CHAPTER FOUR

How to Think about Evolution Non-reductively

Since the middle of the twentieth century, the rigorous mathematical formalism of population genetics has fostered a reductive view of the evolutionary processes. We could reduce species to a pool of their genes, a person to their genotype, and evolution to a change in allele frequencies over time. The reductive view from population genetics reached its climax in Adaptation and Natural Selection by George C. Williams (1966) and The Selfish Gene by Richard Dawkins (1976).

There were, of course, minority voices to remind us that species are not just fields of genes; that a living four-dimensional organism is often not predictable from a genotype; and that evolution encompasses a lot more than just changes in allele frequencies. Cell biologists, for example, had long grappled with the fact that life is hierarchically organized. Even though a human being is composed of just cells and their products, a human body is organized cells, and understanding the nature of that organization is critical to understanding the body. By the same token a species is not simply a cluster of animals, but a special kind of cluster of animals—those that see themselves somehow as potential mates or competitors for mates. Once again, the nature of the organization of the units is what creates the higher-order structure.

The evolutionary biologist Ernst Mayr had challenged the reductive paradigm, calling it “beanbag genetics” in a famous 1959 paper.1 Likewise the population geneticist Richard Lewontin advocated for different “units of selection” in a famous 1970 paper. By the 1980s, a mainstream reaction against the reductive view of evolution was under way, spearheaded by the paleontologist Stephen Jay Gould.²

In challenging the reductive view of evolution, the midcentury ideas of the brilliant developmental geneticist Conrad Waddington were rediscovered (he died in 1975). Waddington had been an unapologetic holist in an age of reductionism, and conceptualized evolution within a hierarchical and cybernetic framework. His invocation of biological levels of organization and interactions among them was more complex than the standard reductive model, and perhaps left his contemporaries a bit intimidated. Nevertheless, it is now clear that Waddington’s systemic idea of evolution is a lot closer to reality than the alternative, and provides a valuable framework for thinking about the evolutionary processes that have produced the familiar modern human condition.³

Waddington envisioned a hierarchy of process, all ultimately accruing to the production of genetic differences between ancestors and descendants.⁴ Waddington, however, centralized organisms rather than alleles. His model began with organisms, but not as static animals—rather, with the “point that the organisms undergoing the process of evolution are themselves processes.”⁵ Waddington deliberately introduces agency into animals, by noting that they choose where to live, and in so doing, they modify their habitat simply by being in it. He called this relationship between the organism and the environment “The Exploitive System.” As it grows and matures, the organism is faced with certain stresses that test its ability to adapt and survive. He called these developmental potentials “The Epigenetic System” and the ability to survive and breed by developing a particular way in that particular setting, “The Natural Selective System.” Finally (and significantly, for its trailing place in the processual evolutionary hierarchy), Waddington called attention to the modification of those “selected potentialities” via mutation in “The Genetic System.”

The salient features of Waddington’s view of evolution are the recognitions that (1) the conceptual units in evolution have overlapping hierarchical relations; (2) organisms do not reside in niches, but partly make them; (3) organisms are not always adults, but grow and develop in response to the particular circumstances of their lives; and (4) organisms vary physiologically in their ability to make those developmental responses, which in turn is an important component of their relative survival and reproduction.

What follows is an expansion and modification of Waddington’s systemic theory of evolution. We will think of the processes of human evolution here in terms of five nested systems: the genetic, the developmental, the exploitive, the cultural, and the natural selective (fig. 1). These systems are not bounded or discrete; they interact with one another and bleed into one another. Seeing evolution in this way, however, helps to make the point that the classical reductive model is really only the starting point of an understanding of evolution.

Figure 1. Hierarchical evolutionary systems.

THE GENETIC SYSTEM

This system is the cellular basis of evolution, the creation of new genetic variation. With the discovery of the structure of DNA in 1953, and subsequently of genome structure, we now know the most fundamental ways in which heritable changes are produced. They are produced by changes in genes, which are units of function within the genome. The genome is made of DNA, and the genes are islands, embedded in oceans of DNA with either no function or very limited and cryptic function.

