompany them, and halfway through the project one of them, Peter Wrege, became a permanent member of the team.
They spent eight years intensely studying bee-eaters, half the time in Africa observing the birds and half back home in Ithaca, poring over notes and data. Assessing the costs and benefits of bee-eater behavior wasn’t easy: the birds live together in groups of several hundred, and they all look pretty much alike, whether male or female: green backs, tan bellies, white and red throat stripes, gracefully curved black beaks. They live in holes a yard or two deep dug into the sides of dirt cliffs; the Emlens caught the birds in mist nets as they emerged from their holes-- positioning those nets meant barefoot walking at night--and eventually marked every bird with a wing tag. Natalie took photographs of all the apartments so each bird could have an address, and as kinship became known, the Emlens and their assistants made flash cards identifying who was related to whom. Hiding behind blinds for most of the day, focusing on one or two apartments at a time, they could see whether birds would go in with food or without, and who greeted whom at the entrance. Natalie invented a sort of ornithological periscope made with dental mirrors, a light, and a battery pack worn around the waist; with it, they could peek inside the deep, foul-smelling burrows to see who was home.
They watched nests with eggs nonstop for two days, watched the same nests again when the hatchlings were young, and then again when the chicks were nearly ready to leave. So we had rigorous information from three different ages, Emlen says, and we’d have those same data for lots of different nests. Every evening they returned to their house, ate a late supper, talked about who was visiting whom, about courtships, births, and deaths, and eventually trespassing, bullying, subtle deceit--all of which grew more exciting as tensions and conflicts became more apparent. It was like watching a soap opera, or the Hitchcock movie Rear Window, says Natalie. Then Stephen would spend hours at night transcribing the tapes they had made in the field and deciding which apartments to watch the next day.
Meanwhile they developed genealogies, which they confirmed with molecular tests. Eventually, Emlen says, you have a whole database on the survival of everybody--a database on who has socially interacted with whom and who has aggressed whom, a database on contributions to the nest, a database on who dispersed and moved between colonies, a database on where they go for their feeding and how successful they are as foragers. Back at Cornell, Natalie and Wrege helped Emlen link all those databases and crunch the numbers.
The longer they studied the bee-eaters, the more they realized that what seemed like simple altruism was based on complex evolutionary calculations. Early on, for example, Emlen discovered that if a nest failed (if a predator ate the eggs, say), a son who was helping at the nest commonly moved into a different apartment and continued helping over there. Emlen thought at first the bird was helping unrelated neighbors, but after several years, as he built up genealogies, he realized that bee-eaters live in large, multigenerational families of up to 17 members that occupy many apartments. Each bird spends a good deal of time visiting parents, grandparents, uncles and aunts, cousins, nieces, and nephews.
They all know each other, they all greet each other, they exclude nonfamily members. So what I missed is that virtually every helper was shifting over to another family member. If your parents’ nest failed-- boom--you jumped over to your brother and sister-in-law. Or if a breeder’s nest failed, it would go over and help its son and daughter-in-law rear grandchildren.
As soon as you have an extended family, Emlen continues, the game becomes much trickier. Each bird has many more choices of whom to help--father, aunt, grandmother, cousin, and so on. Since Emlen knew the calculus that Hamilton had worked out, he thought he should be able to predict who would help whom, and when. If a nest fails, I should be able to predict the number two backup that gets the help. He thought he could also predict the point at which the cost exceeded the benefit and the helper wouldn’t bother. And it turned out, he says, that every prediction was incredibly true. Whether you helped or whether you sat out the season was very much predicted on closeness of kinship. The likelihood of helping just falls off--boom, boom, boom, boom, boom--as the relationship decreases.
Emlen began to perceive a vast web of selfish--as well as cooperative--behavior. For example, female bee-eaters will sometimes try to lay their eggs in the nests of unrelated birds so that they themselves don’t have to expend the effort required to raise their young--a phenomenon called nest parasitism. They can lay an egg in a few minutes and be out. Most of these parasitisms come from outside the family. Mothers and fathers have counterstrategies: they take turns guarding their chamber, and they repel anyone but a family member. If they are both gone from the nest at the same time for a few minutes and return to find a new egg, they throw it out. And a day or two after they lay their eggs, it’s too late for a parasite to sneak in one more. Once the parents’ own eggs hatch, they stop incubating the foreign laggard. So a parasite has to get in right around laying time, Emlen explains--which is not easy.
