Behave: The Biology of Humans at Our Best and Worst

by Robert M. Sapolsky

  In the 1970s anthropologist Joseph Shepher examined records of all the marriages that had ever occurred between people from the same kibbutz. And out of the nearly three thousand occurrences, there was no instance of two individuals marrying who had been in the same age group during their first six years of life. Oh, people from the same peer group typically had loving, close, lifelong relationships. But no sexual attraction. “I love him/her to pieces, but am I attracted? Yech—he/she feels like my sibling.” Who feels like a relative (and thus not like a potential mate)? Someone with whom you took a lot of baths when you both were kids.

  How’s this for irrationality? Back to people deciding whether to save the person or the dog. The decision depended not only on who the person was (sibling, cousin, stranger) but also on who the dog was—a strange dog or your own. Remarkably, 46 percent of women would save their dog over a foreign tourist. What would any rational baboon, pika, or lion conclude? That those women believe they are more related to a neotenized wolf than to another human. Why else act that way? “I’ll gladly lay down my life for eight cousins or my awesome labradoodle, Sadie.”

  —

  Human irrationality in distinguishing kin from nonkin takes us to the heart of our best and worst behaviors. This is because of something crucial—we can be manipulated into feeling more or less related to someone than we actually are. When it is the former, wonderful things happen—we adopt, donate, advocate for, empathize with. We look at someone very different from us and see similarities. It is called pseudokinship. And the converse? One of the tools of the propagandist and ideologue drumming up hatred of the out-group—blacks, Jews, Muslims, Tutsis, Armenians, Roma—is to characterize them as animals, vermin, cockroaches, pathogens. So different that they hardly count as human. It’s called pseudospeciation, and as will be seen in chapter 15, it underpins many of our worst moments.

  Reciprocal Altruism and Neo–Group Selectionism

  There’s not much to say here other than that this is the most interesting stuff in the chapter. When Axelrod got his round-robin tournament all fired up, he didn’t canvass, say, fish for their Prisoner’s Dilemma strategies. He asked humans.

  We’re the species with unprecedented cooperation among unrelated individuals, even total strangers; Dictyostelium colonies are green with envy at the human ability to do a wave in a football stadium. We work collectively as hunter-gatherers or as IT execs. Likewise when we go to war or help disaster victims a world away. We work as teams to hijack planes and fly them into buildings, or to award a Nobel Peace Prize.

  Rules, laws, treaties, penalties, social conscience, an inner voice, morals, ethics, divine retribution, kindergarten songs about sharing—all driven by the third leg of the evolution of behavior, namely that it is evolutionarily advantageous for nonrelatives to cooperate. Sometimes.

  One manifestation of this strong human tendency has been appreciated recently by anthropologists. The standard view of hunter-gatherers was that their cooperative, egalitarian nature reflected high degrees of relatedness within groups—i.e., kin selection. The man-the-hunter version of hunter-gatherers viewed this as arising from patrilocality (i.e., where a woman, when marrying, moves to live with the group of her new husband), while the groovy-hunter-gatherers version tied it to matrilocality (i.e., the opposite). However, a study of more than five thousand people from thirty-two hunter-gatherer societies from around the world* showed that only around 40 percent of people within bands are blood relatives.63 In other words, hunter-gatherer cooperativeness, the social building block of 99 percent of hominin history, rests at least as much on reciprocal altruism among nonrelatives as on kin selection (with chapter 9’s caveat that this assumes that contemporary hunter-gatherers are good stand-ins for ancestral ones).

  So humans excel at cooperation among nonrelatives. We’ve already considered circumstances that favor reciprocal altruism; this will be returned to in the final chapter. Moreover, it’s not just groups of nice chickens outcompeting groups of mean ones that has revivified group selectionism. It is at the core of cooperation and competition among human groups and cultures.

  Thus humans deviate from the strict predictions concerning the evolution of behavior. And this is pertinent when considering three major criticisms of sociobiology.

  THE USUAL: WHERE ARE THE GENES?

