Humans are an astonishing animal. We are the descendants of African apes who stood up and walked across the world. We're not particularly fast, we're not particularly strong, and we're not particularly well equipped to thrive in the many places where we live. Our bodies don't have enough fat or fur for Canada's winters, enough water for Australia's deserts, nor the right proteins for processing many wild foods. And yet, despite our physical limitations, we've built towers that touch the sky, rocketed robots to Mars, and connected our planet through a worldwide communication network. How on earth did we do it?
When most animals encounter a new environment, they're forced to genetically adapt: developing powerful muscles to outrun local predators; fur and fat to prevent freezing; proteins to make plants less poisonous. But genetically, humans have changed very little. The secret to our success is not by genes alone.
Instead of evading local predators, we hunted them. Instead of evolving fat and fur, we wore the pelts of our prey. We didn't process poisons in our bodies, we processed the plants before we ate them. And we did all this by developing tools, techniques, and traditions. We did all this thanks to culture.
Six thousand years ago the first cities were established. Cities then, just as they are today, were a hubbub of human activity. All those people living side by side bickering and bartering with a continual supply of energy, resources, permanent migrants and temporary travelers created a hub in which ideas met and innovations emerged. City dwellers benefited from those innovations and began to consider themselves civilized. Rather than attribute their ways of thinking and technologies to the density of people, they assumed they were something special, in contrast to those around them – the barbarians.
Around 1000 bce the Chinese of the Yellow River Valley distinguished themselves as Huá – the civilized – from the Yí – wild, uncivilized barbarians – that surrounded them. Around the same time the ancient Indians of the Indus Valley considered themselves surrounded by Mlecchá – inferior barbarians. Later, at around 500 bce, Greeks whom both the Indians and Chinese would have considered barbarians, considered themselves civilized in relation to the surrounding foreign-speaking bárbaros, from which the English term ‘barbarian’ is derived. Those barbarians included the Romans, who later considered all non-Romans barbarians, from the trouser-wearing French Gauls to the forest-dwelling German tribes.
The Romans held an even lower regard for those further afield. Roman statesman Cicero, writing to his friend Atticus in the first century bce, worried that there was little to be gained from invading Britain since ‘there isn't a pennyweight of silver in that island, nor any hope of booty except from slaves, among whom I don't suppose you can expect any instructed in literature or music’. No doubt the Aztecs and Mayans of Central America and the Ghanaian, Mali, Yoruba, Great Zimbabwe, and other empires of Africa felt the same way about their neighbors. As people began to communicate with those beyond their immediate borders, we see the same attitude. Ironically, it was often those once labeled barbarians who, upon becoming the nouveau riche in innovation but lacking any memory of the past, began labeling the formerly civilized as barbarians. So the tides turn.
Writing in 1068, Said al-Andalusi, the mathematician, scientist, and qadi (sharia judge) of Toledo in Spain (then under Muslim rule), divided the world into the civilized who concern themselves with knowledge and higher learning, and everyone else. The civilized included the Arabs, Chaldeans, Egyptians, Greeks, Indians, Jews, Persians, and Romans. The rest he considered barbarians, with special mention for the Chinese and Turks as the ‘noblest of the unlearned’.
Today, the terms ‘civilized’ and ‘barbarian’ have fallen out of fashion, but the same attitudes remain. Americans of the east and west coastal cities consider themselves an educated urban elite distinguishable from the rural, uneducated, fly-over states in the Midwest. Western, educated, industrialized, rich, and democratic (WEIRD) societies consider themselves developed compared to the still-developing nations of Africa, South America, and Asia – the majority world countries.
Humans in culturally and technologically more complex societies have also differed in their attitude as to whether ‘primitive people’ can be civilized. The Romans considered the Gauls more civilizable than the Germans. The British considered Indians more civilizable than Africans. Rudyard Kipling's White Man's Burden encouraged Americans to annex the Philippines and civilize the ‘half devil and half child’. In contrast, Senator Benjamin Tillman of South Carolina, who was against taking up this civilizing mission, argued that the people of the Philippines were ‘not suited to our institutions’. But while there was debate as to how the civilized should intervene in the lives of the uncivilized, what was agreed on was that some people were ‘primitive’.
Let's start by unravelling this human universal thinking and then continue by explaining how and why people differ over time and geography.
People around the world have different tools, techniques, traditions, and other aspects of culture to deal with the different problems they face. The systematic documentation of these solutions began with Franz Boas, a German physicist who would come to be known as the father of modern cultural anthropology. Boas was a child of nineteenth-century Europe. He was raised on the assumptions of the time, such as that of the hierarchy of societies from ‘savage’ to ‘civilized’ with Europeans at the pinnacle and everyone else somewhere below. These views were formalized in the cultural anthropology of the time. Another key figure was Edward Burnett Tylor, who was the first to offer a definition of culture that is often still used today: ‘Culture or Civilization, taken in its wide ethnographic sense, is that complex whole which includes knowledge, belief, art, morals, law, custom, and any other capabilities and habits acquired by man as a member of society.’
At the bottom of Tylor's stages were ‘savages’, which literally meant people who lived in the wilds or in the woods, living off the land without agriculture. The Inuit, for example, were savage according to Tylor. Slightly better were ‘barbarians’, such as the Arabs – the same Arabs who would have previously considered Tylor and his fellow Englishmen to be barbarians. For Tylor, the pinnacle of civilization was of course to be found in Europe. To Tylor's credit, he rejected the Victorian belief in a hierarchy of races, believing instead in the universality of humankind. For Tylor, other societies weren't stupid in that they were lacking intelligence, but rather were lacking the knowledge needed to climb the hierarchy of civilization, which in the fullness of time they might achieve. In the late nineteenth century these assumptions about the hierarchy of humans began to be questioned. And this questioning would eventually lead to a more thorough understanding of the human animal and the critical role of society.
In 1883, hoping for adventure, with a head full of such ideas and a freshly minted doctorate in physics, the twenty-five-year-old Franz Boas boarded the Germania on a trip to Baffin Island in the far north of Canada, just west of Greenland. The Germania's mission was to evacuate German scientists from the polar meteorological station. What Boas encountered on Baffin Island wasn't successful scientists living next to ignorant savages, but scientists who struggled to do anything without the Inuit. It shook his sensibilities. Despite coming from a society with a larger energy budget and complex technologies – it was, after all, the Germans visiting the Inuit and not vice versa – German technology and culture were ill-suited to the Arctic. In contrast, the Inuit had deep knowledge about how to live and thrive in this frozen, unforgiving environment. They knew the movements of the caribou and other animals, how to trap and hunt them, which had the best skins, and how to turn the skins into the warmest of clothing. Most impressive to Boas, the landscape of the world's fifth largest island, which seemed plain and formless to him and his fellow Germans, was to the Inuit rich in historical and navigable detail.