Only about 2 percent of the genome is actually functional in the classic genic sense of “coding for proteins” through an RNA intermediary. Somewhat more is obscurely functional, being transcribed into RNA, but not actually expressed in any obvious way physically. But most of the genome lies between genes; or lies within genes, and is deleted from the RNA transcript before protein translation. Consequently it has traditionally been regarded as being of very limited value or utility, although the possibility exists that the cell has uses for some of it that we do not yet see or understand.6

Two lines of evolutionary evidence converged to produce this understanding of the genome. By the 1960s, human diabetes was found to be treatable by injections of insulin derived from a cow or pig pancreas, despite the fact that there are some structural differences among the hormone molecules. Far from being precisely attuned to cow physiology, the bovine insulin molecule works well in humans, which in turn seems to imply a great deal of slop in the genetic system. Discoveries such as these suggested empirically that the genetic system ought to be best understood without the assumption that it has been precisely engineered by natural selection, that is to say, as “non-Darwinian evolution.”⁷

Further, the genomes can be scrambled without apparently compromising the production of organisms, as we noted in chapter 3 with the example of the “siabon.” On the basis of early genomic data like these, the molecular biologist François Jacob famously argued that genetic evolution acted not like an engineer, but like a tinkerer, drawing inspiration from the anthropologist Claude Lévi-Strauss’s work on mythology.⁸ The storyteller, said Lévi-Strauss, does not compose an optimal story from scratch, but rather, cobbles together available motifs and suitable themes, tries them out, sees what parts work well together, and consequently assembles a story that will be familiar and resonant for the audience, without necessarily being efficient, brief, or perfectly suited. In a similar fashion, argued Jacob, nature works with genetic systems that are functional, redundant, and suboptimal, and transforms them into other novel systems with those properties. Thus, at least from the standpoint of genetics, we should see evolution metaphorically not as an engineer, but as a bricoleur, or tinkerer. Both the gibbon and siamang genomes work; they just do so with radically redeployed genes.9

So DNA is not like a blueprint, despite the hype for the Human Genome Project in the 1990s, in a critical way. Most DNA is irrelevant to the production of an organism; you can scramble it up, or even delete chunks of it, often with no apparent ill effects. DNA is not fine-tuned or precisely engineered, or well adapted; it simply gets the job of building an organism done—and there are a lot of genetic ways of getting to the same end point.

Mutations are changes to the DNA in a cell, and usually they don’t matter at all, and simply accumulate in species over time. That is because of the limited functionality of most of the DNA; change it, and it doesn’t make you better or worse. The changes that do occur in functional genes are more likely to make you worse than to improve you. The reason is simply that over the course of the eons of the history of life, our genomes have evolved to produce bodies that function. Random changes are not likely to improve them, any more than random changes to functioning machines are likely to improve them. That is why mutations give you cancer, not X-ray vision. Nevertheless, what differentiates organic “machinery” from the engineered products of human labor is the degree of slop that we find in nature, as opposed to the efficient human engineering of machinery. Randomly change the blueprint and the machine will simply not run as it was designed to, even if it is not supposed to run at maximum efficiency.

Consequently, when we compare DNA across species, we almost always find more differences in non-coding, intergenic DNA than in coding, genic DNA. And within genes, we find more differences across species in DNA regions or sites that do not change the protein products than in places that do change the protein products. Nevertheless this kind of comparison measures only one kind of mutation—nucleotide substitutions, the change of an A, G, C, or T in DNA for one of the other letters. Since the 1980s, it has become clear that there are many more ways to change the DNA—for example, by inserting or deleting small tandemly repetitive DNA sequences, or larger movable DNA sequences, or by using the DNA in one gene as a template to alter the sequence of a gene beside it.

The ultimate result is simply that new DNA sequences are produced, which may have some effect upon the physiology or anatomy of the organism, the things that actually interact with the environment, which of course the genetic system does not.

THE DEVELOPMENTAL SYSTEM

By the late 1930s, Waddington was distinguishing between genetic differences, which exist in the DNA sequences from person to person, and epigenetic differences, which differ from cell to cell within the same person, in spite of genetic uniformity. Both patterns of difference are stably inherited: aside from rare somatic mutations, a body develops mitotically from a fertilized egg, while retaining the same genotype over the course of the life span; and muscle cells give rise to other muscle cells, not to nerve cells. The nature of the epigenetic system proved more elusive than the genetic system, however, and the rise of molecular biology in the 1960s and the Human Genome Project in the 1980s left the question of epigenetics behind. But at some point there is a fundamental difference between a human being and a 170-pound flask of human cells. The nature of that difference, and its role in evolution, is the study of epigenetics, or as Waddington called it, “the causal analysis of development.”10

How do cells make bodies? By turning certain arrays of genes on and off, and making sure they stay that way in descendant cells. The biochemistry of epigenetics lies in the regulation of genes—specializing the cells and organizing them, and in the ways in which that information is transmitted to daughter cells after cell division. The crucial aspect of epigenetics is that two cells with identical DNA sequences can be programmed to look and act differently, and that programming can persist across cell generations and across organismal generations. Moreover, since the conditions of life can affect the epigenetic programming of cells, it now seems possible that those conditions of life (i.e., the environment) can have an effect upon the cellular development of the body, and that this in turn can produce a fit between the organism and environment that is independent of its DNA sequence, and that can be stably inherited as if it were genetic. Thus the body can be seen as reactive and dynamic, rather than as passive and static.