  I pointed out earlier a requirement for neo–group selection, namely that genes be involved in a trait that differs more between than within groups. This applies to everything in this chapter. The first requirement for a trait to evolve is that it be heritable. But this is often forgotten along the way, as evolutionary models tacitly assume genetic influences. Chapter 8 showed how tenuous is the idea that there is “the gene,” or even genes, “for” aggression, intelligence, empathy, and so on. Given that, even more tenuous would be the idea of a gene(s) for maximizing your reproductive success by, say, “mating indiscriminately with every available female,” or by “abandoning your kids and finding a new mate, because the father will raise them.”

  So critics will often demand, “Show me the gene that you assume is there.” And sociobiologists will respond, “Show me a more parsimonious explanation than this assumption.”

  THE NEXT CHALLENGE: IS EVOLUTIONARY CHANGE CONTINUOUS AND GRADUAL?

  The term “evolution” carries context-dependent baggage. If you’re in the Bible Belt, evolution is leftist besmirching of God, morality, and human exceptionalism. But to extreme leftists, “evolution” is a reactionary term, the slow change that impedes real change—“All reform undermines revolution.” This next challenge addresses whether evolution is actually more about rapid revolution than about slow reform.

  A basic sociobiological premise is that evolutionary change is gradual, incremental. As a selective pressure gradually changes, a useful gene variant grows more common in a population’s gene pool. As enough changes accrue, the population may even constitute a new species (“phyletic gradualism”). Over millions of years, dinosaurs gradually turn into chickens, organisms emerge that qualify as mammals as glandular secretions slowly evolve into milk, thumbs increasingly oppose in proto-primates. Evolution is gradual, continuous.

  In 1972 Stephen Jay Gould and paleontologist Niles Eldredge of the American Museum of Natural History proposed an idea that simmered and then caught fire in the 1980s. They argued that evolution isn’t gradual; instead, most of the time nothing happens, and evolution occurs in intermittent rapid, dramatic lurches.64

  Punctuated Equilibrium

  Their idea, which they called punctuated equilibrium, was anchored in paleontology. Fossil records, we all know, show gradualism—human ancestors show progressively larger skulls, more upright posture, and so on. And if two fossils in chronological progression differ a lot, a jump in the gradualism, there must be an intermediate form that is the “missing link” from a time between those two fossils. With enough fossils in a lineage, things will look gradualist.

  Eldredge and Gould focused on there being plenty of fossil records that were complete chronologically (for example, trilobites and snails, Eldredge’s and Gould’s specialties, respectively) and didn’t show gradualism. Instead there were long periods of stasis, of unchanged fossils, and then, in a paleontological blink of an eye, there’d be a rapid transition to a very different form. Maybe evolution is mostly like this, they argued. What triggers punctuated events of sudden change? A sudden, massive selective factor that kills most of a species, the only survivors being ones with some obscure genetic trait that turned out to be vital—an “evolutionary bottleneck.”

  Why does punctuated equilibrium challenge sociobiological thinking? Sociobiological gradualism implies that every smidgen of difference in fitness counts, that every slight advantage of one individual over another at leaving copies of genes in future generations translates into evolutionary change. At every juncture, optimizing competition, cooperation, aggression, parental investment, all of it, is evolutiona
rily consequential. And if instead there is mostly evolutionary stasis, much of this chapter becomes mostly irrelevant.*

  The sociobiologists were not amused. They called the punctuated equilibrium people “jerks” (while the punctuated equilibrium people called them “creeps”—get it? PE = evolution in a series of jerks; sociobiology = evolution as a gradual, creeping process).* Gradualist sociobiologists responded with strong rebuttals, taking a number of forms:

  They’re just snail shells. First, there are some very complete fossil lineages that are gradualist. And don’t forget, said the gradualists, these punctuated equilibrium guys are talking about trilobite and snail fossils. The fossil record we’re most interested in—primates, hominins—is too spotty to tell if it is gradualist or punctuated.

  How fast do their eyes blink? Next, said the gradualists, remember, these punctuated equilibrium fans are paleontologists. They see long periods of stasis and then extremely rapid blink-of-the-eye changes in the fossil record. But with fossils the blink of an eye, a stretch of time unresolvably short in the fossil record, could be 50,000 to 100,000 years. That’s plenty of time for evolution bloody in tooth and claw to happen. This is only a partial refutation, since if a paleontological blink of the eye is so long, paleontological stasis is humongously long.