Boas was so fascinated that he gave up physics and spent the rest of his career trying to understand the ways in which the Inuit lived, survived, and thrived, eventually training his students, including Margaret Mead and Ruth Benedict, to do the same in different societies around the world. Rather than theorizing and speculating about faraway lands or studying long-lost ancient civilizations, these were new ‘cultural anthropologists’, living among people in different societies. They systematically documented how the people in each society were surviving and thriving in vastly different environments using vastly different tools, technologies, societal organizations, and traditions. Each society had different solutions for food, for shelter, for defense, for raising children, for maintaining social harmony, and for all the other problems all humans everywhere need to solve. But how did these societies figure out how to survive and thrive in such a range of environments in the first place?
One obvious answer was emerging in evolutionary biology. Charles Darwin had published On the Origin of Species in 1859 and The Descent of Man in 1871. Darwin himself knew nothing of genetics, but in 1900 – nearly two decades after he died – the careful pea plant breeding experiments of the obscure Augustinian monk Gregor Mendel were rediscovered. Mendel's work had revealed the existence and effect of genes, opening the space for a new field: population genetics.
Scientists, mathematicians, and statisticians spent the first half of the twentieth century reconciling Darwin's theory of ‘natural selection’ with genetics and the rest of biology in what became known as the Modern Synthesis. For example, how could it be that traits were transmitted in discrete genes, but when you looked at something like height, it was a bell curve? Mendel's experiments had revealed a mix of tall and short pea plants, with nothing in between. Shouldn't there be a mix of tall humans and short humans with not much in between, like Mendel's mix? Questions like these had to be mathematically and experimentally reconciled: the answer in this case is that most traits are polygenic, meaning that many separate genes all contribute to height, leading to what looks like a continuous bell-shaped distribution.
The Modern Synthesis forever changed the biological sciences, laying down mathematical theories, theoretical frameworks, and empirical evidence for how all animals developed solutions for food, defense, reproduction, and group living – they did it primarily through genetic evolution. Genes that helped an animal survive better persisted at the expense of genes that did not. And yet, this didn't seem quite enough to explain the success of humans in so many different ecosystems. For sure, humans in different places had different hair, skin, eyes, and noses that may have been adaptations to these ecosystems, but was this what explained their success and different solutions for survival?
Take a cheetah to the Arctic and she will shiver. Take a polar bear to the tropics and he will overheat. But Boas's anthropologists could learn from their indigenous hosts and, with their guidance and knowledge, live and survive despite having very different genes. Indeed, stories of lost European explorers, such as Robert Burke and William Wills in Australia and John Franklin in the Arctic, were appearing around the world. The difference in their survival seemed to be down not to their individual abilities or European technologies, but to their ability to access local people and local knowledge – how to find and process local foods, what plants to look for and which to avoid. There were no primitive people as envisioned by Tylor and others. People around the world were not succeeding because of what was in their genes, but because of what was in their societies. The peculiar cognitive skills and achievements of city dwellers were not due to some general difference in ‘cleverness’. Instead, these specific skills and achievements arose from their access to ideas, metaphors, knowledge, and ways of thinking in larger, interconnected societies.
But how did their societies figure it all out? How did the descendants of an African ape end up learning how to survive in the Arctic? The answer to this question would finally emerge at the intersection of evolutionary biology and cultural anthropology.
The human animal is like any other animal in many ways. Humans eat, poop, mate, reproduce, and then die. But in terms of other dimensions, such as brain size or size of societies, we are an outlier. How this happened or why it didn't happen for other animals isn't clear, but it seems obvious that somehow evolution selected for intelligence in our species. Philosophers of the nineteenth century described humans as the rational animal. Humans were capable of causal reasoning and understood their world in a way that other animals did not, they said. But new experiments in biology and psychology in the late twentieth and early twenty-first centuries, pitting humans against their closest cousins, chimpanzees, revealed that ‘man the rational animal’ might be an illusion. Economists were coming to the same conclusion at the same time, while going through a behavioral science revolution.
Primatologists study non-human primates. Developmental psychologists study human children. Scientists in these two disciplines often collaborate on experiments that pit human children against young chimps and other apes. For example, one set of experiments gave children, chimps, and orangutans a battery of cognitive tests – think of it as an ape IQ test. Half the test involved physical cognition, ranging from having to mentally rotate an object to adding quantities, to causal reasoning, to understanding the functional and non-functional properties of tools. The performances of the children and chimps were indistinguishable; the orangutans did a little worse, but not by much. The other half of the test involved social cognition and ranged from learning a solution from someone else to understanding and producing communicative cues, to figuring out what someone else is trying to do. Here there was no competition, the children beat their chimp and orangutan cousins, particularly when it came to learning from others. The kids killed it when it came to social learning.
The results, published in the journal Science in 2007, confirmed what the sapiens in Homo sapiens really means: ‘wise’. But we're wise not because we're particularly rational, logical, or good at causal reasoning. We're wise because wise people learn from others. Our species learned, as my collaborator Joe Henrich once put it, that it's better to be social than smart. But it's more than that – by being social, we could become smart.
Based on graph and data from: E. Herrmann, J. Call, M. V. Hernández-Lloreda, B. Hare, and M. Tomasello (2007), ‘Humans Have Evolved Specialized Skills of Social Cognition: The Cultural Intelligence Hypothesis’, Science, 317(5843), 1360–6.
One experiment that clearly illustrates this social learning difference took place in two very different locations. The first is St Andrews, Scotland, the birthplace of golf and home of the University of St Andrews, where Prince William met his future wife Catherine (Kate) Middleton. The second is Ngamba Island Chimpanzee Sanctuary, Uganda, home to around fifty or so chimpanzees that have been rescued from across East Africa.
In 2005 two researchers from St Andrews – Victoria Horner and Andrew Whiten – gathered a group of young Scottish children and a group of young Ngamban chimps and gave them the choice to copy or think for themselves. The experimenters presented both groups with a black box. The box had a hole on the top and a hole on the side. Inside the box was a reward: a piece of fruit for the chimps, a sticker for the children. The experimenter showed the chimps and the children how to get the reward by poking a stick through the top hole and then the side hole. They then handed the stick to the chimps.