Unless, he discovered, it’s an inside job, done by someone who has access to the nest and the trust of the parents: in other words, by a mateless helper daughter. The daughter has the whole defense mechanism cracked. She’s taking her turn defending against anybody else. She can synchronize her timing with her mother. So she’s got the timing down, she has access to the nest--she can beat her parents, whereas an external parasite can’t.
The only hitch would seem to be that the daughter isn’t paired. But by monitoring daughters carefully, Emlen discovered the full dimension of their sneakiness. Bee-eaters travel a mile or two from their nests to feeding territories where they catch insects. And there, Emlen says, they saw a daughter behave in a way we have never seen any other time. She just leaves and trespasses on the territory of another family. At first they drive her away. But after doing this for a while, the male on whose territory she is stops aggressing and copulates with her.
Emlen was stunned. Whoa, he remembers thinking, birds are doing this kind of strategic thing? That one example, he says, opened our eyes to the fact that some very subtle, tactical social things are going on, and maybe we had been much too naive in thinking of birds as little simplistic automata.
The older birds were no angels. When young birds tried to pair off and start their own families, their parents or other older relatives would often harass them--essentially choking them with kindness. When the younger birds are paired, Emlen says, and they’re starting to dig their own chamber, there’s a high level of ‘friendly’ interference. An older relative will guard the entrance--very helpful, you might think--but will refuse to recognize the daughter-in-law, who can’t get in. Or a dominant relative might pay frequent visits to a young male whenever he tries to feed his mate. A father, for example, might trespass onto the adjacent territory of his son and daughter-in-law and greet them and greet them and greet them. What’s happening is he’s disrupting the son’s ability to feed his mate.
There is, according to Emlen, a calculated strategy hiding underneath this annoying behavior. By harassing their younger relatives, these bee-eaters increase the chances that the new nests will fail and the young birds will return home to serve as helpers. After all, if they themselves can’t lay eggs, the next best way to spread their genes is to go back home and help Mom and Dad.
Until they were deeply involved in the system, Emlen says, they would have completely misunderstood all these friendly gestures. Ironically, the individuals you have the most leverage over to harass are your closest subordinate relatives. And that can also be mathematically modeled, in terms of exactly who you should harass. So we see brothers occasionally doing it to younger brothers, grandparents occasionally doing it to grandsons, but mostly it’s fathers harassing their sons. So there’s a lot going on below the surface of this harmonious, altruistic picture--all sorts of tugs-of-war.
It’s a complicated scenario, and figuring out what is the optimally selfish way to behave in any given situation is not easy, even for a human--Emlen needed a huge amount of data and a computer to test his predictions of bee-eater strategies. So how could lowly, birdbrained bee- eaters work out the same calculations so precisely? Apparently, natural selection has already done the computing for them, through many thousands of years of trial and error. Suppose we have several bee-eaters, Emlen says, and they all have systems for weighing the relative importance of their various kin. Now suppose one of them weighs a certain kinship more heavily than another does. If it’s a bit more likely to feed a grandson, say, than it is to feed a cousin, that will have consequences on the success of rearing those kids, and have impacts on the number of gene copies down the pike. If the genetic payoff is better for helping a grandson than a cousin, then the genetically based system that recognizes grandsons survives, while the one that predisposes the bird to help cousins gradually wanes. Eventually the helping behavior of a bird species will become fine-tuned to the opportunities its particular family structure allows.
Emlen is quick to point out that the result of this natural selection is not a rigid, deterministic set of reflexes. We’re not talking about the evolution of a gene for altruism. It isn’t the behavior per se the gene is regulating. The gene--or rather, a complex of many genes--is instead regulating the ability to evaluate social situations. An animal might develop genes that help it distinguish its kin by sight, smell, memory, tail-quivering password, or all of the above. And that complex of genes might include some that pump up different hormones for different emotions, according to what’s perceived. All this would raise the odds of behaving in a certain way without making that behavior totally gene-driven. With so much of neurobiology yet to be understood, Emlen treats the specific chains of genes, hormones, and other triggers for behavior as nothing more than a black box. He prefers to call them decision rules. It’s as if you have a toolbox of all sorts of behaviors, Emlen says, and what natural selection fine-tunes is the ability to choose the right behavior from the toolbox and use it in the right context.