  They’re missing the important stuff. A key rebuttal is to remind everyone that paleontologists study things that are fossilized. Bones, shells, bugs in amber. Not organs—brains, pituitaries, ovaries. Not cells—neurons, endocrine cells, eggs, sperm. Not molecules—neurotransmitters, hormones, enzymes. In other words, none of the interesting stuff. Those punctuated equilibrium nudniks spend their careers measuring zillions of snail shells and, based on that, say we’re wrong about the evolution of behavior?

  This opens the way for some compromise. Maybe the hominin pelvis did indeed evolve in a punctuated manner, with long periods of stasis and bursts of rapid change. And maybe the pituitary’s evolution was punctuated as well, but with punctuations at different times. And maybe steroid hormone receptors and the organization of frontocortical neurons and the inventions of oxytocin and vasopressin all evolved that way also, but each undergoing punctuated change at a different time. Overlap and average these punctuated patterns, and it will be gradualist. This only gets you so far, though, since it assumes the occurrence of numerous evolutionary bottlenecks.

  Where’s the molecular biology? One of the strongest gradualist retorts was a molecular one. Micromutation, consisting of point, insertion, and deletion mutations that subtly change the function of preexisting proteins, is all about gradualism. But what mechanisms of molecular evolution explain rapid, dramatic change and long periods of stasis?

  As we saw in chapter 8, recent decades have provided many possible molecular mechanisms for rapid change. This is the world of macromutations: (a) traditional point, insertion, and deletion mutations in genes whose proteins have amplifying network effects (transcription factors, splicing enzymes, transposes) in an exon expressed in multiple proteins in genes for enzymes involved in epigenetics; (b) traditional mutations in promoters, transforming the when/where/how-much of gene expression (remember that promoter change that makes polygamous voles monogamous); (c) untraditional mutations such as the duplication or deletion of entire genes. All means for big, rapid changes.

  But what about a molecular mechanism for the stasis? Plunk a random mutation into a transcription-factor gene, thereby creating a new cluster of genes never before expressed simultaneously. What are the odds that it won’t be a disaster? Randomly mutate a gene for an enzyme that mediates epigenetic changes, thereby producing randomly different patterns of gene silencing. Right, that’s bound to work out swell. Parachute a transposable genetic element into the middle of some gene, change a splicing enzyme so that it mixes and matches different exons in multiple proteins. Both asking for major trouble. Implicit in this is stasis, a conservatism about evolutionary change—it takes very unique macro changes during times of very unique challenge to luck out.

  Show us some actual rapid change. A final rebuttal from gradualists was to demand real-time evidence of rapid evolutionary change in species. And plenty exist. One example was wonderful research by the Russian geneticist Dmitry Belyaev, who in the 1950s domesticated Siberian silver foxes.65 He bred captive ones for their willingness to be in proximity to humans, and within thirty-five generations he’d generated tame foxes who’d cuddle in your arms. Pretty punctuated, I’d say. The problem here is that this is artificial rather than natural selection.

  Interestingly, the opposite has occurred in Moscow, which has a population of thirty thousand feral dogs dating back to the nineteenth century (and where some contemporary dogs have famously mastered riding the Moscow subway system).66 Most Moscow dogs are now descendants of generations of feral dogs, and over that time they have evolved to have a unique pack structure, avoid humans, and no longer wag their tails. In other words, they’re evolving into something wolflike. Most likely, the first generations of these feral populations were subject to fierce selection for these traits, and it’s their descendants who comprise the current population.*67

  Feral Moscow dogs

  Rapid change in the human gene pool has occurred as well with the spread of lactase persistence—a change in the gene for the enzyme lactase, which digests lactose, such that it persists into adulthood, allowing adults to consume dairy.68 The new variant is common in populations that subsist on dairy—pastoralists like Mongolian nomads or East African Maasai—and is virtually nonexistent in populations that don’t use dairy after weaning—Chinese and Southeast Asians. Lactase persistence evolved and spread in a fraction of a geologic blink of an eye—in the last ten thousand years or so, coevolving with domestication of dairy animals.