If you've ever seen videos of chimps speaking with a language board, scrolling Instagram, or whizzing through a working memory task, you'll know that they are smart. They imitated the experimenters perfectly, first poking the stick through the top and then through the side to get the fruit. Happy chimps.
The experimenters then did the same thing with the children – showing them how to poke the stick through the top hole and then the side hole. The children, just like the chimps, poked the stick through the top and then through the side to get the sticker. Happy children.
Then came the key treatment condition. The experimenters replaced the black box with a clear, transparent box that was otherwise identical. Because the box was clear, both the chimps and the children could now see that the first action – poking the stick through the top hole – did nothing. In fact there was a separation inside the box, so only the side hole accessed the reward. The top hole was irrelevant. But again, the experimenters poked the stick through the top hole and then through the side hole. They handed the stick to the chimps. What did the chimps do? Chimps are smart! They ignored the top hole and just retrieved the fruit from the side hole. They engaged in what scientists call ‘emulation’ rather than ‘imitation’.
Then, once again the experimenters poked the stick through the top hole and then through the side hole. They then handed the stick to the children. What did the children do? Children are smart! Or are they? The kids continued to poke the stick through the irrelevant top hole and then through the side hole.
It's not that the children didn't understand the causality – later experiments confirmed that they understood it just as well as the chimps. But the children assumed that the adults knew something that they didn't. So instead of emulating and trying to reverse engineer the reasons for the adults’ actions, they simply imitated. They copied all actions because they assumed that the seemingly pointless two-poke method is just how people do it. Human children ape better than apes do. And in so doing, the human children weren't making a mistake, they were becoming brilliant.
By copying successful behaviors, beliefs, tools, and ways of thinking from the previous generation, even without understanding why these work, human children engage in a process unique to our species that allows them to surpass the limits of their own cognitive abilities. It is a cultural evolutionary process that has led to antibiotics, democratic governments, and nuclear reactors. It has allowed humans to build on each other's work and to take for granted the work already done in the past in order to narrow the set of things we presently need to develop and innovate on. Chimps, on the other hand, still try to figure out everything on their own, and so, as comparative psychologist Bennett Galef famously put it, still sit naked in the rain.
In this case, the children were mistaken about one thing: they assumed that the adults were demonstrating the best strategy. In fact the adults were deliberately inefficient for the purposes of the experiment (which also hints about movements and moments of mass human folly, which I will discuss later).
But in the world beyond this experiment, and by and large as far as human progress has been concerned, children learn from relatively wise adults who, when they themselves were children, imitated their adults, who when they were children imitated their adults, and so on back in time. Each generation of children, the next generation of adults, thereby acquired a head full of successful recipes – tools, techniques, and traditions. Not through understanding but by selective trust. Our lives are filled with acquired recipes, the origins of which we have long forgotten. This reliance on socially transmitted information is, in essence, a shortcut to brilliance.
When we cook, we cut the meat a certain way and add salt at the beginning or end, because that's how our parents taught us to cook, or because that's what Gordon Ramsay told us to do (no one wants Gordon swearing at them). You don't need to understand the biology or chemistry of cooking to produce a tasty, cooked meal full of predigested and bioavailable calories and nutrients. We check our emails on computers, a technology that, for many, may as well be magic. A sequence of clicks takes you to your inbox, crammed with more emails than you can get through. If the Wi-Fi goes down, turning it off and on again might fix it. There's no need to understand how a processor, software, and random access memory produce an image on the screen or how your router can use radio waves to show you Facebook.
The children were right to copy the adults but were wrong to assume that the adults knew what they were doing. The funny thing is, although adults often don't know what they're doing or why, they often feel like they do. They have what psychologists call an illusion of explanatory depth.
The illusion of explanatory depth is perhaps one of the most profound but lesser-known discoveries in psychology. It's so profound and impossible to irradicate that even after reading this section you might not believe it. But I feel compelled to mention it before we continue. The essence of the bias is that we all think we know more than we actually do. We assume we understand and have reasonable causal models for our beliefs, behaviors, and the technologies we use – it threatens our sense of self to think otherwise. The illusion is only shattered when we're tested. When we're asked to explain specifics.
In the now classic experiments, researchers first asked people if they understood how a flushing toilet works. Flushing toilets are a centuries-old technology that, with a little luck and enough fiber, you interact with every day. You probably have some sense that you understand how they work. But think about it – if someone asked you, how does the water in the bowl flush everything away and return to the same level? – could you explain it? Could you make a toilet without copying another one? In expounding the details, you might realize your understanding is shallower than you initially believed. Or take a wine bottle and turn it upside down. You'll see a punt on the bottom – any idea why? In other cases, you might even have explanations you think you understand. For example, public health messaging warns us to complete our courses of antibiotics. Why? To prevent superbugs. But surely, since not all bacteria will die, the fittest that survive will be the most antibiotic resistant, no? Surely finishing your antibiotics is what would lead to superbugs!
If you look up the answers and learn about toilet pressure differentials, the geometry and history of wine bottles, and selection pressures on antibiotic kill curves, you will once again feel like you understand the mechanisms. Indeed, the answer may feel obvious in hindsight. This feeling too is part of the same illusion. You can't shake it and some people refuse to accept it. The illusion of explanatory depth is a quirk of our psychology that is almost impossible to overcome.
Right now, you may be sitting on a chair or lying on a bed or driving a car that you could not recreate from scratch (processing wood, mining ore and making metal, creating and then weaving fibers, and so on). You are reading a book whose production process is more complicated than you think. This illusion has many implications.
One implication is that other people's problems have a seductive reductiveness. The solutions to your friend's relationship or career problems seem so obvious to you. Perhaps even the solutions to societal problems, especially if they're another society's problems. In all these cases, your psychology shields you from the true complexity of any issue.
All of this is to briefly illustrate that the world is not only complicated but even more complicated than our psychology allows us to believe. But if we don't have these causal models then who does? Where did the Inuit, Wall Street bankers, Silicon Valley superstars, or any of us for that matter, get all the information that allows us to succeed? Where did it ultimately come from? How do we know what to do?
