1
A COLD SHUDDER
You can live some sort of life and die without ever hearing the name of Darwin. But if, before you die, you want to understand why you lived in the first place, Darwinism is the one subject that you must study.
Richard Dawkins, foreword to John Maynard Smith’s
The Theory of Evolution, 1993
In October 2000, I was embarking upon a natural sciences degree at Cambridge University, not quite sure what I had let myself in for. As was usual, four of us had squeezed into a small office to have our weekly meeting with one of the college tutors, where we would discuss the course material and make sure that we understood the content. Our tutor, Veronica, was a botanist with a bashful manner. She spoke in a quiet voice, barely more than a whisper, and offered her remarks on our written work in light pencil strokes as if afraid of stating her case too forcefully. At the end of an hour’s discussion about something “planty,” the details of which I cannot recall, Veronica announced that she wanted us to go away and write a 2,000-word essay on “why contemplating the eye gave Darwin a cold shudder.”
I panicked. I had picked the “Evolution and Behavior” course for the “Behavior” part, believing (mistakenly as it turned out) that this lay within the purview of psychology, not biology. I hadn’t the faintest idea about evolution and hadn’t read any Darwin. How was I supposed to fathom what might bring him out in a cold sweat? In those days, it was not as easy to rely on Google or Wikipedia to glean any clues as to what the hidden meaning behind this obscure question might be. I started to wonder if coming to Cambridge had all been a big mistake. If I couldn’t even understand the question, how on earth was I going to come up with an answer? I don’t remember what I wrote for the essay but I received a decidedly average mark for it.
Though I was wholly unprepared to answer it, the cold-shudder question cuts to the heart of the difficult problem of explaining the appearance of complex design in nature. The human eye is a marvelously intricate organ. It has a lens, allowing us to focus on objects whether they are far away or right under our nose. Along with other mammals, we have color vision, meaning that we can see somewhere in the region of 100,000 to 10 million different colors. We have different photoreceptor cells, called cones and rods, that are specialized for their roles in daytime and night vision. This is not a book about the eye and I don’t wish to labor the point, but the eye is a pretty amazing bit of sensory gear. From an evolutionary perspective, however, the eye is a bit of a puzzle because the Darwinian mantra is that complex adaptations derive from a series of small, gradual steps, each one bringing a slight improvement in performance and advantage to the bearer. If an eye is only functional once it can actually see things, what possible use is a half-formed version?
Darwin himself acknowledged this difficulty with his grand theory. In response to a colleague’s gentle criticism of his book, Darwin confessed, “About weak points I agree. The eye to this day gives me a cold shudder.” Even now, proponents of creationism brandish the eye as incontrovertible proof that Darwin’s theory must be wrong, and that the appearance of design implies an intelligent omniscient designer. Perhaps this was why he worried about it.
But did he, though?
While Darwin admitted that viewing the eye as being the product of natural selection seemed “absurd in the highest possible degree,” he went on to speculate how a complex eye could, in principle, evolve in successive steps, starting simple and becoming increasingly sophisticated. For this to happen, the tiny changes would have to be passed from parents to offspring and be beneficial to individuals who inherited them. Although it was not until 150 years later that he was vindicated, Darwin’s hunch turned out to be prescient. We now know that complex eyes did evolve in a series of gradual steps—starting out as simple layers of photosensitive cells that allowed individuals to regulate their daily cycles, and incrementally layering on successive features that, at each step, proved advantageous to the bearer.
Variation in traits is most commonly passed from parents to offspring via genes: packages of information that are transmitted, unchanged, from one generation to the next. Genes contain the instructions that your cells need to build proteins; it is these proteins that are the workhorses of life. Your bones, your skin, your fingernails, and your hair are all made of protein. And so is your brain. Your thoughts, feelings, and moods are all events taking place in structures made of protein.
