Hare is a Professor of Evolutionary Anthropology, Psychology, and Neuroscience at Duke and a core member of the Center for Cognitive Neuroscience. Woods is the Director of the Duke Puppy Kindergarten (a longitudinal study researching the effective rearing of service dogs), as well as a writer and journalist.
Goodreads Blurb: For most of the approximately 300,000 years that Homo sapiens have existed, we have shared the planet with at least four other types of humans. All of these were smart, strong, and inventive. But around 50,000 years ago, Homo sapiens made a cognitive leap that gave us an edge over other species. What happened?
Since Charles Darwin wrote about "evolutionary fitness," the idea of fitness has been confused with physical strength, tactical brilliance, and aggression. In fact, what made us evolutionarily fit was a remarkable kind of friendliness, a virtuosic ability to coordinate and communicate with others that allowed us to achieve all the cultural and technical marvels in human history. Advancing what they call the "self-domestication theory," Brian Hare, professor in the department of evolutionary anthropology and the Center for Cognitive Neuroscience at Duke University and his wife, Vanessa Woods, a research scientist and award-winning journalist, shed light on the mysterious leap in human cognition that allowed Homo sapiens to thrive.
But this gift for friendliness came at a cost. Just as a mother bear is most dangerous around her cubs, we are at our most dangerous when someone we love is threatened by an "outsider." The threatening outsider is demoted to sub-human, fair game for our worst instincts. Hare's groundbreaking research, developed in close coordination with Richard Wrangham and Michael Tomasello, giants in the field of cognitive evolution, reveals that the same traits that make us the most tolerant species on the planet also make us the cruelest.
Survival of the Friendliest offers us a new way to look at our cultural as well as cognitive evolution and sends a clear message: In order to survive and even to flourish, we need to expand our definition of who belongs.
Humans have evolved special cognitive skills to help us cooperate.
In experiments in which food was hidden under a cup, chimps were actually pretty bad at finding food under a cup the experimenter pointed at. Dogs turned out to be much better. The author first noticed it in his dog Oreo. And it wasn't just following the movement, it worked with gaze and a variety of other manipulations. It also worked with all dogs they tried. And they didn't need trial and error learning like chimps. However, it remains unclear if dogs understand the intention of our gestures. We know they at least instinctively follow them. Why can dogs do this task but chimps can’t? Likely because we’ve domesticated dogs. Throughout history, we’ve fed and bred dogs that follow our commands. Dogs seem to have begun domesticating themselves by selective pressure being applied to wolves that started close to hunter gatherer groups and eating their garbage. Probably more animals will self domesticate as human urban population grows. This gave an evolutionary advantage to ones that cooperate well with human communication. Chimps haven’t had this evolutionary pressure; therefore, they haven’t evolved the cognitive ability to understand our gestures.
It turns out human infants can also do this task well from about 9 months old.
It shows that even as children, humans can recognize that other people can harbor knowledge and intentions beyond their own. This is called theory of mind, and it’s one of humanity's greatest achievements according to this book.
Friendliness is a genetic trait that corresponds to greater communication abilities.
In 1959, the geneticist Dmitry Belyaev moved from Moscow to the far off city of Novosibirsk in Siberia. He attempted to domesticate wild foxes. The experiment has now been continuously breeding foxes for over 50 years.
Fox domestication experiment: divided a wild population into two groups, one was bred based on how they reacted to people, what authors are calling friendliness is friendliness to people in this case. The other group was bred randomly. Interestingly, the foxes not only got friendlier to people, they developed a host of other seemingly unrelated traits like splotchy coat coloring, floppy ears, smaller teeth, and the ability to breed year round. Their bodies grew larger while their brains grew smaller. This trait clustering has been observed in many domesticated species, even those as dissimilar to use as domestic chickens bred from wild jungle fowl.
Perhaps most notably, foxes from the friendly group demonstrate a much sharper ability to communicate with humans. If you present wild foxes with the two cup test, they fail about half the time. In contrast, the friendly foxes are able to follow human gestures to select the right cup. Remarkably, they retain this ability even if they’re raised by foxes from the control group.
This clustering is noteworthy because it shows that sociability and communication skills are linked by the same genetic process. If evolutionary pressure is selecting for one, it’s also improving the other. In fact, this correspondence is also observed in other domesticated animals, such as ferrets and Bengalese finches.
What causes seemingly unrelated mammalian domestication traits like friendliness and coat splotchiness to cluster together? Migratory neural crest cells may link some of them together. These cells are involved in the development of the adrenal medullae (which are smaller in domesticated animals), as well as tail and ear cartilage, skin pigment, and facial bone and tooth development. They also involved in brain development, specifically how the brain responds to serotonin and oxytocin.
Note that the authors don’t present a full discussion on traits found in domesticated mammals. Rather, they focus selectively on the ones that fit this particular hypothesis.