Three sources of information guide our behavior. First, there's what you learn from your own life experience. In the 1930s Harvard psychologist B. F. Skinner put rats into boxes and rewarded them when they pushed one lever and shocked them when they pushed another. The rats quickly learned which lever to push. Skinner laid down the laws of reinforcement learning, laws which would eventually lead to modern machine learning.
The process is simple: if you want your dog to poop in the right place, reward them when they do so and scold them when they do not. All animals learn through this process of pleasure and pain. When we touch a hot stove, we get burned and we don't do it again. When a pick-up line lands, we might try it again. But we're not blank slates. We're born with preferences. Even without reinforcement, rats and humans will naturally prefer sweetness over bitterness: bananas are tastier than Brussels sprouts. And that's thanks to genetic evolution.
Genetic evolution is the second source of information. It is also a kind of reinforcement learning, but one in which the delay on feedback from actions is too long for individual reinforcement learning to learn. Sweet compounds predict nutrients and bitter compounds predict poisons, but by the time you learn that rule, through trial and error, you'd be dead. And that's what happened – mammals that coded sweet sugars as tasty and bitter alkaloids as not tasty were rewarded with better health, longer lives, and more offspring than those that did not. And so those preferences persisted over generations all the way to us.
All animals, including us, have a genetic inheritance from our parents. Things that historically helped us survive, thrive, and reproduce feel pleasurable, and things that might harm, kill, or end our lineage feel painful. In humans, these genes affect, for example, the shape of our eyes, color of our skin, blood type, ability to drink milk in adulthood (unique to a minority of human populations and unique among all mammals), and our propensity for various diseases. These genes are also correlated with cognitive and behavioral differences.
But the law of evolution discovered a new way to use reinforcement learning in humans: not just genes, not just individual learning, but an extensive cultural inheritance.
When a baby is born, it spends the next two to three decades catching up on the last several thousand years of human history. Sometimes, we need to override our genetic predispositions – we need to learn to like the bitterness in broccoli and Brussels sprouts. We spend our earliest years acquiring beliefs, values, norms, traditions, technologies, habits, and behaviors – in short, culture.
Culture is socially transmitted information; when we communicate, we're creating and evolving culture. When we salaciously gossip about people's love lives and life choices, secretly judge or copy their parenting style, or give and withhold hearts on social media; when we learn something new or decide to try something different; when we inspire others or are inspired by others – we are changing culture. This realization that humans inherit both genetic and cultural information became known as dual inheritance theory. The rules by which culture evolves became known as cultural evolution. The mathematical theory of our dual inheritance and cultural evolution began with the question, when will natural selection lead to the evolution of learning from others?
Genetic evolution leads to genes adapting to environments, but the math says that socially learning from others will be favored not for a particular environment but when the environment is changing. Specifically, when it is changing a little, but not too much. Like in the fairy tale Goldilocks and the Three Bears, the Goldilocks zone of environmental variability is not too stable and not too unstable. This Goldilocks zone is dependent on an animal's lifespan and generation length, but when the environment fluctuates for long enough within this zone, any animal would focus on what others were doing rather than learning by themselves or blindly following their genetic instincts. We can understand one of the key insights from this model without diving into the math.
When the environment is highly stable – that is, when things are not changing over long periods of time, then genes and genetic evolution are the best way to adapt. Consider something like the amount of sunlight, which varies seasonally but is otherwise stable as a function of latitude (i.e. there's more sunshine in Australia than Austria). Here a genetic solution would be favored, as is the case with human skin color, which optimizes for UV radiation. If you're a dark-skinned person living in Europe, you risk vitamin D deficiency without sufficient supplements, and if you're a light-skinned person living in Australia, you risk skin cancer without sufficient sunscreen. Human skin color is a genetic adaptation to stable levels of sun exposure. The effects of sun exposure are too delayed to learn individually or culturally.
At the other extreme, if the environment is highly unstable then this selects for large brains and trial-and-error learning. Today the water is here, tomorrow it's there. Today we eat blue berries, tomorrow they're gone and we need to figure out if those red berries are edible. The environment is changing too quickly for both genetic evolution and the knowledge of our ancestors and so we must rely on individual reinforcement learning – trial and error. But what if the environment is changing too quickly for genes but too slowly for individual learning? Between these two extremes, where the environment fluctuates in a way that matches generation length, is where your parents and grandparents have some knowledge worth learning.
Imagine you're a child living in an ancient society. We haven't yet invented physics, chemistry, biology, economics, or psychology. Your society goes through a cyclical drought. You may have never experienced a drought and maybe your parents haven't either, but Grandma remembers that when she was a child there was a drought and they went left of the mountain, beyond the forest, and found water, and so with this knowledge, she leads her tribe to safety. Under these conditions, past generations have knowledge worth paying attention to, and so it pays to socially learn. Indeed, this is also why respect for the elderly decreases during times of rapid technological and cultural change. When the world is quickly changing, old knowledge becomes out of date, losing the elderly their place in society as founts of knowledge.
Without this ancestral optimally variable environment, people wouldn't care as much about what others were doing. Instagram would never have been invented. This reliance on social learning moved us from relying on causal reasoning to relying on cultural reasoning.
Sometimes what it takes for science to move forward is to let go of assumptions. Earth, for example, looks pretty flat and pretty stable, and our eyes clearly see a sun tracing the sky from east to west. But when we let go of the assumption of a flat earth and moving sun, we get a better model of the solar system. And with this knowledge we can go to Mars. Time feels like it ought to flow the same for all people and all places, but when we let go of that assumption, we get a better model of the universe. As the 2014 movie Interstellar correctly taught us, time literally moves faster at higher altitudes. And with this knowledge we can build a global positioning system (GPS). The biological world looks pretty chaotic and humans are clearly set apart, but when we let go of that assumption, we get a better model of nature. And with that knowledge we can edit genes and engineer messenger ribonucleic acid (mRNA).
The seemingly obvious assumption that has held back the human and social sciences is the assumption that human uniqueness is grounded in human intelligence. Don't get me wrong, we are bright. But not because of our ability to reason. We can reason to some degree, but without culture we can't do it much better than other animals. I imagine you're at least a little skeptical at this point – this assumption is hard to see past, but persist if you can. Because when we let go of the assumption that human intelligence is a product of our hardware and not our software then we get a better model of ourselves. And with that knowledge we not only get to a theory of everyone, but we can use that knowledge to build a better society.
