The Evolution of Everything: How Small Changes Transform Our World

Hume’s swerve

The first person to dent this cosy consensus was David Hume. In a famous passage from his Dialogues Concerning Natural Religion (published posthumously in 1779), Hume has Cleanthes, his imaginary theist, state the argument from design in powerful and eloquent words:

Look around the world: Contemplate the whole and every part of it: You will find it to be nothing but one great machine, subdivided into an infinite number of lesser machines … All these various machines, and even their most minute parts, are adjusted to each other with an accuracy, which ravishes into admiration all men, who have ever contemplated them. The curious adapting of means to ends, exceeds the productions of human contrivance; of human design, thought, wisdom, and intelligence. Since, therefore the effects resemble each other, we are led to infer, by all the rules of analogy, that the causes also resemble. [Dialogues, 2.5/143]

It’s an inductive inference, Dennett points out: where there’s design there’s a designer, just as where there’s smoke there’s fire.

But Philo, Cleanthes’s imaginary deist interlocutor, brilliantly rebuts the logic. First, it immediately prompts the question of who designed the designer. ‘What satisfaction is there in that infinite progression?’ Then he points out the circular reasoning: God’s perfection explains the world’s design, which proves God’s perfection. And then, how do we know that God is perfect? Might he not have been a ‘stupid mechanic, who imitated others’ and ‘botched and bungled’ his way through different worlds during ‘infinite ages of world making’? Or might not the same argument prove God to be multiple gods, or a ‘perfect anthropomorphite’ with human form, or an animal, or a tree, or a ‘spider who spun the whole complicated mass from his bowels’?

Hume was now enjoying himself. Echoing the Epicureans, he began to pick holes in all the arguments of natural theology. A true believer, Philo said, would stress ‘that there is a great and immeasurable, because incomprehensible, difference between the human and the divine mind’, so it is idolatrous blasphemy to compare the deity to a mere engineer. An atheist, on the other hand, might be happy to concede the purposefulness of nature but explain it by some analogy other than a divine intelligence – as Charles Darwin eventually did.

In short, Hume, like Voltaire, had little time for divine design. By the time he finished, his alter ego Philo had effectively demolished the entire argument from design. Yet even Hume, surveying the wreckage, suddenly halted his assault and allowed the enemy forces to escape the field. In one of the great disappointments in all philosophy, Philo suddenly agrees with Cleanthes at the end, stating that if we are not content to call the supreme being God, then ‘what can we call him but Mind or Thought’? It’s Hume’s Lucretian swerve. Or is it? Anthony Gottlieb argues that if you read it carefully, Hume has buried a subtle hint here, designed not to disturb the pious and censorious even after his death, that mind may be matter.

Dennett contends that Hume’s failure of nerve cannot be explained by fear of persecution for atheism. He arranged to have his book published after his death. In the end it was sheer incredulity that caused him to balk at the ultimate materialist conclusion. Without the Darwinian insight, he just could not see a mechanism by which purpose came from matter.

Through the gap left by Hume stole William Paley. Philo had used the metaphor of the watch, arguing that pieces of metal could ‘never arrange themselves so as to compose a watch’. Though well aware of Philo’s objections, Paley still inferred a mind behind the watch on the heath. It was not that the watch was made of components, or that it was close to perfect in its design, or that it was incomprehensible – arguments that had appealed to a previous generation of physicists and that Hume had answered. It was that it was clearly designed to do a job, not individually and recently but once and originally in an ancestor. Switching metaphors, Paley asserted that ‘there is precisely the same proof that the eye was made for vision, as there is that the telescope was made for assisting it’. The eyes of animals that live in water have a more curved surface than the eyes of animals that live on land, he pointed out, as befits the different refractive indices of the two elements: organs are adapted to the natural laws of the world, rather than vice versa.

But if God is omnipotent, why does he need to design eyes at all? Why not just give animals a magic power of vision without an organ? Paley had an answer of sorts. God could have done ‘without the intervention of instruments or means: but it is in the construction of instruments, in the choice and adaptation of means, that a creative intelligence is seen’. God has been pleased to work within the laws of physics, so that we can have the pleasure of understanding them. In this way, Paley’s modern apologists argue, God cannot be contradicted by the subsequent discovery of evolution by natural selection. He’d put that in place too to cheer us up by discovering it.

Paley’s argument boils down to this: the more spontaneous mechanisms you discover to explain the world of living things, the more convinced you should be that there is an intelligence behind them. Confronted with such a logical contortion, I am reminded of one of the John Cleese characters in Monty Python’s Life of Brian, when Brian denies that he is the Messiah: ‘Only the true Messiah denies his divinity.’