Epigenetics emphasizes two features of the body that the DNA sequence misses. The first is adaptability, the property of a body to adjust developmentally to environmental insults.11 We noted some examples of this feature in chapter 3: hypoxia, tanning, and callousing, for example. The second is canalization, the property of a body to find a “normal” way to develop, in spite of environmental or genetic variation. In a famous 1956 experiment, Waddington subjected fruit flies to a chemical interruption of their development, and found that most of them died, but a few of them survived, while developing a weird condition: a second thorax. This wasn’t a new mutation (which would only have originated in a single fly, and there is a mutation that mimics this condition), but a different pathway of development, stimulated by the presence of ether in the fly egg’s atmospheric environment. Waddington artificially selected for those flies that were able to make this developmental adjustment, and soon had a strain of flies that could consistently develop the bithorax phenotype under the environmental stimulation of the ether. Waddington had successfully selected for the physiological property of adaptability; he had a line of flies whose physiology had allowed them to survive by developing very weirdly when appropriately stimulated—rather than simply dying. Then he began to breed and select those flies under conditions of lower ether concentrations, and soon he had a strain of flies that developed the bithorax phenotype without ether at all. He had selected for canalization, so that the flies had found a “new normal.”12

This appeared to mimic the pattern of Lamarckian inheritance, or the inheritance of acquired characteristics, but Waddington explained the pattern by a strictly Mendelian process. The genes involved were not genes for phenotypes, as the reductive population geneticists saw things, but rather, genes for the ability to physiologically adjust. In the first phase of the experiment, he was selecting for genes (which have still not been isolated) that allowed the fly to develop weirdly, rather than simply to die in the toxic conditions (adaptability). And in the second phase, he was selecting for genes that permitted the weird phenotype to become the normal one (canalization).

How might these ideas be applicable to humans? Consider our most fundamental feature, bipedality. Under the reductive model, where genes code directly for phenotypes, we have often imagined bipedalism emerging gradually from the successive fixation of uprightness mutations, as a quadrupedal ancestor at a 45-degree angle had mutations that allowed its descendants to walk at a 60-degree angle, who then had mutations that allowed their descendants to rise to 70 degrees, and eventually to perfect 90-degree verticality. And all this presumably was accompanied by the gradual accumulation of mutations that altered the pelvis, knees, legs, spine, and cranial base in parallel.

The problem is that all of those intermediate states never existed. They are certainly not evident in the fossil record. Apes usually walk quadrupedally, but sometimes do walk bipedally. They do it over short distances and clumsily, but discretely, and both modes are part of their locomotor repertoire. They can sometimes walk on two feet, although not for long, when they want to. It stands to reason that our own ancestors could do it also. Consequently, we must think of the evolution of bipedalism not as the acquisition of a brand-new feature, but as a transition from a facultative to an obligate manner of walking. That is, an ancestor that could walk bipedally, essentially chose to do more of it, and now has descendants that can do nothing but.

But now, instead of an empirical problem, we have a theoretical problem, for that is just not supposed to happen. Choices that you make in your life can’t get into your DNA and be passed on to your descendants. You can choose to root for the Red Sox, but your children might root for the Yankees. You inherit your DNA, but you don’t modify it. It’s like bodily mutilations. If you cut the tails off of fifty generations of mice, the fifty-first generation has tails as long as the first. Why? Because you changed their tails, not their DNA. How might the choice to walk upright more frequently have occurred genetically? Waddington’s ideas are useful here.