The friendly bonobo shows all the telltale signs of self-domestication.
Chimpanzee society can be quite brutal. Male chimps routinely patrol their territory and will aggressively attack any interlopers. During mating season, they’ll violently assault potential mates. Female chimps often fight each other and sometimes even kill their rivals’ children.
Bonobos on the other hand, while closely related to chimps, exhibit much more friendly and cooperative behavior.
Bonobos vs chimps: bonobos have smaller brains, smaller teeth, and retain juvenile features like white tufts and play (including sexual play). They often have splotchy coloring on the lip. They also don’t have an alpha male, like chimps do. In chimps high ranking males get to have sex with ovulating females. Aggressive males have an advantage and infant killings can increase their fitness, perpetuating cycle of violence. Bonobos don't have visible ovulation. Females are aggressive together to keep aggressive males in check. Female bonobos band together, even if they're not related, unlike chimps. One theory about why they differ is that perhaps bonobos are more friendly because they have more resource availability in their natural habitat. Only high ranking chimps are guaranteed enough food, but all bonobos usually have access to food. No bonobos have ever been seen to kill another bonobo. Bonobo societal structure works so well that the most successful bonobos are far more successful at fathering offspring than the most successful alpha chimp. So, it seems like female preference for nonaggressive males may have led to enough selective pressure to create this structure.
In addition to fewer fights about mates, bonobos don’t descend into violence over food. In one illustrative experiment, scientists placed a series of bonobos alone in a room with fresh fruit and a small door keeping out another bonobo. They found that the first ape typically opted to let in the second, even if they’d never met before. Chimps, when given the same choice, just eat everything themselves.
Chimps are exposed to more testosterone in development. Bonobos have double the serotonin axon density in basal and central amygdala which is one of the first changes seen in experimentally domesticated animals. However, the authors later show that infant bonobo sexual behavior is likely related to infant bonobos having the same T levels as juvenile chimps, so it’s not a totally clear cut relationship.
And, just like the domesticated foxes, bonobos show signs of better communication and cooperation skills. When presented with a task that requires collaboration - for instance, a puzzle that requires two apes to simultaneously pull ropes to get a treat - bonobos can quickly work together. However, chimpanzees are typically not able to do this.
These distinct physical traits and more egalitarian behaviors seem to indicate that bonobos have undergone a natural process similar to domestication, but in the wild (self-domestication). Their success as a species is evidence that in some cases it must be evolutionarily advantageous to cultivate traits like friendliness and cooperative social skills.
Human evolution seems to have favored friendliness.
Our early ancestors may have undergone the same self-domestication process as bonobos. Fossil records show that the evolution of early humans may have selected for friendliness as well.
Modern humans, (Homo sapiens) are currently the dominant primates on the planet. However, as recently as 50,000 years ago, we weren’t alone. At 100ka Homo erectus was the most widespread, lasted longer than Homo Sapiens has already, made beautiful tools and was a clear hominid frontrunner. At 75ka, neanderthals were skilled hunters, making and using complex tools, buried their dead and wore jewelry. But despite these promising starts, Homo Sapiens began to pull ahead. We made projectiles like the spear thrower, started solving complex problems, and communicating in detail. We moved out of Africa, made complex social structures, painted mythical creatures, and traded things or carried them long distances.
But how? What happened? The authors think it was cooperative communication. Neanderthals could only live in groups of 10-15, but Homo Sapiens were living in groups of over 100. We can communicate with strangers about shared goals. An increased aptitude for getting along allows humans to communicate more effectively; form denser, more stable social structures; and collaborate on developing new technologies. The authors argue that friendship and community made our species more robust. They think our friendliness evolved through self-domestication, which the authors define as selection pressure on friendliness, just like in dogs and bonobos. But our friendliness has a dark side, we can unplug a threatening group from our friendliness, and we can easily dehumanize them.
Testosterone, a hormone linked with competition and aggression, also contributes to specific facial features like pronounced brow ridges and jaws. When we examine the fossil record, we find that as humans became more successful, the average brow ridge and jaw significantly shrunk, starting about 80ka. The 2D:4D ratio became more feminized, suggesting lower androgen exposure during development. Our skulls also shrunk by about 5% over the last 20,000 years, which is a trait often seen in domesticated animals.
The authors also argue that in modern faces, the fact that we have white sclera is a side effect of selecting for friendliness, as these white areas of our eyes make it easier to see the direction of someone’s gaze, a useful communication tool. However, note that there is some variance in whether chimps have sclera, some do, sometimes only in one eye. Additionally, the feature seems more common in juveniles. Finally, 70 other mammal species have been documented to have white sclera appearing in at least some individuals (Clark et al., 2023). The authors actually argue that “increasing eye contact would have promoted the expression of oxytocin, encouraging bonding and social communication” but they don’t present any evidence for this reverse inference, instead assuming it would have been true given that in some experiments, humans in a laboratory setting who make prolonged eye contact produce more oxytocin and report feeling more bonded to strangers. However, note that a review found that peripheral (serum) oxytocin is not correlated with CNS (brain) oxytocin levels under baseline conditions, so many of the findings in the oxytocin literature may be misleading (Valstad et al., 2017).