Within economics, the much maligned ‘economic man’ or Homo economicus was based on nineteenth-century philosophical views about humans as the rational animal, mathematically formalized in the twentieth century using physics-inspired approaches complete with metaphors for friction and elasticity. But that assumption is wrong.
At least three Nobel Prizes in Economics have been awarded to work revealing different ways in which these rational assumptions fail to describe actual human behavior. Herbert Simon won the Nobel Prize in 1978 for his theory that rationality must be bounded by limited information, limited time, and limited cognitive capacity. We just don't ever have enough information, the time to figure out all the possibilities, or even the brain space to be truly rational. Daniel Kahneman won the Nobel Prize in 2002 for his work with Amos Tversky on prospect theory, a particularly prominent example of seeming irrationality – loss aversion – that people are more averse to a possible loss than they value the same possible gain. If I offered a flip of a coin to win $100,100 or lose $100,000, most people won't take the coin flip despite odds in their favor. Most recently, Richard Thaler won the Nobel Prize in 2017 for his work on nudging – applying our irrational biases to understand economic decision-making and applying these insights to successfully change behavior. Organ donation rates on death, for example, were close to 100% in Austria but around 12% in Germany. That's not because Austrians are nicer than their German cousins (I'll leave it to my German-speaking friends to debate this). The difference is simply that in Germany people have to opt in to donate their organs whereas Austrians had to opt out to not donate their organs. The difference was the default option. People usually stick with the default for decisions that don't have an impact on their immediate lives.
If you don't believe that you're irrational, perhaps you can at least believe that other people are. We often feel like we're the only rational person in the room. And so does everyone else. This body of work in behavioral economics is a slow realization that even if it feels like we're rational (and simultaneously that others aren't!), our behavior is not really rational, at least not in the sense of individually maximizing utility or being consistent in our preferences. Instead, we are a product of evolutionary rationality – fitness maximized and preferences tuned by millions of years of genetic evolution, thousands of years of cultural evolution, and a short lifetime of experience.
Herbert Simon won his Nobel Prize for realizing that we have limited information, limited time, and limited cognitive capacity. But long before Simon, evolution realized it too. Since we didn't have information, time, or intelligence to compute the answers by ourselves, we distributed the computation across the population and solved the answer slowly, generation by generation. Then all we had to do was socially learn the right answers. Using this approach, we could limit our reason and causal understanding to minor tweaks with partial causal models of the world. You don't need to understand how your computer or toilet works, you just need to be able to use the interface and flush. All that needs to be transmitted are which buttons to push – essentially how to interact with technologies rather than how they work. And so instead of holding more information than we have mental capacity for and indeed need to know, we could dedicate our large brains to a small sliver of a giant calculation. We understand things well enough to benefit from them or attempt to make improvements, but all the while we are making small calculations that contribute to a larger whole – like a wisdom of the crowds. We are just doing our part in a larger computation for our societies’ collective brains. Yet sometimes our societies’ wisdom of the crowds is instead madness of the masses.
This evolutionary rationality doesn't always arrive at the right answer. The answer can be mismatched, leading us to overweigh losses relative to gains, overeat cake over cauliflower, and over-rely on information we hear a lot. And these sources of information can give different signals – genetic evolution doesn't necessarily want you to be happy, your culture might lead you to fewer children, and your lifetime of experience might find shortcuts to pleasure from pornography to Pringles that simply don't optimize genetic, cultural, or even physical fitness. But on average, evolutionary rationality is or was adaptive. (We'll return to this shortly.)
Is it really true that we can't causally reason? On the one hand, experiments show that even simple feedback systems are difficult for people to disentangle. On the other hand, some researchers argue that the difference between what artificial intelligence can do and what human intelligence can do is causal reasoning. Experiments with children seem to show that they can discover simple causal patterns sometimes better than adults. And certainly it feels like we're capable of what Kahneman called System 2, deliberate causal reasoning. But at the same time, we're flush with systematic mistakes. There's a reason why students are warned that correlation isn't causation – we tend to assume temporal precedence or non-causal association is causal, which leads us to superstitions and repeating what we did last time. And there's a reason we try to make people aware of common logical fallacies like ad hominem – we naturally evaluate the person behind the argument as much or more than the argument itself. And there's a reason psychology gave up on Freudian introspection – we are just not a good guide to our own thinking. Most humans are not logical analytic philosophers, but we do have a limited form of causal reasoning. It's a messy kind of logic called enthymemes.
Enthymemes are a form of cultural reasoning: pseudo-logical reasoning with unstated premises filled by assumptions shared by others in our society or culture; assumptions that we have inherited from generations gone by. Take the classic deduction ‘All humans are mortal. Socrates is human. Therefore Socrates is mortal,’ as an example. This is something Star Trek's favorite Vulcan, Mr. Spock might say, but most humans won't. Instead, humans reason along these lines: ‘Socrates is mortal because he's human.’ The unstated premise is that humans are mortal, which we learn vicariously even before we first experience a death. We do this all the time. ‘She is sick since she has a fever’; ‘Benjamin is a typical Canadian therefore he is polite.’
We find cultural causal reasoning intuitive in a way that formal logic is not. For example, this is difficult: ‘If it is a Monday, the local band wear their uniform. The local band is wearing their uniform. Therefore it is a Monday.’ Is that logical? This version with the same structure is much easier: ‘If I get wet, I need to change my clothes. I changed my clothes. Therefore I got wet.’ When restated in cultural causal terms, the illogical nature of both statements becomes more obvious.
An experiment that reveals the difference between causal reasoning and cultural reasoning is the Wason selection task. Given the following choice, which cards must you turn over in order to test the truth of the rule ‘If a card shows an even number on one face, then its opposite face is a star’?
If you haven't seen it before, try it. No, seriously, give it a go and remember your answer.
When the same logic is stated in cultural causal terms, the answer is immediately obvious. Which cards must you turn over to test the rule ‘If you are drinking alcohol, then you must be over twenty-one’?
Incidentally, the typical answer to the first case is the number 8 card and the star card. The correct answer is the number 8 card and the triangle card since you're looking for a violation of the rule to falsify it. In the second case, it's more obvious that you need to check the beer drinker and the sixteen-year-old. The difference between the two cases is typically discussed as a tendency toward detecting cheaters – that humans have an evolved tendency toward seeing who is cheating, in this case drinking alcohol when they shouldn't be, but I argue that it's part of a more general psychology of cultural reasoning, in this case about norms.