Evolution is change over time—in a biological sense it refers to the rise and fall of different gene variants1 in a population. This ebb and flow of gene variants can occur because of natural processes. Mutations introduce new variants into populations, while stochastic events, like asteroid strikes or volcanic eruptions, can wipe out an entire gene lineage purely by chance. But there is just one force that consistently pushes gene variants in one direction or another: we call it natural selection. This is the process by which gene frequencies change because of the effects the genes have on the bearer. Selection is a blind sorting mechanism: when there is variation in traits, and variation can be inherited by offspring, gene variants that code for beneficial traits will tend to accumulate in the population. As Darwin emphasized, these differences could be slight: “a grain in the balance” yielding just the “slightest advantage” would be sufficient to drive this great engine of change.
All else being equal, a gene variant will tend to increase in frequency in a population when it codes for a trait that confers a survival or reproductive advantage on its bearer, allowing it to outcompete individuals that don’t carry that gene variant. Variants which are advantageous to their bearers—which affect physical or cognitive traits that either increase survival or reproductive success—will tend to accumulate in the population. To use the language of evolutionary biology, these genes will be under positive selection. When we seek to explain why certain behavioral traits, such as aggression or caring for offspring or being kind to strangers, exist and persist within populations, we are implicitly asking how the genetic variants associated with those traits are favored by selection. This doesn’t imply that behavior is either exclusively or deterministically governed by genes, or that genes exert the same effects in all the bodies or environments that they find themselves in. Nevertheless, for traits that do have some genetic component, however small, we can ask how likely it is that these genes will find their way into subsequent generations based on the effects they help to produce.
Taking the gene’s perspective is sometimes called the “gene’s-eye view,” most famously championed by Richard Dawkins in his treatise of “selfish” genes. Genes are selfish but it is important to clarify what this rather loaded term really means. Describing genes as selfish is not implying that they are immoral or conniving or any of the other unsavory character traits we might attribute to selfish people. It is also not a shorthand for describing genes that tend to be associated with selfish characteristics, residing only in the bodies of the most nefarious individuals. On the contrary, every one of the approximately 25,000 genes in your body can be described as “selfish” or, less controversially, as “self-interested.” Genes are self-interested in the sense that they each have a singular overriding “concern”2: to ensure that they appear in subsequent generations.
Taken at face value, the selfish-gene perspective seems to imply that any heritable trait that lowers an individual’s reproductive success or survival (and the gene variants that underpin it) will be ruthlessly weeded out of the population. But if we accept this worldview—with its narrow reading of Darwinian logic—then how do we account for the many examples of cooperation we see in the world around us?
As a concrete example of the phenomena we are trying to understand, think back to the case of the suicidal ants we encountered earlier. At first glance, the existence of such extreme altruism seems to pose a serious challenge to Darwin’s theory. Darwinian logic hinges on the assumption that individuals are driven by self-interest. Most creatures try to survive and have as many offspring as they can, even if this means that there are more hungry mouths than the environment can support. Natural selection acts as a hidden sorting mechanism, a metaphorical sieve: when there is not enough to go around, then only the strongest, the fastest, the fittest3 survive. How and why would evolution have promoted these heroic tendencies, that impose the gravest costs on their bearers while directing benefits to others?
The key to understanding the curious ant behavior is to appreciate that members of the same colony are highly related. It is no coincidence that some of the most striking examples of cooperation in nature occur within the confines of family groups. To understand why charity so often begins at home, we need to expand our view of the ways in which an individual’s actions might yield downstream benefits to its genes—and we can do this by taking the gene’s-eye view. A gene that is present in both my and my brother’s body doesn’t care how it gets into the next generation. From the gene’s perspective, it is irrelevant whether it is transmitted via my own children or via my nieces and nephews. Costly helping behaviors can be favored by selection if the benefits to relatives (in terms of increased offspring) sufficiently compensate the costs (in terms of foregone reproduction) that the helpful individuals face.
This overarching framework for explaining the evolution of social traits is called “inclusive fitness theory” and its logic allows us to make specific predictions about where helping behavior will evolve, and to whom it will be directed. The ants that sacrifice their lives to seal the colony from outside don’t incur a personal reproductive cost because ant workers are sterile. Moreover, scores of relatives stand to benefit when they perform this selfless act, which can explain how such extreme sacrifice could be favored. Relationships among siblings in human families can also be affected by relatedness: research shows that full siblings tend to see more of one another and invest more in the relationship than they do with half-siblings, even when half-siblings are raised under the same roof like full siblings.