Our brains have evolved to form powerful social bonds, even with strangers.
As humans evolved to become more friendly, strong social bonds became an essential tool for survival. The authors argue that there are clear benefits to living in stable, collaborative social environments.
So, how are human bonds forged? Oxytocin may play a role. It may interact with signaling in various brain regions (medial prefrontal cortex, temporal parietal junction, superior temporal sulcus, and precuneus, what the authors call collectively the “theory of mind network”) and influence emotional reactivity. Low emotional reactivity is related to the speed at which we develop the ability to cooperate and communicate. Serotonin also increases the impact of oxytocin, as does decreased testosterone. Part of self-domestication is increased available serotonin and decreased testosterone, thus making oxytocin more effective at promoting social bonding. Note that the authors present this info in very general terms, without specifying brain regions or defining terms like “impact” or “effective.” The authors go on to claim that “oxytocin may disrupt the connection between the mPFC and the amygdala…blunting the amygdala’s fear and disgust response.” However, they later state exactly the opposite: “by binding to neurons in the mPFC, oxytocin amplifies the amygdala threat signal and blunts the response of the mPFC during social interactions.” These contradictory pieces of information highlight the importance of presenting neurobiology findings complete with their specific context, lest they otherwise become totally meaningless.
Another key tool may be behavioral inhibition, or self-control. Raw brain size correlates well with inhibition ability, and in primates, there’s more bang-for-the buck as well, as brains are also more densely packed with neurons are they grow larger, increasing computing power even more than expected given increased brain size. This self-control combined with reduced emotional reactivity may have been the key combination. It led to a runaway virtuous cycle: better tech allowed us to hunt more food with less effort, which allowed us to live in denser groups, denser groups led to even better tech etc.
This dynamic helps our species build social networks far beyond our immediate families. In essence, it allowed the creation of a new type of social category, the intragroup stranger. This is someone who you don’t know personally but still treat as part of your group due to some unifying cue (religious symbol, style of dress etc). Neighborhoods, cities, and countries - all large communities we rely on - are possible thanks to our friendliness toward intragroup strangers.
Many of our self-domestication traits may be a result of developmental stalling
Example, cockroaches and termites are closely related. Adult termites retain many of the features of juvenile cockroaches (friendliness, very powerful gut bacteria). Taken to a logical extreme, they’re the ultimate example of cooperation being tied up in their success: they’re eusocial. This type of developmental arrest has been shown in other species in which retaining a juvenile feature helps them cooperate as adults (blue streak wrasse fish retain cleaning behavior and juvenile mouth morphology for example).
The authors claim that selection for friendliness is really selection for an expanded window of social development. In dogs and bonobos, this means that key traits for social flexibility develop earlier and continue growing later.
Our strong social bonds can make outsiders seem less human.
How can we understand our barbarous impulses without abandoning our concept of friendliness as an evolutionary advantage?
To start, we must look at the downside of forming strong social bonds. While our brain’s sensitivity to oxytocin helps us empathize with other members of our group, it has an ugly side effect. Oxytocin also makes mammals more aggressive to outsiders perceived as threats (e.g a mother bear will carefully coddle her cubs but fiercely attack any nearby animal that seems suspicious).
So, while humans will extend help and support to strangers perceived as part of their group, those seen as outsiders are subject to the opposite treatment. This is partially because our brains apprehend outsiders with less theory of mind. That means we spend less time empathizing with the thoughts, feelings, and experiences of non-group members. On a neurological level, we treat them as less human. “People in the low competence, low warmth category (e.g homeless people and drug addicts) elicited more activity in the amygdala than the others had, suggesting that people felt threatened by them and were less likely to attribute a fully human mind to them.” Note that this is a big reverse inference.
Interestingly, group identity, not rule violations, best predicted the activation of what the authors call “theory of mind networks”, resulting in either tolerance or punishment.
It seems like in bonobos, they developed friendliness towards strangers, but we extend it only to strangers who share a group identity with us. In every hunter-gatherer group researchers have studied, men preemptively carry out lethal raids against outsiders.
When one powerful group oppresses another, it often depicts the outsiders as non-human animals. This is sometimes called simianization and is apparent in everything from propaganda posters to racial slurs. These depictions serve to stoke the natural tendency to hate outsiders. Simianization may be particularly effective because unlike other animals invoked to demean people (e.g. rats) great apes fall into the uncanny valley, invoking more discomfort and disgust. People are also more likely to think a dehumanized group is less sensitive to pain.