An ancestor that began to walk upright more frequently would have considerably different stresses placed upon its skeleton. Its center of gravity would lie atop its pelvis, rather than ahead of the pelvis; its feet would be supporting its full body weight, rather than just its rear weight, for example. These stresses would result in developmental modifications of the body, such as the curvature of the lumbar region of the spine—like the bithorax fruit flies, but less bizarre. Some early humans would be better able to make these skeletal adjustments than others—this would be natural selection for adaptability. Subsequently, natural selection for “the new normal”—canalization—would facilitate the developmental appearance of these features. There might well be tweaking to be done, in the fixation of mutations affecting body proportions, for example; and further, bipedalism is crucially a learned behavior in humans (see below)—but to model it effectively as a genetic process, we need to think of it as a developmental system, rather than as a static set of mutations-for-traits.13 Indeed, a parallel argument can be made for the locomotor transition from fish to tetrapod.¹⁴

The relationship between the genetic system and the epigenetic or developmental system is also highly political. In chapter 1, we observed that the punch line of the very first textbook of Mendelian genetics was that “the creature is not made, but born.” Whether true or not, it certainly has considerable bio-political content. You inherit your genes from your ancestors, so are you any more than their genes? Those at the top of a hereditary aristocracy would certainly like to think not. If you are simply a reconstitution of your ancestors, then the possession of a noble pedigree is all you need to establish your superiority to the rest of the world. This mode of thought has always been there in science: by the end of the nineteenth century, biology was very polarized between two bio-political views. Followers of the biologist August Weismann called themselves “neo-Darwinians” and held that the germ cells comprise a link between the generations, and the body (Greek, soma) is simply a cellular dead end. Thus, through the “continuity of the germ-plasm” you are simply a reconstituted product of your ancestors, an argument that of course resonated strongly with political conservatives in fin-de-siècle Europe. But if you are not simply a reconstitution of your ancestors, then what else are you made of? Scientists with left-leaning political views found other things to study that shaped human existence: notably, culture, parenting, and the direct influence of the biological environment. Genes (“nature”) and the conditions of life (“nurture”)—in the euphonious opposition from Shakespeare’s The Tempest—stood as opposing one another.15

These “neo-Lamarckian” geneticists, on the other hand, maintained that you are crucially a product of your upbringing and circumstances. One of the most prominent, and last, of this school was a biologist named Paul Kammerer, who came to America on a lecture tour in 1923, hoping to teach the human race “to avoid acquired degenerate tendencies.” His research, argued Kammerer, would permit us “to eliminate race hatred.”¹⁶ A noble thought, to be sure, but hardly derivable from the mating habits of toads, which is what he studied. Kammerer committed suicide a few months after the revelation that his prize toad specimen had been injected with India ink to emphasize the features that were supposed to have been produced by the inheritance of acquired characteristics.¹⁷

The hereditarian scientific ideology reached its zenith twice in the twentieth century—first, with the rhetoric of the eugenics movement in the 1920s, and second, with the rhetoric of the Human Genome Project in the early 1990s. In the 1920s, genetics provided a rationalization to sterilize the poor and restrict the immigration of Italians and Jews into the United States. A popular 1925 college genetics textbook warned students of “a great many people who are always on the border line of self-supporting existence and whose contribution to society is so small that the elimination of their stock would be beneficial.”18 Not by coincidence, in the 1990s, political conservatives jumped on the “geno-hype” being generated by the molecular biologists promoting the Human Genome Project, and the result was the infamous best seller The Bell Curve, which reiterated old arguments about imaginary racial differences in intelligence being at the root of social inequality.

Epigenetics can thus be seen as the modern scientific reaction against the hereditarian thought that rode in on the rhetoric of the Human Genome Project, which was busily justifying itself with claims like “we now know, in large measure, our fate is in our genes.”¹⁹ It provides an explanation in cellular Mendelian genetics for the influence of the environment upon the body, and as well for the manner in which we are actually more than our own DNA sequences, and more than our ancestors’ DNA sequences.

THE EXPLOITIVE SYSTEM

The third evolutionary system once again highlights the non-passivity of the organism. It is the relationship between organisms and their surroundings. Animals live where they are familiar and safe. Classically, ecologists recognized the fit between an organism and its environment, and saw the environmental niche as a static “given” to which the organism’s ancestors had gradually become adapted. Subsequent generations of scholars, however, came to appreciate that the environment is itself dynamic and reactive, because the organism doesn’t simply “occupy” a niche, but interacts with its environment and transforms it.²⁰ The organism is not an automatic outgrowth of its genotype, but a reactive agent; and the environment is neither stable nor independent of the organism. The descendants thus have to coadapt in harmony with the new environments created by their own ancestors. The most important such transformation was probably engendered by the cyanobacteria hundreds of millions of years ago, photosynthesizing on an earth without atmospheric oxygen. By transforming the atmosphere, they made it possible for multicellular animal life to evolve.

In the case of human evolution, these generalizations reach their apotheosis. The ecological focus of human evolution involves the extent to which our ancestors were not embedded as “animals” in a local “environment,” but rather, brought environments with them, created familiar environments in unfamiliar places, and proceeded to transform wherever they were into images of what they wanted it to be.