Relatedness is a crucial factor in explaining why individuals help one another, and this expanded notion of self-interest allows us to reconcile otherwise puzzling behaviors with Darwin’s theory. But relatedness can’t do all the work. The benefits and costs must stack up as well. Benefits and costs are ecological parameters that also dictate the circumstances under which cooperation will be favored. One way to give cooperation a shove is to reduce the costs associated with helping others—this can happen when it is difficult for individuals to breed independently.
A particularly charming example is that of the long-tailed tit. These gregarious birds are seasonal breeders that flit about in large, chattering flocks but usually try to breed independently, in pairs. Despite their dainty size, long-tailed tits build the most elaborate nests of any European bird species. The nest is a completely covered dome, with a small cavity at the mouth for the parents to pass in and out, and is constructed of materials that seem to have been drawn from a witch’s spell. The exterior calls for scraps of moss and lichen, bound together with spiders’ silk; while the interior is lavishly furnished with thousands of plush feathers. The tits spend over three weeks constructing fairy-tale palaces, a sizeable investment in the context of the barely three-month-long breeding season. Nevertheless, for many birds, the effort is in vain: most nests are discovered by predators, with fewer than one in five pairs successfully fledging their young. If their nest fails when it is close to the end of the breeding season, the hapless pair will typically seek out relatives nearby and help at that nest instead. In this example, the cost of helping is very small: if there is no time left in the breeding season to try for another independent attempt, then the reproductive sacrifice is minimal. These small costs make it relatively easy to tip the balance toward helping relatives instead.
When we look at evolution this way, we can start to make more nuanced predictions about when we should expect help to arise and be favored by selection. For example, individuals should be less willing to help old or extremely sick relatives, since there is a smaller chance of the investment being converted into fitness benefits. This rather unromantic calculation seems to be performed by African Megaponera ants when they raid termite nests. Termite colonies are defended by specialized soldier morphs, which aggressively attack the marauding ants, sometimes wounding them in the process. Injured ants release a pheromone—a chemical cry for help—that stimulates their comrades to rescue them from the jaws of a termite soldier and carry them back to the nest. At the nest, the stricken ants are tended by their sisters, whose licking and grooming prevents the wounds from becoming infected. There’s a twist however: ants that are either too badly injured (for example, having lost five legs rather than one) or too old are not rescued. There is no evolutionary advantage to rescuing a nest-mate who is going to die soon anyway. Ants themselves seem to know when they aren’t worth the trouble and are less likely to send a distress signal if they are aged or too badly injured.
A study of US households in 1910 reveals something similar might occur in humans. In this database, childless couples were more likely to help their relatives (just like the long-tailed tits I described above). Following the logic that not all relatives are equally valuable, these childless couples were more likely to take a niece or nephew into their home than an aging parent. In relatedness terms the parents should be more valuable but, in terms of the potential fitness benefits, aging parents are an evolutionary lost cause: there is more potential to reap a benefit from helping a young niece or nephew instead.
So far, we’ve seen that selection can favor genes associated with cooperative and even heroic actions when this benefits copies of these genes that are found inside other individuals. Genes may be selfish but this doesn’t preclude cooperation, under the right circumstances. Although taking the gene’s-eye view can help to explain many puzzling phenomena, you may have noted a problem with the story I have told. Here’s the difficulty: gene variants are commonly and rightfully viewed as the entities that natural selection works upon, and whose frequency either increases or decreases over evolutionary time. But genes are bundled up into collectives—which we call organisms or individuals—and it is the effects that genes have on these individuals that are exposed to selection. To put it another way, genes do not bear the adaptations that determine their own success. Instead, these design features are carried at a higher level of biological organization, by the individual.
The invention of individuals was an evolutionary masterstroke. Take a moment to consider what a multicellular individual really is: you and I and every other multicellular being on Earth is a collective that operates as a whole rather than as a bunch of parts.
Earlier I said we could think of genes as if they were tiny agents, pursuing their own agenda. But the individuals we see all around us seem to be goal-driven in a similar way. An individual oak tree reaches toward the sun as if its goal is to grow ever taller; an individual great tit carries food to the nest as if its goal is to help the young chicks grow and survive. Behavioral ecologists, like me, tend to talk of individuals, rather than the genes, as being the entities that pursue an evolutionary agenda because it is individuals, rather than genes, that we can see and whose behavior we observe.