When given an Ascent of Man scale, people readily attribute less than 100% human status to many groups of outsiders. People who dehumanize outsiders are also more likely to sanction their torture or killing. Precuneus activity is correlated with whether we view a group as fully human or not, a finding the authors attribute a lot of significance to given that “the explosive growth of the PC was mainly responsible for the unique globular shape of our head, which appeared only after we separated from Neanderthals.”
Even so, violence isn’t inevitable. By cultivating empathy between groups, it’s possible to foster the feelings of shared humanity which enliven our best instincts.
We can foster tolerance with close, casual contact with others.
What made the thousands of people who protected Jews at great personal cost during WWII so committed to helping a group others had so easily dehumanized? According to sociologists Pearl and Samuel Oliner, most of these resistors shared one thing in common. They all had close friendships with Jewish people.
The dehumanizing forces that led to the holocaust have not disappeared. Unfortunately, all around the world, there are still people committed to hating and oppressing those viewed as outsiders. One prominent example is alt-right ideology. These individuals score high on measures of Social Dominance Orientation, meaning they believe some groups, such as racial minorities, are inherently inferior.
How can a society curtail such harmful views from spreading? Some argue that people with backward beliefs should be violently repressed. However, this approach could backfire. Remember, humans tend to dehumanize outsiders when they feel threatened. Heightening that feeling of vulnerability could intensify that outward hatred.
A better approach may be to create more spaces for non-threatening interactions between groups. Positive, casual social interaction erodes the negative feelings that fuel hate. For instance, college students randomly assigned roommates of different racial backgrounds are found to be much more tolerant of racial differences later in life. Likewise, in the 1940s, white residents of desegregated neighborhoods were much more likely to support further desegregation.
If we want to stop our societies from slipping further toward animosity, we need to ensure people have contact across all sorts of demographic lines. Currently, cities and towns in the United States are often spatially divided. There are rich areas and poor areas, white neighborhoods and Black neighborhoods.
However, Americans can begin to roll back this entrenched segregation by consciously changing cities to encourage interaction. It’s important to fund more mixed-income housing developments, build more inclusive public spaces, and tear down physical barriers that keep people apart. In short, it’s important to build cities where people know all their neighbors.
Our Congress is now more polarized than it has been since the civil war. The authors believe that Newt Gingrich broke party friendliness by reducing personal interactions across party lines: reducing the work week to three days so politicians don't move their families and fraternize on weekends, spear headed using unfriendly language, banned pork barrel laws so members of opposite parties can't use them as incentive, they don't need each other.
Our treatment of animals mirrors our treatment of each other.
People who foster interspecies friendships are more likely to extend kindness to other humans as well. Psychologists Gordon Hodson and Kristof Dhont studied this dynamic. They found that people who attributed more thoughts and feelings to animals also scored higher on measures of tolerance.
Another study, conducted by Brian Hare and his graduate student Wen Zhou, looked at this phenomenon even more closely. In a survey, the scientists asked people if they agreed that some breeds of dog were inherently superior to others. Those who saw truth in the statement also scored highly on measures of Social Dominance Orientation. That is, they also favored hierarchies in human society.
Clearly, how we see the animal world, and our relationship to other living creatures, has implications for how we see each other. Perhaps, if we learn to cultivate a nurturing attitude toward all nature, we’ll be more likely to tap into our greatest strength, our capacity for friendship. Again, kind of a reverse inference here.
Conclusions:
Humans have evolved to be an inherently social species. Throughout history, our ability to survive has depended on cultivating social skills, interpersonal communication, and mutually supportive communities. Unfortunately, our hard-wired tendency to bond with those close to us can also make us wary of outsiders who seem threatening to us. To build a more peaceful world, we should strive to form friendships with those around us, even if they seem different on the surface. This can perhaps be accomplished with repeated friendly contact with many different groups of people.
Thought questions:
- Did you think the authors presented a convincing case that humans were self-domesticated? What were the strongest pieces of evidence they presented? If you’re unconvinced, what evidence might convince you?
- Do you think that using ways that humans communicate (e.g. pointing) as benchmarks of cooperation in general is appropriate? It seems like that makes sense in species that evolved to communicate with humans specifically (dogs, domestic foxes), but does it make sense to apply this to chimps?
- Do you think the self-domestication hypothesis could be approached genetically? Could we formulate a polygenic score that predicted domestication phenotypes across species and use that to evaluate whether humans had a high domestication score?
- Would you rather live in your current society or as a bonobo in a bonobo society? Why or why not?
- What are some ways we might be able to increase frequent casual contact with people outside our “in group” both in the design of cities and in our personal lives?
- What are your own most salient identity markers? In other words, how do you evaluate whether someone shares an identity group with you or not?
- Did the evidence the authors presented on how readily humans dehumanize others surprise you? Why or why not?