Darwin on the eye

Nearly six decades after Paley’s book, Charles Darwin’s produced a comprehensive and devastating answer. Brick by brick, using insights from an Edinburgh education in bottom–up thinking, from a circumnavigation of the world collecting facts of stone and flesh, from a long period of meticulous observation and induction, he put together an astonishing theory: that the differential replication of competing creatures would produce cumulative complexity that fitted form to function without anybody ever comprehending the rationale in a mind. And thus was born one of the most corrosive concepts in all philosophy. Daniel Dennett in his book Darwin’s Dangerous Idea compares Darwinism to universal acid; it eats through every substance used to contain it. ‘The creationists who oppose Darwinism so bitterly are right about one thing: Darwin’s dangerous idea cuts much deeper into the fabric of our most fundamental beliefs than many of its sophisticated apologists have yet admitted, even to themselves.’

The beauty of Darwin’s explanation is that natural selection has far more power than any designer could ever call upon. It cannot know the future, but it has unrivalled access to information about the past. In the words of the evolutionary psychologists Leda Cosmides and John Tooby, natural selection surveys ‘the results of alternative designs operating in the real world, over millions of individuals, over thousands of generations, and weights alternatives by the statistical distribution of their consequences’. That makes it omniscient about what has worked in the recent past. It can overlook spurious and local results and avoid guesswork, inference or models: it is based on the statistical results of the actual lives of creatures in the actual range of environments they encounter.

One of the most perceptive summaries of Darwin’s argument was made by one of his fiercest critics. A man named Robert Mackenzie Beverley, writing in 1867, produced what he thought was a devastating demolition of the idea of natural selection. Absolute ignorance is the artificer, he pointed out, trying to take the place of absolute wisdom in creating the world. Or (and here Beverley’s fury drove him into capital letters), ‘IN ORDER TO MAKE A PERFECT AND BEAUTIFUL MACHINE, IT IS NOT REQUISITE TO KNOW HOW TO MAKE IT.’ To which Daniel Dennett, who is fond of this quotation, replies: yes, indeed! That is the essence of Darwin’s idea: that beautiful and intricate organisms can be made without anybody knowing how to make them. A century later, an economist named Leonard Reed in an essay called ‘I, Pencil’, made the point that this is also true of technology. It is indeed the case that in order to make a perfect and beautiful machine, it is not requisite to know how to make it. Among the myriad people who contribute to the manufacture of a simple pencil, from graphite miners and lumberjacks to assembly-line workers and managers, not to mention those who grow the coffee that each of these drinks, there is not one person who knows how to make a pencil from scratch. The knowledge is held in the cloud, between brains, rather than in any individual head. This is one of the reasons, I shall argue in a later chapter, that technology evolves too.

Charles Darwin’s dangerous idea was to take away the notion of intentional design from biology altogether and replace it with a mechanism that builds ‘organized complexity … out of primeval simplicity’ (in Richard Dawkins’s words). Structure and function emerge bit by incremental bit and without resort to a goal of any kind. It’s ‘a process that was as patient as it was mindless’ (Dennett). No creature ever set out mentally intending to see, yet the eye emerged as a means by which animals could see. There is indeed an adapted purposefulness in nature – it makes good sense to say that eyes have a function – but we simply lack the language to describe function that emerged from a backward-looking process, rather than a goal-directed, forward-looking, mind-first one. Eyes evolved, Darwin said, because in the past simple eyes that provided a bit of vision helped the survival and reproduction of their possessors, not because there was some intention on the part of somebody to achieve vision. All our functional phrases are top–down ones. The eye is ‘for seeing’, eyes are there ‘so that’ we can see, seeing is to eyes as typing is to keyboards. The language and its metaphors still imply skyhooks.

Darwin confessed that the evolution of the eye was indeed a hard problem. In 1860 he wrote to the American botanist Asa Gray: ‘The eye to this day gives me a cold shudder, but when I think of the fine known gradation my reason tells me I ought to conquer the odd shudder.’ In 1871 in his Descent of Man, he wrote: ‘To suppose that the eye with all its inimitable contrivances for adjusting the focus to different distances, for admitting different amounts of light, and for the correction of spherical and chromatic aberration, could have been formed by natural selection, seems, I freely confess, absurd in the highest degree.’