Even with brains half the size of our own, our ancestors were looking at the world in a wholly new way, asking what they could do with the things around them. Not only did they transform rocks into tools, but the rocks eventually reciprocally transformed their hands into better tool-using appendages. Chimpanzees don’t do much with tools for two reasons: they have small, weak brains and small, weak thumbs. Using their hands to either hang from or to support their weight when on the ground, apes have long fingers and short thumbs. Probably the only test of strength in which you could beat a chimpanzee is in the children’s game of thumb wrestling. In other words, tools coevolved with manual dexterity.21 The net effect was the evolution of a creature who had not only the desire, but the ability, to see the world as composed of ingredients or raw materials to make things out of.

One interesting consequence of banging rocks together, or rubbing things vigorously, is that sometimes they get warm or throw off sparks. If you choose the right materials, and work at it carefully for a few hundred thousand years, you can become very adept at producing fire when you need it. And the most obvious value of fire is that it allows you to take your environment with you wherever you go. It’s warmth and protection from predators, at the very least. It’s a light in the dark. It also permits the transformation of inedible or indigestible foods into edible and more digestible foods.22 Between the ability to control fire and the ability to skin and work animal hides with their sharp stones, our hominid ancestors could construct environments for themselves in places that would have uninhabitable for their own ape ancestors.

Along with tools and fire, animal hides could be used to make a second skin (i.e., clothing), as well as to help construct a shelter from the elements (i.e., early dwellings). We don’t know when this began, but our ancestors were probably doing it by a few hundred thousand years ago.

The fourth mode by which early humans constructed niches involved importing raw materials from far away, so that they had these objects at their disposal where the objects did not occur naturally. This involved networks of exchanges and reciprocal obligations with other human groups in different areas—in a word, trade.²³ Unlike the networks imagined by modern economists, based on modern markets in which every participant tries to maximize gain and get the most for the least, the networks of early humans probably involved cooperation and ritualized exchange, based instead on ethnographic inference.²⁴ Personal gain at someone else’s expense was probably less of a motivation than mutual aid,²⁵ at very least since these partners would probably be standing in some sort of permanent relationship with one another, linked by bonds of kinship and an understanding of their future expectations from one another.

Finally, early people developed relationships with other animal species that once again created environments previously unknown. Long before these species were maintained and selectively bred (only a few thousand years ago), they probably coexisted with people symbiotically, in ways that benefited both parties. We can only imagine what those ways might have been, but the fact that early humans were drilling holes in mammalian teeth, and sometimes leaving their dead with animal parts,26 suggests that they theorized their coexistence with other species. The earliest carvings we know of, from about 40,000 years ago, are half human, half lion—which suggests that early people thought about their relationships with animals in complex and symbolic ways. The fact that early people utilized shells and rocks and plants, but generally depicted only other mammals, likewise attests to the idea that they thought a lot about, and interacted intimately with, other mammalian species long before penning them up and breeding them.

The most significant aspect of this mode of niche construction came about with the domestication of plants and animals, and the decision to begin producing food. This took place in different parts of the Old World, with different kinds of species, between about 12,000 and 4,000 years before the present. This permitted human societies to control the means of their own subsistence and to store and redistribute the surpluses, although it led immediately to nutritional imbalances, and eventually to gross disparities in wealth and status. One could reasonably argue that much of the modern world is a direct social and economic consequence of the choice that those people made, a few thousand years ago, to begin messing around with the gene pools of their familiar animal and plant species. (This is quite different from modern issues surrounding genetically modified foods, however, since Monsanto isn’t exactly “people,” and the question of the scope, goals, and consequences of such modification today are not comparable to those of several millennia ago.)

THE CULTURAL SYSTEM

This evolutionary system (omitted by Waddington) involves learned behaviors, which exist in other species, but which are elaborated and embellished in human evolution by the development of symbolic thought, essentially creating environments and adaptations out of the imagination. To the extent that these imaginings may be realized, people live very different kinds of ecological lives than do our close relatives, the apes. At the very least, culture transforms what are ecological relationships in other species into economic relationships in humans. Anthropologist Clyde Kluckhohn noted (in quaintly sexist terms) that “culture can be regarded as that part of his environment that is the creation of man.”27