  Other genes have spread in humans even faster. For example, a variant of a gene called ASPM, which is involved in cell division during brain development, has emerged and spread to about 20 percent of humans in the last 5,800 years.69 And genes that confer resistance to malaria (at the cost of other diseases, such as sickle-cell disease or the thalassemias) are even younger.

  Still, thousands of years counts as fast only for snail shell obsessives. However, evolution has been observed in real time. A classic example is the work of the Princeton evolutionary biologists Peter and Rosemary Grant, who, over the course of decades of work in the Galapagos, demonstrated substantial evolutionary change in Darwin’s finches. Evolutionary change in humans has occurred in genes related to metabolism, when populations transition from traditional to Westernized diets (e.g., Pacific Islanders from Nauru, Native Americans of the Pima tribe in Arizona). The first generations with Westernized diets develop catastrophically high rates of obesity, hypertension, adult-onset diabetes, and death at early ages, thanks to “thrifty” genotypes that are great at storing nutrients, honed by millennia of sparser diets. But within a few generations diabetes rates begin to subside, as there is an increased prevalence in the population of “sloppier” metabolic genotypes.70

  Thus, there are examples of rapid changes in gene frequencies in real time. Are there examples of gradualism? That’s hard to show because gradual change is, er, gradual. A great example, however, comes from decades of work by Richard Lenski of Michigan State University. He has cultured E. coli bacteria colonies under constant conditions for 58,000 generations, roughly equivalent to a million years of human evolution. Over that time, different colonies have gradually evolved in distinctive ways, becoming more adapted.71

  Thus, both gradualism and punctuated change occur in evolution, probably depending upon the genes involved—for example, there has been faster evolution of genes expressed in some brain regions than others. And no matter how rapid the changes, there’s always some degree of gradualism—no female has ever given birth to a member of a new species.72

  A FINAL CHALLENGE LACED WITH POLITICS: IS EVERYTHING ADAPTIVE?

  As we’ve seen, variants of genes
that make organisms more adapted to their environment increase in frequency over time. But what about the reverse—if a trait is prevalent in a population, must it mean that it evolved in the past because it was adaptive?73

  “Adaptationism” assumes this is typically the case; an adaptationist approach is to determine whether a trait is indeed adaptive and, if so, what the selective forces were that brought it about. Much of sociobiological thinking is adaptationist in flavor.

  This was subject to scathing criticism by the likes of Stephen Jay Gould and Harvard geneticist Richard Lewontin, who mocked the approach as “just so” stories, after Kipling’s absurdist fantasies about how certain traits came to be: how the elephant got its trunk (because of a tug-of-war with a crocodile), how the zebra got its stripes, how the giraffe got a long neck. So why not, supposedly ask the sociobiologists in this critique, how the baboon male got big cojones while the gorilla male got little ones? Observe a behavior, generate a just-so story that assumes adaptation, and the person with the best just-so story wins. How the evolutionary biologist got his tenure. In their view, sociobiological standards lack rigor. As one critic, Andrew Brown, stated, “The problem was that sociobiology explained too much and predicted too little.”74

  According to Gould, traits often evolve for one reason and are later co-opted for another use (fancy term: “exaptation”); for example, feathers predate the evolution of bird flight and originally evolved for insulation.75 Only later did their aerodynamic uses become relevant. Similarly, the duplication of a gene for a steroid hormone receptor (as mentioned many chapters ago) allowed one copy to randomly drift in its DNA sequence, producing an “orphan” receptor with no use—until a novel steroid hormone was synthesized that happened to bind to it. This haphazard, jury-rigged quality evokes the aphorism “Evolution is a tinkerer, not an inventor.” It works with whatever’s available as selective pressures change, producing a result that may not be the most adaptive but is good enough, given the starting materials. Squid are not great swimmers compared with sailfish (maximum speed: sixty-eight miles per hour). But they’re damn good for something whose great-great-grandparents were mollusks.