But he then went on to set out how he justified the absurdity. First, the same could have been said of Copernicus. Common sense said the world stood still while the sun turned round it. Then he laid out how an eye could have emerged from nothing, step by step. He invoked ‘numerous gradations’ from a simple and imperfect eye to a complex one, ‘each grade being useful to its possessor’. If such grades could be found among living animals, and they could, then there was no reason to reject natural selection, ‘though insuperable by our imagination’. He had said something similar twenty-seven years before in his first, unpublished essay on natural selection: that the eye ‘may possibly have been acquired by gradual selection of slight but in each case useful deviations’. To which his sceptical wife Emma had replied, in the margin: ‘A great assumption’.

Pax optica

This is exactly what happened, we now know. Each grade was indeed useful to its possessor, because each grade still exists and still is useful to its owner. Each type of eye is just a slight improvement on the one before. A light-sensitive patch on the skin enables a limpet to tell which way is up; a light-sensitive cup enables a species called a slit-shelled mollusc to tell which direction the light is coming from; a pinhole chamber of light-sensitive cells enables the nautilus to focus a simple image of the world in good light; a simple lensed eye enables a murex snail to form an image even in low light; and an adjustable lens with an iris to control the aperture enables an octopus to perceive the world in glorious detail (the invention of the lens is easily explained, because any transparent tissue in the eye would have acted as partial refractor). Thus even just within the molluscs, every stage of the eye still exists, useful to each owner. How easy then to imagine each stage having existed in the ancestors of the octopus.

Richard Dawkins compares the progression through these grades to climbing a mountain (Mount Improbable) and at no point encountering a slope too steep to surmount. Mountains must be climbed from the bottom up. He shows that there are numerous such mountains – different kinds of eyes in different kinds of animal, from the compound eyes of insects to the multiple and peculiar eyes of spiders – each with a distinct range of partially developed stages showing how one can go step by step. Computer models confirm that there is nothing to suggest any of the stages would confer a disadvantage.

Moreover, the digitisation of biology since the discovery of DNA provides direct and unambiguous evidence of gradual evolution by the progressive alteration of the sequence of letters in genes. We now know that the very same gene, called Pax6, triggers the development of both the compound eye of insects and the simple eye of human beings. The two kinds of eye were inherited from a common ancestor. A version of a Pax gene also directs the development of simple eyes in jellyfish. The ‘opsin’ protein molecules that react to light in the eye can be traced back to the common ancestor of all animals except sponges. Around 700 million years ago, the gene for opsin was duplicated twice to give the three kinds of light-sensitive molecules we possess today. Thus every stage in the evolution of eyes, from the development of light-sensitive molecules to the emergence of lenses and colour vision, can be read directly from the language of the genes. Never has a hard problem in science been so comprehensively and emphatically solved as Darwin’s eye dilemma. Shudder no more, Charles.

Astronomical improbability?

The evidence for gradual, undirected emergence of the opsin molecule by the stepwise alteration of the digital DNA language is strong. But there remains a mathematical objection. The opsin molecule is composed of hundreds of amino acids in a sequence specified by the appropriate gene. If one were to arrive at the appropriate sequence to give opsin its light-detecting properties by trial and error it would take either a very long time or a very large laboratory. Given that there are twenty types of amino acid, then a protein molecule with a hundred amino acids in its chain can exist in 10 to the power of 130 different sequences. That’s a number far greater than the number of atoms in the universe, and far greater than the number of nanoseconds since the Big Bang. So it’s just not possible for natural selection, however many organisms it has to play with for however long, to arrive at a design for an opsin molecule from scratch. And an opsin is just one of tens of thousands of proteins in the body.

Am I heading for a Lucretian swerve? Will I be forced to concede that the combinatorial vastness of the library of possible proteins makes it impossible for evolution to find ones that work? Far from it. We know that human innovation rarely designs things from scratch, but jumps from one technology to the ‘adjacent possible’ technology, recombining existing features. So it is taking small, incremental steps. And we know that the same is true of natural selection. So the mathematics is misleading. In a commonly used analogy, you are not assembling a Boeing 747 with a whirlwind in a scrapyard, you are adding one last rivet to an existing design. And here there has been a remarkable recent discovery that makes natural selection’s task much easier.

In a laboratory in Zürich a few years ago, Andreas Wagner asked his student João Rodriguez to use a gigantic assembly of computers to work his way through a map of different metabolic networks to see how far he could get by changing just one step at a time. He chose the glucose system in a common gut bacterium, and his task was to change one link in the whole metabolic chain in such a way that it still worked – that the creature could still make sixty or so bodily ingredients from this one sugar. How far could he get? In species other than the gut bacterium there are thousands of different glucose pathways. How many of them are just a single step different from each other? Rodriguez found he got 80 per cent of the way through a library of a thousand different metabolic pathways at his first attempt, never having to change more than one step at a time and never producing a metabolic pathway that did not work. ‘When João showed me the answer, my first reaction was disbelief,’ wrote Wagner. ‘Worried that this might be a fluke, I asked João for many more random walks, a thousand more, each preserving metabolic meaning, each leading as far as possible, each leaving in a different direction.’ Same result.