As he watched the family sagas of bee-eaters play out, Emlen says, he increasingly came to see the families as very humanlike. And so, right or wrong, I got to thinking more and more about the fact that bee- eaters were structured that way. I didn’t know it when I started out. Bee- eaters are monogamous, they have about 85 percent fidelity rates--higher than humans currently but in the same ballpark, in the last two centuries at least. The parallels go on and on--there is divorce. Pairs that are unsuccessful in their early reproductive attempts have higher probabilities of divorce.
The similarities are even more striking when you consider that for hundreds of thousands of generations, our ancestors, like bee-eaters, probably lived in large extended families. In prehistoric societies, people were probably surrounded from birth to death by uncles and aunts, parents and grandparents, brothers and sisters, and all their relatives’ children. And with limited resources from hunting and gathering, many of the tensions between relatives over the raising of children (and the solutions they reached) may have mirrored those found among bee-eaters.
The extended human family survived even the huge changes wrought in the way people lived by the agricultural revolution. Only in the past few centuries did it begin to disintegrate. In many countries, technological and economic factors like wealth and mobility have caused families to dwindle down to the nuclear, Ozzie-and-Harriet form and then keep dwindling, with single-parent families becoming more common. (In the full sweep of human evolution, both these forms of the family are abnormal.) The same forces are creating more stepfamilies than ever before, which, in an evolutionary sense, amounts to the same thing as the single- parent family: fewer close genetic relatives living together. Yet as we try to adjust to these new kinds of families, we still carry with us the unconscious decision rules we evolved over hundreds of thousands of years in extended families.
Stepfamilies--human or otherwise--are one of Emlen’s favorite examples. The prediction on the straight selfish-gene logic, he says, is that we should see a real change in behavior after a mate change. Let’s say the kids remain with their mother, and a stepfather comes into the picture. If the mother and stepfather have children, her original kids should be less willing to help raise them, because half-brothers and sisters share only 25 percent of their genes. That’s the same as nieces and nephews, so moving out to help a brother and sister-in-law should be equally likely. Also, older kids should be more likely to leave the family altogether, because the payoff for staying decreases.
Years ago Emlen started testing these predictions among bee- eaters by watching what happened when long-term mates split apart and chose new ones. The probability of kids helping raise hatchlings produced by their own parent and a step-parent dropped, while their probability of shifting over to help others in the extended family increased, and their probability of leaving the family altogether increased. The kids from the first pairing say, ‘Thank you, no. I’m doing much less, provisioning much less. I’m going to spend much more time over there with my other family members.’
Does the same logic apply to families across the animal kingdom? Emlen is now trying to create a unified theory of family behavior, one that can explain the family dynamics to be found in humans, other mammals, insects, and birds. In this magnum opus, Emlen synthesizes not only his own work but the labors of a whole generation of biologists who have studied family behavior in vertebrates ranging from fish to birds to mammals, including humans. In a preliminary version--An Evolutionary Theory of the Family, which appeared in Proceedings of the National Academy of Sciences in 1995--he managed to summarize in only seven pages every kind of family situation any novelist could imagine. A table of 15 predictions covers the likely behaviors: helping, infanticide, sexual conflict, dispersal. In that paper he explains his work in plain English; now he’s working on the next version, which will be a set of formulas that any behaviorist can test by plugging in data.
Why do families form in the first place? It’s in the theory. Families form when there are limited resources and limited opportunities for successfully mating and raising kids. Families are also inherently unstable. In plentiful times, maturing offspring are more likely to strike out on their own than stay home as helpers. And yet rich families are more stable than poor families. If a family of mongooses controls an exceptionally good hunting territory, for instance, the offspring are more likely to delay reproducing and will instead hang around helping the elders, because the prospects of inheriting the family goodies look so much better than the prospects of trying to raise kids while claiming and defending a new territory.
Who will help whom among families, and when? It’s in the theory. Almost every species among the hundreds that have been studied follows Hamilton’s selfish-gene logic to the letter, whether the creatures live in bizarre families like those of bees (sisters, sisters, which one of you zillions is not my sister?), or family structures more familiar to us, like the bee-eaters. What happens when a replacement mate enters the family? It’s in the theory. The payoff of staying at home goes down for children, while new sexual tensions arise. Stepfamilies will therefore be less stable than intact biological families.