What permits this organically lies in the product of our brain, that is to say, our mind. The human mind seems to be uniquely capable of four processes, which shape the way we interact with our world, and essentially create it.²⁸ The first is that we think hierarchically, not in terms of all the world’s elements being equivalently elemental, but in terms of “this is a kind of that.” That’s the basis of classifying, which we do to everything from relatives to colors to plants. Often there are many possible dimensions by which to classify: for example, we could focus on the use of a chair and classify it along with a bed as “furniture,” or on the quadrupedal structure of a chair and classify it along with a deer as “something with four legs,” or on the composition of the chair and classify it along with a tree as “made of wood.” The choice we make depends on the purpose of the classification, and sometimes on simply arbitrary decisions that our ancestors made for us. Is a dolphin a kind of fish on account of where it lives and how it moves, or a kind of mammal on account of its physiology and evolutionary history? We’ll call it a kind of mammal because we will privilege the second set of criteria over the first, but certainly the first set makes a certain degree of sense too.

The second way we think is symbolically, making arbitrary associations between things that have no intrinsic connection to one another. The most basic example of this is pointing, which humans are doing in their first year, but chimps just don’t do. They can be intensively trained to do it a little, rather like they can be trained to walk and to smoke cigarettes while riding bicycles, but pointing is just not a chimpanzee thing. It is simply an imaginary connection between the tip of the index finger and an object out there, but it exists only in the mind of the pointer and of someone with a similarly built brain.29

The third way we think is creatively, taking information from different domains and putting them together in new ways. Probably the most basic way of doing this is by the use of simile: a mountain may be like a molehill, a cloud may be like the silhouette of a face, a lion may be like a brave, strong friend. These juxtapositions or combinations may have been thought of before or they may be brand-new, but this manner of thought opens up essentially an infinitely expandable array of possibilities. Anything can in principle be associated with anything else, if you just think about it the right way.

And finally, we think abstractly. That is to say, we conceive of things that don’t exist or will never exist or have never existed—and we can treat them as if they were just as real as things that do exist or did exist or will exist. Burying the dead, for example, was being carried out by prehistoric peoples 100,000 years ago, despite being a waste of time and protein. The reason has something to do with love or respect or memory—but it involves a conception of the idealized past or the imagined future, not the lifeless present.

Human thought, however, is only the merest aspect of being human—for it is organic and internal. The more significant part for our evolution is what it allows us to do among ourselves, the “superorganic” aspects of human existence, which involve the relations among people formed by our communication system, quite unique in the history of life.

What language does for us is not simply to allow us to have abstract thoughts, but it compels us to share them, and thus opens up a social universe of imagining, planning, and cooperating. This, in turn, permits us to construct our own niches—but not simply in relation to our physical survival and comfort. We create historical and social environments as well, which we were born into and we have to adapt to. Language is the most fundamental of these environments, both a function of our organic, cognitive processes, and yet also a construction of history and culture.

The primary effect of language is that it allows us to know what is going on in someone else’s mind—because, unlike other species, they can tell us. This forms the basis for the coordinated activity that characterizes human behavior. Along with the ability to tell someone what you are thinking comes the ability to reinforce that information with highly developed facial musculature, eyebrows, and eye whites, which readily combine to communicate a range of gross feelings, such as happiness, rage, disgust, surprise, boredom, and sadness, as well as more subtle things, like bliss, irony, and romance. It also comes with the ability to deceive others into misreading your intentions, for your own benefit, which in turn has raised the possibility in the minds of some scientists that we evolved not so much to be cooperators, but rather to be schmucks, for our intelligence is there by virtue of having facilitated the development of deceit in our ancestors.30 The fact, however, that we can do something does not mean that we evolved to do it, a well-known fallacy known as adaptationism. The fact that we can do cartwheels does not mean we evolved to do cartwheels; it only means that the properties that we did evolve also permit that activity. Seeing humans as naturally prosocial or antisocial simply recapitulates an old philosophical argument—for example, Thomas Hobbes in the mid-seventeenth century seeing people primordially as solitary and competitive, and Giambattista Vico in the early eighteenth century seeing people as primordially cooperative and social.³¹ Our evolutionary history involves the propensities to be both cooperative and manipulative, but the cooperative, prosocial features seem to be the ones that got us where we are in the history of life.

What we do not know is whether language (as vocal symbolism) emerged from primordial ape vocalizations that became symbolic, or from primordial symbolic acts that became vocal. Ape vocalizations are not conversational (that is to say, alternating, so that one ape goes “oo-oo-oo” and then gives another ape a turn). They appear, rather, to be contagious, like laughter; that is, one ape goes “oo-oo-oo,” and the others join in. Further, we humans control our breathing so that we vocalize almost exclusively while we exhale. That’s not true for the chimp vocalizations. The inference, then, is that ape vocalizations are homologous to laughter rather than to speech, which in turn suggests that ape vocalizations are not the evolutionary source of human language.32 Instead, it seems more likely that human language is the result of symbolic acts—like pointing, gesturing, and dancing—whose cognitive associations became transferred and eventually co-opted by the vocal apparatus.