But this mental shortcut is justified. Evolution creates individuals by aligning the interests of the genes within them. An individual that pursues her own evolutionary agenda is therefore pursuing the agenda of all the genes from which she is made. This equivalence allows us to think of individuals as being goal-driven agents, safe in the knowledge that we can translate back to the gene’s-eye view at any point.
The evolution of individuals was the first crucial step on the road to ever-increasing social complexity, foreshadowing the transitions to families, to communities, and to large-scale societies. But how do we know when a collection of genes and cells becomes an individual? Why should the person reading this book have the special status of being the individual, rather than conceiving of every cell in your body as an individual or, indeed, every selfish gene?
It feels intuitive to conceive of ourselves as coherent individuals, while simultaneously denying that status to a flock of seagulls or to a herd of wildebeests. But appeals to intuition are an unreliable tool for demarcating these evolutionary boundaries. For example, it probably feels rather counterintuitive to conceive of a colony of ants as an individual, but many evolutionary biologists would argue that ant colonies are superorganisms in their own right. On what basis, other than intuition, can we draw the line? Are you really an individual and, if so, why?
2
INVENTING THE INDIVIDUAL
The individual is, accordingly, a unified commonwealth in which all parts work together for a common end.
Rudolf Virchow, Atoms and Individuals, 1859
Evolution invents new individuals by sewing the interests of the parts tightly together with the whole. Remember the Russian-doll analogy, with the inner dolls representing genes and genomes and cells. The inner dolls have only one route to the next generation: they must travel inside the outer doll, the “individual.” This imprisonment aligns the interests of the inner dolls, meaning that they are incentivized to work together, rather than against one another. Their joint mission is to create the best individual they possibly can, with their ticket to the next generation depending more or less entirely on their success in this collective venture.
Individuals can exist as single cells, as multicellular organisms, or even as entire colonies in some cases. To distinguish individuals from groups or collectives, it helps to recognize that natural selection is not just a process, but an engineer: a force that assembles an entirely new product from a collection of parts. Human engineers frequently have design goals in mind: the people designing a new iPhone will have a target size and weight in mind, as well as other specifications for the quality of the camera and the battery life, for example. Evolution has no foresight, but nevertheless shapes and molds the design features of individuals in a similar way, by sifting out the less fit (or worse adapted) variants from the population. We can identify individuals by spotting the level at which the design features seem to cohere, and to pull in the same direction.
This is a bit abstract so, to illustrate, let’s use a real-life engineering example: the car. The buyer of the car is the hand of selection, and the car is the individual. Just like a multicellular organism, a car is made up of functional subunits. For example, the crankshaft, spark plugs, and pistons that comprise the engine are analogous to the cells, nerves, and muscles that form organs like the heart in your body. These are subunits rather than units in their own right because their functionality is only obvious when considered as part of the car. Assuming that you had no prior knowledge, if you were to find a steering wheel or a piston or even a full engine abandoned on the street absent the vehicle, it would be difficult to determine what these parts were designed to do, whereas their role becomes obvious when they are working in synchrony with the car’s other parts. An individual therefore has the appearance of being goal-driven and possesses functionality in the same way that a car does—but that the component parts lack.
What’s more, when working together, the subunits of the car give rise to features—or adaptations—that are only apparent as properties of the car itself and not of the constituent parts. These design features are what we consider when selecting which car to buy. We don’t pore over the sump and the fan belt—instead, we select based on the properties and features of the car itself. Does it look good? How fast can it go? Is it reliable? By selecting on these properties, we are of course exerting a selection pressure on the component parts despite the fact that we don’t see (or really care) about them. If speed or reliability are things that consumers care about to the extent that they affect differential car sales, then this selection pressure will drive changes in the componentry, in that some kinds of pistons and crankshafts will be more likely to end up in car engines than others. Natural selection sifts among genetic variants in a population in much the same way, by acting upon the design features of the individuals that carry these genes rather than directly upon the genes themselves.
Copyright © 2021 by Nichola Raihani