We frame our arguments in enthymemes and cultural causal logic all the time. Scientists do too and will often fail to realize the degree to which they are doing this until they try to model their theories using mathematics or computation. Suddenly all those assumptions and all that unstated and culturally shared logic in our verbal theories has to be precisely stated. The unstated invalid assumptions and leaps of logic become transparently obvious. Which is why many sciences favor formal mathematical and computational models in the development of theory, which force the modeler to write down all assumptions and logic.
Nonetheless cultural evolution has improved our ability to use logic over centuries and transmitted that improved formal reasoning ability through our education system. Everyone around us is educated, so it leads to the impression that the kind of formal rationality we value in our society is what makes us different from other animals. In other words, thanks to schools, we're a lot better at actual formal causal reasoning than we once were. To see this, we need to look at populations without schooling.
In the 1930s Russian psychologist Alexander Luria wanted to understand the effects of education on cognition. He headed to Uzbekistan, where an educational revolution was occurring: schools were being introduced to farming communities with no previous formal education. Luria ran some simple tests. What he found shocked many people.
We often consider reason and logic to be a part of what separates us from other animals. (In fact, this too is not quite true – even fish and fruit flies have some logical reasoning abilities.) Luria tested simple logical propositions such as ‘If p then q’, famously explaining once that ‘In the far north, where there is snow, all bears are white.’ As part of his test he asked participants the following: ‘Novaya Zemlya is in the far north, and there is always snow there. What color are the bears there?’ This was a simple question readily answered by even young children in our society and individuals with access to education in Uzbekistan. But this simple question was met with puzzlement from those not exposed to formal education. Here is a typical response: ‘I don't know. I've seen a black bear; I've never seen any others.’
Luria's results are hard to believe and, in any case, the early twentieth century didn't have the fancy experimental controls and statistical methods we now have. So anthropologist Helen Elizabeth Davis and I decided to run the same experiment among the herding Himba people of Namibia and Angola. We found what Luria found.
The Himba, who never went to school, behaved the exact same way as the unschooled Uzbeks. For example, Helen asked them: ‘There is a country in which boats are made of sand. I have a boat from this country. What is it made from?’ The answers from those without formal education were similarly ‘I don't know’ or ‘Probably wood’. The willingness to deal with hypotheticals and ease of reasoning in this way is a taught skill, culturally evolved and transmitted through our education system. The Himba who never went to school can reason, but the way they reason is different from the Himba who went to school. Today, most people we know have some amount of elementary and high-school education, and these concepts and ways of thinking are embedded, reinforced, and implicitly taught in TV shows, books, and how people around us speak. Children emerge in this world and readily learn to reason more formally. This ubiquity of more formal reasoning in our society tricks us into believing that humans are naturally rational animals.
Innovations occur as an evolutionary process, a giant calculation of the collective brains of our societies. A key feature of this system is not just social learning but knowing who to socially learn from.
Let's return to that child living in an ancient society before modern science and technology. In that child's village lives a hunter. Let's call him Bruce.
Bruce isn't just any hunter, he's the best hunter. He's strong and well respected. Bruce brings in the most game and has the biggest house. Men are envious, women have crushes, and all the little kids want to be like Bruce. But the trouble is, they don't know exactly what it is that makes Bruce successful. There's no evidence-based, peer-reviewed study of successful hunters; no business or self-help books to tell them how to unlock their inner hunter. The kids just know that Bruce is cool.
They may use their informal cultural reasoning to hypothesize that some things may be more related to Bruce's success – the weapons he uses, the time of day he goes hunting, or where he hunts. But the world is complicated and contains many complex causal relationships. Bruce's success may be because of his weapons and training, but could also be because of the shoes he wears, the beard he sports, or the gods he worships. The best strategy a little hunter can adopt is to copy everything about Bruce and hope that maybe they will grow up to be just like him.
That psychology, a pay-off strategy, is still with us today. It's why celebrities have undue influence in areas that have nothing to do with their expertise. It's why Beyoncé's preference for Pepsi or perfume might increase sales despite neither what she drinks nor how she smells having anything to do with her rise to fame. It explains the Kate Middleton effect, where dresses worn by the Princess of Wales quickly sell out in Britain. And it's why you once believed fat made you fat (which makes sense: eat fat get fat) and then believed that sugar made you fat (which also somehow makes sense: unused calories are stored as fat) and now maybe believe it's calories overall. Or why you floss – or at least tell your dentist that you floss – despite a weak evidence base for its effectiveness. Or, indeed, it's why you hold many beliefs, about eating locally (often a higher carbon footprint because transport costs are quite low), recycling (often not economically worth it and so ultimately ending up in the landfill), supporting or not supporting big-game hunting (often an old or problematic animal is killed to provide funds that would otherwise not be there to support conservation), or even that the Earth is spherical (this one is true, or more accurately the Earth is an approximate oblate spheroid). It's also why many approaches to fake news and misinformation fail. It's not about the information; it's about whom we trust. So how do we acquire these beliefs?
Remember, any adaptive evolutionary system – artificial, natural, cultural, or even a genetic algorithm in machine learning – requires just three ingredients:
All three are present in the cultural evolutionary system of human social learning.
Diversity is a fact of life – people do all kinds of things for all kinds of reasons. Some parents are tiger moms and dads, others prioritize creativity and free expression. Some people like the sciences, others like the humanities. Some people are extraverts, risk takers, with short-term mating on their mind; others are the opposite. The key to cultural evolution lies in transmission and selection.
Transmission needs to be faithful – not too much information should be lost. In genetic evolution, this is achieved by low mutation rates when genes are transmitted from parent to child. In cultural evolution, this is achieved by imitation rather than emulation when we learn from others. Remember Horner and Whiten's experiment with young Ngamban chimps and young Scottish children? Both animals were socially learning, but the chimps loosely emulated and the children faithfully imitated. The children copied without understanding. But no one copies at random.
We copy more when we're uncertain or when the cost of figuring it out by ourselves is high. We copy people who are successful, who are experts, who others are copying. We copy people who are similar to us. We copy less as we get older. And we copy people who meet these requirements, but especially if they seem honest and sincere. It's not a perfect process, but over generations it filters the best behaviors and beliefs into the population until the population itself becomes a source of learning. At this point, you can start following trends and conforming to the majority.