There is precious little information about humans that can be used to judge whether Emlen’s predictions capture our family life as well. But what statistics there are are supportive. Human stepparents invest less in offspring from previous pairings, and children in stepfamilies are more likely to be physically abused. There’s more conflict between stepsiblings than biological siblings, as well as the opposite: the stepkids just don’t interact--they ignore one another. Stepchildren leave home earlier. And stepparents commit infanticide at a rate 60 times higher than biological parents.
Emlen spends much of his time now traveling from conference to conference to make his case for an evolutionary view of family problems, and one objection he often hears is that birds, the source of his most convincing data, are too different from humans to be used as a model. People can accept primates, he says, but birds, for crying out loud? It’s as if there’s a cement wall that people can’t cross. And I see my role as breaking that wall. We may be able to learn more from ‘lower’ animals than we think. Obviously, we are closer biologically to the other primates. They are, Emlen grants, excellent models for mental capabilities and cognition. But their families are not like ours. Chimpanzees, for example, live in troops in which males play no role in raising their offspring.
It’s birds that are overwhelmingly monogamous and form pair- bonds, Emlen argues. It’s birds where males play a role in parental care. This is not a typical primate trait at all. Animals that live in very similar types of societies will have had long evolutionary histories of encountering the same types of social choices and therefore will have developed very similar rules for how to behave. And the best examples of extended family-based societies that Emlen knows of are the bee-eaters.
Knowing how our own genes have predisposed us to act, Emlen tells his audiences, can help us cope with old-family behaviors that come bursting out of our modern, new-family containers. These behaviors are going to be difficult to change, he warns, if they are in fact evolutionary. But they are predispositions, not determined, which means they can be consciously overcome.
One way to reduce the tension between our evolutionary history and modern family life would be to manipulate our social environment. Emlen isn’t advocating legislating ourselves back into a Pleistocene family structure. But it might be helpful, he muses, to offer a tax break to grandparents if they sell their condo in Arizona and move back to the town where their grandchildren live, to pitch in and help raise them.
Another approach is to recognize how certain family situations can turn into flash points for violence and abuse. One of the most explosive situations, Emlen suggests, may be a stepparent coming into a family with a sexually mature son or daughter. Sexual abuse in stepfamilies is much higher, about eight times higher, than in biologically intact families, he says. The same thing happens in birds. When a new mate comes in, there had been no sexual interest whatsoever between a son and his mother. There is great interest between the son and his stepmother, and great aggression as the father attacks the son, the father guards the mate. The human data show analogous trends. Consider a newly remarried mother, for instance, who has a teen-age daughter. Even with the best intentions, Emlen says, there is going to be a higher probability of subsurface tension and competition--though you might deny it constantly-- between that daughter and the mother, vis-à-vis the stepfather. Knowing this, anticipating this, saying this is natural, seeing why it may all have come about, I think can help defuse the flash point.
Of course, you can’t defuse what you don’t anticipate. Emlen would like families and counselors to accept the greater likelihood of conflict in stepfamilies. A statistically greater chance, he says, does not mean that every family is going to have these problems, but people might be counseled to expect problems and prepare for them. We are a species that has the ability to consciously overcome predispositions that we feel are inappropriate. And counseling is better than a surprise. I can have my antennae out, I can be more alert. I can nip most of these problems in the bud.
Aside from flash points, stepparents also find themselves having to cope with a lingering sense of guilt. As I talk to social scientists and read the literature, Emlen says, there’s a lot of information about parents in stepfamilies feeling guilty because they just don’t engender the same emotional feelings toward the stepchildren as toward their own. This is precisely what you’d expect. Understanding the evolutionary pressures that make us feel differently toward biological children and stepchildren can help one handle the challenge of making a stepfamily work. It helps you realize you have to go the extra mile, that biology is not going to help you.
Ultimately, Emlen hopes that we will be able to deal with evolution’s influence on our behavior in much the same way medicine deals with genetic vulnerabilities. If you have an inherited disposition for hypertension, you cut down on salt; if you enter a family situation that tends to lead to unhappy endings, you can look for counterstrategies.
At first it sounds ominous, says Emlen, saying that if you’re in this kind of family situation, you’re in trouble. But I argue that it should be empowering. Knowledge should be empowering.