Wagner and Rodriguez had stumbled upon a massive redundancy built into the biochemistry of bacteria – and people. Using the metaphor of a ‘Library of Mendel’, in which imaginary building are stored the unimaginably vast number of all possible genetic sequences, Wagner identified a surprising pattern. ‘The metabolic library is packed to its rafters with books that tell the same story in different ways,’ he writes. ‘Myriad metabolic texts with the same meaning raise the odds of finding any one of them – myriad-fold. Even better, evolution does not just explore the metabolic library like a single casual browser. It crowdsources, employing huge populations of organisms that scour the library for new texts.’ Organisms are crowds of readers going through the Library of Mendel to find texts that make sense.

Wagner points out that biological innovation must be both conservative and progressive, because as it redesigns the body, it cannot ever produce a non-functional organism. Turning microbes into mammals over millions of years is a bit like flying the Atlantic while rebuilding the plane to a new design. The globin molecule, for example, has roughly the same three-dimensional shape and roughly the same function in plants and insects, but the sequences of amino acids in the two are 90 per cent different.

Doubting Darwin still

Yet, despite this overwhelming evidence of emergence, the yearning for design still lures millions of people back into doubting Darwin. The American ‘intelligent design’ movement evolved directly from a fundamentalist drive to promote religion within schools, coupled with a devious ‘end run’ to circumvent the USA’s constitutional separation between Church and state. It has largely focused upon the argument from design in order to try to establish that the complex functional arrangements of biology cannot be explained except by God. As Judge John Jones of Pennsylvania wrote in his judgement in the pivotal case of Kitzmiller vs Dover Area School District in 2005, although proponents of intelligent design ‘occasionally suggest that the designer could be a space alien or a time-traveling cell biologist, no serious alternative to God as the designer has been proposed’. Tammy Kitzmiller was one of several Dover parents who objected to her child being taught ‘intelligent design’ on a par with Darwinism. The parents went to court, and got the school district’s law overturned.

In the United States, fundamentalist Christians have challenged Darwinism in schools for more than 150 years. They pushed state legislatures into adopting laws that prohibited state schools from teaching evolution, a trend that culminated in the Scopes ‘monkey trial’ of 1925. The defendant, John Scopes, deliberately taught evolution illegally to bring attention to the state of Tennessee’s anti-evolution law. Prosecuted by William Jennings Bryan and defended by Clarence Darrow, Scopes was found guilty and fined a paltry $100, and even that was overturned on a technicality at appeal. There is a widespread legend that Bryan’s victory was pyrrhic, because it made him look ridiculous and Scopes’s punishment was light. But this is a comforting myth told by saltwater liberals living on the coasts. In the American heartland, Scopes’s conviction emboldened the critics of Darwin greatly. Far from being ridiculed into silence, the fundamentalists gained ground in the aftermath of the Scopes trial, and held that ground for decades within the educational system. Textbooks became very cautious about Darwinism.

It was not until 1968 that the United States Supreme Court struck down all laws preventing the teaching of evolution in schools. Fundamentalists then fell back on teaching ‘creation science’, a concoction of arguments that purported to find scientific evidence for biblical events such as Noah’s flood. In 1987 the Supreme Court effectively proscribed the teaching of creationism on the grounds that it was religion, not science.

It was then that the movement reinvented itself as ‘intelligent design’, focusing on the old Aquinas–Paley argument from design in its simplest form. Creationists promptly rewrote their textbook Of Pandas and People, using an identical definition for intelligent design as had been used for creation science; and systematically replaced the words ‘creationism’ and ‘creationist’ with ‘intelligent design’ in 150 places. This went wrong in one case, resulting in a strange spelling mistake in the book, ‘cdesign proponentsists’, which came to be called the ‘missing link’ between the two movements. This ‘astonishing’ similarity between the two schools of thought was crucial in causing Judge John Jones to deem intelligent design religious rather than scientific as he struck down the Dover School District’s law demanding equal time for intelligent design and evolution in 2005. Intelligent design, according to the textbook Of Pandas and People, argued that species came into existence abruptly, and through an intelligent agency, with their distinctive features already present: fish with fins and scales, birds with feathers.