We even use these strategies to learn how to learn. You weren't born with an understanding that a suit signals success, unless you're in Silicon Valley, and then its more a signal that you're in sales. You learned that. Nor that a nice typeface meant New York Times-level credibility. That's a learned strategy and the reason that your grandparents often confuse weird websites with newspaper-like web templates as legitimate sources of information. When you decide who to invest with on eToro or who to follow on Instagram, you're probably looking at popularity and who is already following the person – applying conformity to infer success – a so-called prestige strategy.
The law of evolution found a new way to increase efficiency and cooperation and harness energy – by giving us rules for who to learn from that reinforced beliefs, behaviors, and technologies that worked and discarded those that did not.
We can think about these strategies as homing in on who's got knowledge worth paying attention to, limited by how much information we have about them. If we've seen someone succeed by becoming a football player, we seek out those with the best football skills to learn from – a skill or expert strategy.
But sometimes, we don't know which specific skills lead to success and so, instead, we use cues of success overall such as the appearance of wealth – a success strategy. We identify those who are successful and copy a bit more generally, not knowing what exact traits led to their success.
Further, we may not even know who is successful or what constitutes success, and so we look to see who others are paying attention to and, not knowing which traits matter, over-copy these individuals – a prestige strategy.
The prestige strategy helps to explain the famous-for-being-famous phenomenon – a self-created false signal of success that becomes real. It's a bootstrapping of the prestige strategy. We copy someone because other people are copying them (or appear to be copying them).
Even our prestige strategy leads us to be more nuanced than just the number of connections a person has. We care about who those connections are. If a person on Twitter is followed by Barack Obama and Elon Musk, that counts for more than a greater number of less-known followers. We can quantify prestige using what's called eigenvector centrality, which is also how Google got so good at ranking Web pages. Before Larry Page and Sergey Brin's algorithm, search engines used the contents of pages to help you find them. In the world before Google, when you searched for ‘fish’ you'd get pages that mentioned fish. In contrast, PageRank effectively calculated the prestige of a website by looking at how many and which other pages that mention fish linked to it. Using the same technique that evolution also discovered led to search results that were so much better. Google completely replaced Ask Jeeves, AltaVista, Dogpile, and all the other ’90s search engines.
Alongside the prestige strategy there may also be a pariah strategy – an over-avoidance of traits possessed by those who others avoid. For example, the name Adolf and the toothbrush moustache were both popular in Germany when Adolf Hitler became Chancellor. But both precipitously fell in popularity following the end of the Second World War because of their association with him. And yet, it is unlikely that being christened Hans Hitler and sporting a goatee would have prevented Hitler's evil atrocities. We have a natural inclination to avoid even associations with failure, unpopularity, or, in this case, evil.
When people in a society deploy these social learning strategies, they filter out things that don't work and make things that do more popular. And so, an even easier way to absorb a package of adaptive traits is to just copy what most people do – conformity. In WEIRD societies we think of conformity as something to avoid. But the reason we warn people not to conform is because it's such a strong human tendency. The negativity around being conformist is a mostly Western idea. Many societies assume those who can learn from and copy others are the smart ones.
We unconsciously conform and apply these strategies to all aspects of our lives. One of my favorite examples is a neuroscience study that shows that we look to others to learn who is attractive.
When our friends find someone attractive and say so, the target of their hot-or-not judgement goes up in our own estimation. We might predict that the effect is stronger if we find our friends attractive. Thus, even though attractiveness in the context of mating is a product of well-conserved cues for health and fertility, such as symmetry and clear skin, it is also subject to cultural evolutionary forces. And this explains changing standards of beauty.
Once upon a time, having more fat revealed that you had enough calories and were therefore high status and wealthy. Today it indicates the opposite; that you don't have access to high-quality foods or time to exercise. Once upon a time, being pale meant you were probably wealthy and were able to spend time indoors and being tanned meant you worked in the fields. Today being pale indicates you're stuck in an office and have less time for vacations and basking in the sunshine.
These social learning strategies are incredibly clever, as we shall see. While other animals also socially learn, humans have the most sophisticated social learning repertoire in the animal kingdom. We are the most strategic social learners.
We are not blind copiers. For example, when we deploy the above strategies, we consider their relevance to ourselves. Michael Phelps, the Baltimore Bullet, credits his outstanding swimming success to copying his GOAT predecessor Ian Thorpe, the Thorpedo. Boys will tend to focus more on older males, girls on older females. And people will often pay more attention to those of the same ethnic, political, or social group, or who have things in common with them. This tendency to copy those like ourselves is why representation and role models are so important.
We also consider the sincerity or accuracy of beliefs. We hate hypocrites and pay attention to signs of insincerity. For example, if you met someone who said they took a particular supplement every day that made them super-productive and offered you some, you should be suspicious. First, you would evaluate whether they really are successful and second whether they actually took the supplement. Consuming supplements can be costly or even dangerous and you want to be sure that at a minimum you're copying an actually successful person who actually believes that the supplements are helping them. Seeing them take the supplement every day is a much stronger signal for copying. The founder of Nike, Phil Knight, intuitively understood this. Rather than listing all the features of his shoes and telling you how amazing they were, he made the brilliant business decision to sponsor sports stars. When people see sportspeople actually wearing Nike while competing and ideally winning, it's far more persuasive than any verbal endorsement.
None of this is to say that every weird belief of a successful celebrity is adaptive, but it is more likely to get copied. Cultural evolution is the process by which these social learning strategies over time weed out things that don't work on average and keep things that do.
Costly and sincere cues are more important when the relationship between a behavior and successful outcomes is less obvious. The transmission of religion offers a compelling example, because religious beliefs are often not directly observed and the causality is often not immediately obvious. Therefore, the transmission of religion relies heavily on costly displays, sincerity cues, and prestige.
You are more likely to become a believing Muslim if your parents prayed five times a day even when they weren't observed. If they only went to Friday prayer where they could be observed by all their friends, you might infer that their piety is a product of community pressure rather than sincere belief. This might be why many religions call for private prayer as a cultural adaptation to help transmission of the faith from parents to children, who can observe private prayer. Islam's prayer style, which involves touching one's forehead to the ground, can also create the zebibah prayer bump, a callus or dark spot on the forehead that serves as a hard-to-fake signal that the person engages in frequent prayer. If your parents were successful or you see successful Muslims, this adds to the likelihood of internalizing the faith.