One such symbolic act involves bodily decoration, a distinctly human feature. Clothing is not just utilitarian, but communicative; and early humans were probably decorating themselves in other ways—with pigments and jewelry—at least as soon as they began dressing. Indeed, cutting and tending the hair on our head must have coevolved with the technology to do so, and again is far more functional symbolically than biologically. The earliest depictions of the human form, the Venus figurines from about 25,000 years ago, show the hair being carefully tended, back in the Stone Age. This is classic symbolic anthropology: we associate short hair with convicts, soldiers, and businessmen; and long hair with hippies, geniuses, and artists. The connection is subtle but wide ranging, and it seems to be about being close to the nexus of social power, either having it imposed on you, or wielding it yourself. Long hair is symbolically associated with being less controllable. The point is that hair communicates social information about its bearer.³³ It requires constant tending, and it’s uniquely human; apes don’t have to worry about it. But humans have to, because if they don’t, it overgrows their sensory apparatus, which would be patently maladaptive. Head hair had to coevolve with the ability and interest in taking care of it. And what that suggests is that we are dealing with minds that are familiar; they are like our own in some fundamental way, making statements about who we are through our personal grooming habits. This, once again, emphasizes the fact that internal human mental processes are creating external meanings and relationships that connect humans invisibly and symbolically to one another. What is unusual from the standpoint of evolution is that those external, or extrasomatic, connections can outlive the bodies of their bearers, which is in large measure the distinguishing feature of human culture.34

Consider two fundamental human attributes: language and kinship. A human is born into both. You probably didn’t choose to learn English, and you definitely did not choose to be a son or daughter, brother or sister, grandchild, nephew or niece. Those slots existed before you appeared, and you learned how to occupy them; and English existed before you appeared, and you learned how to speak it. Moreover, although you learned English, you didn’t learn all of English. Nobody knows all of English; it’s larger than any single person’s scope of knowledge, and always has been. Likewise, nobody knows all of kinship; in most cultures, the knowledge of how to be a son is different from the knowledge of how to be a daughter, for example; the people who know what to do when a woman is delivering a baby may not know how to trade properly with neighboring peoples or how to make an arrowhead. Consequently, it is not quite right to identify culture as the knowledge that an individual possesses, as biologists and psychologists sometimes do, for no individual in any culture possesses all the knowledge of that culture. Culture, in other words, is bigger than the individuals who possess its knowledge. That doesn’t mean that it can’t be directed or influenced; just that it can’t be possessed, only sampled.

The most directly observable way that humans adapt is technologically, and technological evolution has autocatalytic properties that are quite distinct from the organic properties of the natural world. This arises from the fact that the same technology used for survival and food procurement may be useful in aggression and defense. If you don’t have the most up-to-date and efficient militia or defenses, your neighbors are likely to. And even if they decide not to try and annex you with their technological superiority, they will be just a bit more likely to be able to ward off an attack when your common enemy sweeps down from the north. We now live in an age in which technology is not just a military commodity but forms the basis of our entire economy. And we have arrived at a familiar situation, in which you expect your children’s technological world to be unfamiliar to you. And yet, only few generations ago, most people in the world anticipated that their children’s lives would be pretty much the same as their own.

As we noted in chapter 2, when viewing technology, the long lens of history sees progress and acceleration. Other aspects of culture invariably change, often in reaction to technology, but not necessarily toward objective improvement. Kinship changes (for example, with the large number of single working parents, and the introduction of the kin term “baby daddy”), and language changes (with “twerking” and “selfies”) , but it’s not clear whether those constitute improvement, degradation, or some kind of random motion.

THE NATURAL SELECTIVE SYSTEM

The variations produced by mutation may ultimately be preserved or perpetuated in future generations if they are favorable, or they may be rejected or destroyed if they are injurious, as Darwin recognized. But variations in what? Darwin clearly meant body parts, but subsequent generations of geneticists transferred Darwin’s meaning to genes—by simply equating a species with a gene pool, a body with a genome, and particular attributes of bodies with genes themselves. Thus, the geneticist Theodosius Dobzhansky could reduce evolution to “a change in the genetic composition of a population” or “a change in gene frequency through time.” Natural selection would simply be the disproportional representation of one or the other genetic variants in future generations.

