The surprising evolutionary link between bowerbirds and human art

Sexual selection may explain the start of art. As [Charles] Darwin put it in The Descent of Man: “The playing passages of bower-birds are tastefully ornamented with gaily-coloured objects; and this shews that they must receive some kind of pleasure from the sight of such things.” The satin and great bowerbirds courting females with bottle tops and chili peppers are of course driven mostly by instinct, says the conventional wisdom, while human beings have culture. 

Hmm. I think both ends of that claim are partly wrong: People are driven more by instinct than we admit, and bowerbirds have more culture than we assume. Bowerbirds are unusually large-brained birds. The Cambridge zoologist John Madden surveyed the behaviors of bowerbirds and concluded that “despite a paucity of data in comparison with primate studies, it could be argued that bowerbirds may be considered to fulfill the same criteria on which we base our use of the term culture when applied to our close relatives, the great apes.” For example, a spotted bowerbird that found itself swept off course by a storm, ending up in satin bowerbird country, learned to collect blue items instead of its usual white, green, and red.

As for whether people have instincts, plenty of experiments show that people have innate tendencies and that the way these work is often through making people more likely to learn some things than others, so culture and instinct are not opposites, but work together. Nature operates via nurture, not versus it. As the anthropologist Joe Henrich has documented, this means, for instance, that when people make mistakes they tend to be in an adaptive direction such as mistaking safe animals for dangerous ones rather than vice versa. Learning is a means of evolving. 

This means it is highly likely that when smaller-brained human ancestors first began decorating their bodies, their homes, their clothes, and their tools, it was likely very much more at the behest of instinct rather than the result of any kind of rational calculation. Bowerbirds might help us understand how that happened.

Take, for example, the Acheulean handaxe. This is the name given to a widespread tool used mainly by Homo erectus, which turns up again and again in the archaeological record, over an immensely wide area from Europe to southern Africa to Asia and over an unbelievably long period from almost 1.8 million years ago to just 300,000. It’s tear-drop-shaped, symmetrical, and sharp-edged. Years ago I spotted that the one that sits on my desk – my wife found it on eBay, but it probably came from North Africa originally — was exactly the same size and shape as the computer mouse I used at that time. This sent a shiver down my spine: two objects designed to fit the human hand but separated by at least half a million years. 

Nobody knows quite what Acheulean handaxes were mainly used for but skinning animals, slicing vegetables, and whittling wood are probably on the list, though throwing them as sharp-edged projectiles has also been suggested. It is not immediately obvious to us today why this design was so well suited to some particular task, let alone why it worked equally well in places as far apart as Africa, Europe, and India.

The far greater problem the handaxe poses is why it never changed much either in space or time. It experienced more than one and a half million years of extreme technological stagnation. Every other technology hominins invented, from the spear thrower to the Swiss army knife has evolved pretty rapidly thanks to innovation. Not the Acheulean handaxe. That it could be so universal and change so little over such a long period — hundreds of thousands of generations — defies cultural explanation. 

If it was a product of culture, it would surely show fairly rapid change and significant geographical variation: That’s what culture does. Yet archaeologists have consistently and universally assumed that the Acheulean technology was indeed purely the product of culture: within-tribe imitation according to a social tradition. They have assumed that because, well, they always assume that about human beings. The possibility that we are looking at a product of instinct is just never even considered. In the words of four scientists arguing for a different explanation: “When nonhuman animals display complex behavior, the default assumption is that it is under genetic control. For complex behavior in humans and other hominins, however, the default position is to invoke culture and not genes.”

It was the anthropologist Rob Foley who first made the suggestion in the 1980s that Acheulean handaxes might be partly the product of genetically inherited habits rather than purely cultural learning. He was largely ignored. Two evolutionary anthropologists, Pete Richerson and Rob Boyd, took up the idea in 2005, describing it as “bewildering” that so little variation would appear in a cultural tradition. Then four anthropologists based in Canada and the Netherlands, led by Raymond Corbey of Leiden University, revisited the debate in 2016, adding several new strands of argument. 

They pointed out that scientists struggle to explain the sudden arrival of fast-changing and cumulative stone technology in the upper Paleolithic after about 300,000 years ago, usually reaching for a “cognitive upgrade or increased population size” as the trigger, neither of which is persuasive. Instead, Corbey argues, if the previous technology was largely instinctive then the arrival of cumulative cultural innovation is more easily explained: It was a switch from expressing an instinct for learning something to a pattern that was very much more dominated by social learning. Corbey points out that evolution is fond of endowing even highly intelligent animals such as bowerbirds with genetically determined behaviors if only because it is a shortcut for the individual to get the behavior down pat. When I contacted Corbey to explore this further, he reminded me that “the relevant database is a gorgeous challenge. It comprises a few million known handaxes and billions of unfound ones, on a conservative estimate.”

Of course, most cultural anthropologists reacted with disdain to this genetic theory, rabbiting on instead about “preferred cultural conservatism” and “the psychological bias for majority imitation that subsequently became a social norm,” which is little more than a restatement of the problem. Corbey responded to his critics by pointing to the “Baldwin effect,” an idea proposed by the American psychologist James Mark Baldwin in 1896. 

Baldwin said that animals which adopted new habits or found themselves in new environments would then find that any genetic mutations which helped them cope with those circumstances would be favored. As those mutations spread, the new habits would become more heritable and instinctive, therefore. So what starts as a fully learned behavior could end up as a partly inherited one. 

This might superficially resemble the inheritance of acquired characteristics, but it would not be. In Baldwin’s words: “It would look as though the acquired traits had sunk into the hereditary substance in a Lamarckian fashion, but the process would really be neo-Darwinian.” 

Jean-Baptiste Lamarck famously proposed a theory of evolution in which offspring inherited traits that were acquired during their [parent’s] lifetime: a body-builder would have muscular children, for example. But the Baldwin effect is subtly different. For example, by taking up the consumption of milk, human beings in a few parts of the world put themselves unwittingly in a good position to promote mutations that allowed the digestion of lactose in adulthood, and sure enough, the genomes of people changed. Culture can therefore drive genetic change. I cannot emphasize enough just how much both most social scientists and most evolutionary biologists struggle to see this point.

In the case of handaxes, the significance of this insight is that if early hominins started knapping stone tools, any mutant human being with a genetic propensity to be good at this skill from the start would thrive, and leave more offspring, so the skill would itself gradually become more heritable. As Corbey puts it: “If phenotypically plastic individuals grow up time and again, over hundreds if not thousands of generations, in a technological niche while manipulating stone, and provided that the cost/benefit ratio is right: Would not selection in the long run favor features of the organism befitting their technological capacities, so crucial for survival?”

This goes so against the grain for cultural anthropologists — and most of the rest of us — who think of human beings evolving away from instincts and towards culture […]. But evolutionary biologists who have looked into the mechanics are agreed that the Baldwin effect is a real phenomenon. So even if Homo erectus’s ancestors did not have a genetic tendency to enjoy making stone tools, their descendants probably did. Even if the first bowerbirds were simply copying successful rivals, their descendants found themselves with highly specific instincts to (learn to) build highly specific bowers. It is no more fanciful to suppose that Homo erectus had an instinct to make handaxes than that a satin bowerbird has an innate instinct to make a highly complex bower to a particular design. 

At the very least the hypothesis should be taken seriously. This does not mean handaxe-making requires no learning; bowerbirds still have to learn how to perfect the building of bowers by watching others and practicing. But it is directed learning, towards an instinctive, adaptive end. I repeat: A lot of learning is like that — nurture reinforcing nature, rather than overcoming it.