Religion often has extreme forms of costly behaviors that signal true belief and increase the belief of others. Tertullian, one of the early fathers of the Catholic Church in the second century, was right when he said that ‘the blood of martyrs is the seed of the Church’. What could demonstrate true belief in an afterlife more than giving up your current life with no obvious personal or group gain?
People also care more about some kinds of information than they do about other kinds – content strategies. For example, children will pay more attention to information about scary (and therefore dangerous) animals and avoid potentially poisonous plants. And people care a lot about gossip – information about people's love lives and information about their reputations. And because people tend to try to present their best selves, we pay more attention to negative information.
Let's say I have two pieces of information about a popular celebrity. The first is something amazing she did. The second is something terrible she did. You can only choose one. Which do you choose?
Some aspects of culture are obviously and objectively true, with clear causality. For example, steel axes are better at cutting than stone axes and guns are more effective weapons than swords. And so, steel axes replaced stone axes and guns replaced swords, diffusing beyond group boundaries. In contrast, other aspects of culture, such as marriage practices, are less likely to diffuse beyond group boundaries, because each culture's wedding or marriage traditions are usually not objectively better than those of other cultures.
A key cultural mutation in the spread of the subjective and causally obscure is a claim that it is objective. One key move that may have led to the spread of Christianity, for example, was the claim that its beliefs were objective.
Many ancient religions claimed that their god was better than your god and left it to the battlefield or prosperity to decide which was true. But Judaism was among the first religions to claim that there was only one god. Not that the Jewish God was better than your god, but that your god didn't even exist. The claim was now objective, making it easier for an evangelical offshoot of Judaism keen on converting the world – Christianity – to spread beyond group boundaries.
These are just a handful of the many social learning strategies that have been identified. Once you understand these strategies, you'll begin to see them everywhere. People trying to sell you things, special interest propaganda, psyops, or simply the nature of peer pressure and bullying. But while you can begin to recognize these strategies or even see how they're embodied in, say, the latest popular business books, we use them naturally and automatically, internalizing what we learn.
We deploy these social learning strategies effortlessly and often without conscious awareness. Take grammar for example. You might recall the grammar lessons you suffered through in school – present and past participles, gerunds, and so on – but much of grammar, the rules of language, are implicitly socially learned.
For example, there's an order to describing adjectives in the English language that some non-native speakers are taught explicitly, but that if you're a native English speaker you might not even know that you know unless you read Mark Forsyth, who popularized this fact. It goes like this: opinion-size-age-shape-color-origin-material-purpose noun. So you can have a lovely little old rectangular green French silver whittling knife. But if you're a native English speaker, try moving adjectives around and it just sounds off: a green French old little lovely silver rectangular whittling knife – ugh!
Similarly, we acquire how to say words – our accent – from those around us, with a focus on the prestigious or majority. Immigrant children, for example, will often acquire the accent of the majority rather than their parents. Accents are thus a hard-to-fake and generally reliable signal that reveal the sources of much cultural influence. Indeed research reveals that accent trumps race when it comes to judgements. That is, people will individualize and ‘forget’ race if they can distinguish someone by better cues of culture, such as accent, clothing, and mannerisms. We did not evolve to pay attention to skin color. For most of our history, the groups around us had the same skin color and generally looked physically similar. But they were distinguishable through speech and behavior.
Accent reveals a lot and can be a source of discrimination. British university students face discrimination for having an accent from regions of England – or even regions of London – associated with working-class or lower education levels. A similar effect is found in the United States for certain southern accents, particularly the Appalachian accent, often derogatively labelled ‘hillbilly’.
Accent is a shibboleth, demarcating communities and indicating the degree of integration and cultural transmission between communities. When groups in the same society speak with different accents, it is an indication of a group divided, a fractured society.
Implicit, internalized social learning writes the software of your mind, but its implicit nature also means that you are cursed by your own knowledge about what you think is and isn't true, what is right and wrong. You have no time or ability to recheck everything you've learned and no need if your life is going well. In fact, you're doing it right now, implicitly (or perhaps explicitly!) using these very same social learning strategies to decide if what I'm saying is sensible and useful. Testing the degree to which my claims fit your intuitions based on what you think you know and whether what you're learning will serve you well, what my politics might be, whether I belong to your ingroup, whether I'm sincere, and so on. Don't worry, I've provided a peppering of examples for you to test these claims for plausibility against your own life experience and what you think you already know.
Different people in different parts of the world have different culturally evolved and socially learned beliefs and behaviors. They have different software shaping their brains’ hardware – literally changing the size and function of different parts of their brains. That software constitutes different culturally adaptive packages, filtered by generations of sophisticated selective social learning that has enabled our species to live in every ecosystem on earth. We feel like we understand why cassava should be soaked and boiled, why certain foods should be avoided during pregnancy, why bending a bow in a certain way makes a better weapon, why we should brush our teeth twice a day, eat more vegetables, stretch before exercising, or wash our hands before eating. But typically, we do not. Washing hands, for example, seems obvious to those of us who grew up in a world where everyone takes germ theory for granted, but it was less than 150 years ago that doctors refused to believe that fewer mothers might die if they washed their hands between examining a dead body and delivering a baby.
We acquire all these practices and more through social learning strategies that write the software of our brains and our societies. Our ingenuity is a product of this software, which has forced our brains’ hardware to keep up, co-evolving with our genes to predispose us toward certain kinds of information, such as language, norms, and gossip. In essence, what we had to learn got so large, we maxed out on brain size and began focusing on only the most useful and relevant information, taking the rest for granted.
Our cultural package, much like energy, is part of the water we swim in – essential to our lives, but mostly imperceptible. The inability to see our cultural package, our incredible, information-dense inheritance, different for different people, leads us to judge one another as intrinsically more and less sophisticated. It leads us to assume a human nature that is instead based on culturally transmitted skills and fail to appreciate the many ways in which mental software is written by those around us and those long gone. But seeing that cultural package and where it comes from is essential to understanding who we are and how we got here.
When I think of myself as a recipient of a cultural package that generations before me have added to and subtracted from, grown and made more efficient, and of the origins of which I have little to no knowledge, I feel a mix of humility and awe. Maybe even a little unease and fear. But understanding how we acquire this package and then contribute back to it is critical to understanding how the laws of evolution, innovation, cooperation, and energy manifest in our species.
This understanding is necessary to figuring out how to maximize our chances of reaching the next level of energy abundance. Part of this requires understanding how these strategies have made us clever, because people often assume that the secret to our success is our intelligence.