By the 1980s, a bifurcation had occurred within the study of natural selection. The behaviorists or ethologists were adopting the reductive definition and extending it even more broadly—now talking about competition among alleles, and “the selfish gene.” On the other hand, mainstream evolutionary biologists were rejecting the reductive definition altogether, for its failure to grapple with the interaction among genes in producing phenotypes, a failure that Ernst Mayr called “beanbag genetics.” The reductive approach failed to problematize the body, which was implicitly simply the sum of its genes; and failed to conceptualize variation and competition among elements at different levels of a natural hierarchy—between organisms or populations or species.35 After all, the reductive definition addressed simply the transformation of a lineage through time, and not the multiplication of lineages.

The “change in gene frequencies in a population” was evolution all right, but it constituted evolution’s minor features; evolutionary biologists like George Gaylord Simpson were interested in its major features.³⁶ No one doubted—as Darwin took great pains to demonstrate—that the differences one observed within populations, and the processes that produced them, were effectively the same as, but smaller in scale than, those that one observed between species. Nevertheless, it was difficult to see how a well-studied microevolutionary genetic phenomenon—like the spread of the allele for sickle-cell anemia in Africa—actually afforded an adequate description of, say, bipedalism, assuming one could just wait and observe for hundreds of thousands of years.

The problem lay in the facile translation between genetic constitution and body, or between genotype and phenotype. Fruit-fly genetics and human medical genetics had converged on a system of discovering and naming genes, which focused on their major pathological effects. Consequently, geneticists had learned rather little about how genes build a normal, working body, and rather much about how to screw up that process. There are, after all, many more ways to make a bad soufflé than there are to make a good one. Even today, in the wake of the Human Genome Project, we know that it takes two genomes to build a person, and almost nothing about why one or three won’t cut it.37

Natural selection, then, involves the often passive competition between biological forms, for representation in future generations. Such competition requires two attributes: reproducing, or copying; and interacting in some way with an external world that promotes or inhibits that replicative process. We find those properties in three kinds of biological forms: cells, organisms, and species. Cells generally reproduce mitotically and interact physiologically; organisms generally replicate sexually and interact socially; species generally replicate geographically and interact ecologically.

The cells in your body are programmed for division (mitosis), harmonious interaction with other cells, and death (apoptosis). Cells that cheat, by replicating uncontrollably, manage to outbreed the other cells in the short run, but kill the organism of which they are a part. Hereditary cancer consequently is a disease primarily of the middle-aged and elderly, who have finished reproducing—for cancers that strike young people essentially doom themselves. The life cycle of the organism thus places constraints upon the behavior of its cells.

In parallel, the population can place constraints on what organisms can do. In a classic argument, animals can’t reduce their breeding for the good of the group, since anyone who doesn’t do it (“cheaters”) will quickly outbreed the rest. (Since breeding represents evolutionary fitness in the most literal sense, this is the most quintessentially altruistic act in biology, from which all others are simply mathematical deviations.) The only way that lowering your own fitness for the good of the group could happen is if the organisms had foresight (which of course they don’t) or coercive mechanisms by which to discourage cheaters (which they don’t either). On the other hand, you don’t have to think too hard to come up with one species that has both foresight and coercive institutions, so the constraint that “things can’t evolve for the good of the group” does not carry weight in Homo sapiens. As the evolutionary geneticist Francisco Ayala put it,

The fitness advantage of selfish over altruistic behavior does not necessarily apply to humans, because humans can understand the benefits of altruistic behavior (it benefits the group but indirectly it benefits them as well) and thus adopt altruism and protect it, by laws or otherwise, against selfish behavior that harms the social group.38

The differential replication of variants due to the constraints of their setting can thus take place at different levels of a natural hierarchy, and can impact the patterns discernible at other levels. Rates of speciation and extinction of populations, for example, may affect what appear to be the simple proportions of alleles or organisms in a species. More significant, however, is that we formally distinguish between natural selection, as a consistent and non-random bias across generations that shapes the gene pool to fit its circumstances, and genetic drift, as a one-off random blow or tweak to the gene pool. Consequently, studies that examine a snapshot of behaviors or alleles at a single time, and find them to be more-or-less in tune with a hypothesis, and conclude that this is evidence of selection at work, are not really finding evidence of selection, because its most salient point is that multigenerational consistency. We know that the genome can produce non-adaptation and maladaptation, and that bodies can make do with a lot of physical noise, while still maintaining a passable level of functionality; that is the crucial distinction between Herbert Spencer’s “survival of the fittest” and the Darwinian survival of the fit.