Life Finds a Way

by Andreas Wagner

  Citation records provide further evidence for the blindness of human creativity by proving that creativity does not depend on a person’s age. Dirac put it most brutally when he called a physicist “better dead than living still, when once he’s past his thirtieth year.”23 Other fields have different myths about when creativity runs out, but it turns out they are just that: myths. The fraction of nuggets found in mental gravel does not change much over a lifetime. That’s what Simonton and others found when surveying areas as different as history, geology, physics, and mathematics. For example, in a survey of more than four hundred mathematicians, work by younger and older mathematicians receives similar recognition.24 Younger creators do not always find more nuggets than older ones, nor do older creators benefit from greater experience.

  A recent analysis of more than two hundred thousand physics publications—and more than half a million articles in fields like biology and economics—also disproves Dirac’s callous statement: physicists and other scientists produce important work at any age, with no discernible trend. And what’s most important, a creator’s best work tends to emerge whenever she produces the most work. That’s the constant probability of success all over again.25

  Lack of foresight. Serendipitous discovery. Many misses. Flops of great minds. Constant probability of success. All these patterns of human creativity underscore the same point. New variants of ideas, images, and concepts are created blindly, in the same sense as new variants of DNA are. Like biological evolution, we are blind to the future success of our creations.

  Although selection’s role in human creativity is more obvious than that of blind variation, it also leaves room for misunderstanding. Clearly, unless our minds select a useful idea, image, or concept for further elaboration, improvement, or publication, that idea will eventually disappear. Such mental selection is just as essential as natural selection is in biological evolution. However, this does not mean that it is the only force driving creativity, or even just the most important one, for the same reason that natural selection’s uphill drive is insufficient for biological evolution: it cannot conquer the complex landscapes of creation, because it is unable to accept inferior solutions that lie on the path to better ones.

  Because we know little about how our brains encode ideas, it may be a long time before we can map the landscapes our minds explore. But we are beginning to understand that our brains organize much information about the outside world—even abstract concepts—in some spatial form. For the most immediate information provided by our senses, this has been known for more than a century.26 Take color, where our minds perceive three major dimensions—hue, saturation, and brightness—such that an object’s color occupies a location in a space of colors.27 One advantage of encoding such information spatially is that objects in a space have a distance, which helps our minds judge instantly whether two colors are similar, like bright orange and bright yellow, or very different, like bright orange and dark purple. Another example is the pitch of a sound, which depends on its frequency and can thus be represented in a single dimension. This is how pitches are mapped onto the linear dimension of our inner ear’s cochlea and how they are encoded deeper in the auditory cortex of our brain.28

  With this knowledge, it takes no great leap to accept that our brains encode other, more complex or abstract concepts, like the sounds of spoken letters, the identity of animals (cat, dog, cow, etc.), or the properties of fruits (color, texture, taste, etc.), in a similar spatial manner. A forceful advocate of this view is Peter Gärdenfors, a Swedish cognitive scientist, who calls such encodings conceptual spaces and argues that we need to understand that thought has a geometry.29 He shows that such spaces can help explain how we compare concepts, how we learn new concepts, and how we create new combinations of concepts.

  Recent experiments prove that Gärdenfors is on to something. In these experiments, participants were shown a cartoon image of a bird on a computer screen and were taught how to use a computer program that can manipulate two aspects of this bird’s body shape: the length of the neck and the length of its legs. The researchers running the experiment trained the participants to use this program to create various bird shapes and to morph these shapes into one another. The researchers wanted to find out how these shapes actually become encoded in the participants’ brains, knowing that they could be encoded spatially; for example, in a two-dimensional conceptual space of varying neck and leg lengths. The trained participants did not know this, but their brains apparently did. When they watched birds morphing into one another, the same brain regions became activated that people use when navigating physical space, in a pattern that is highly characteristic of such navigation. In other words, not only do our brains encode these bird shapes spatially, but in doing so they piggy-back on the same neural circuitry we use to navigate the world.

  The shape of a bird is a pitifully simple concept when compared to the sophisticated methods our minds use to solve complex problems. It may be many years before we know how such ideas are organized in our brain, how a brain explores the mental spaces in which they exist, and how many dimensions these spaces have—I suspect many. Fortunately, these are all small details compared to the most fundamental principle from previous chapters: difficult problems share the fundamental property that their solutions form a rugged landscape. Regardless of how our brains encode these solutions, and whatever space we must explore to find the best ones, we need to overcome this ruggedness. In other words, we can be sure that creative solutions to hard problems anywhere—including those our minds solve—will require finding the high peaks of a rugged landscape in some mental space.

  And that’s where the steady improvement and hill climbing of selection gets into familiar trouble: it does not allow things to get worse before they get better, and it seals off the valleys that need to be crossed. To cross these valleys, nature reins in the power of selection. It turns out that our minds can do something just like that. Picasso’s Guernica helps make that point.

  In a study that retraced Picasso’s journey towards Guernica, Dean Simonton presented all of Picasso’s forty-five sketches to four independent judges after shuffling them to erase any information about the order in which they had been created. The judges’ task was to order these sketches such that the first sketch would be the least similar and the last sketch the most similar to the final painting. Simonton then compared the judges’ ordering of the sketches with the actual order in which Picasso had created them. If Picasso’s sketches had improved steadily toward the final painting, then the judges would order the sketches in the same order in which Picasso had created them.

  But they didn’t. Ordered by resemblance to the final painting, the sketches were scrambled with respect to creation date. Picasso’s mind did not climb a single hill toward his final painting. His path would lead uphill only to go down again, moving in multiple zig-zags toward the final result. Some of his sketches are similar to a figure in the final painting, only to be followed by others that bear little resemblance. Some sequences of sketches appear to steadily improve toward the final painting, only to drop into a valley of little resemblance. What is more, the path zig-zagged also in sequences of sketches experimenting with the same motif—the woman with the dead child, the fallen warrior, or the screaming horse.30 And some motifs didn’t make it into the final painting at all, such as the outstretched fist of the warrior, which became transformed in later versions and eventually disappeared.31

  Simonton’s systematic study quantifies what creative people have known all along: the path toward a creative product is neither straight nor all uphill. Perhaps Rainer Maria Rilke referred to this path when he described in dark imagery that a poet must “have been among the shades.… You have to sit down and eat with the dead.”32 His words echo mythological journeys into the underworld, like those of Orpheus, Virgil, and Dante—themselves powerful metaphors for the trials of creation. Writer Margaret Atwood puts it like this: “Poets travel the dark roads. Inspiration is
a hole that leads downward.”

  The nineteenth-century mathematician and polymath Henri Poincaré—a father of chaos theory—connected his creative journey to another familiar one. It’s the one nature uses to create molecules like bucky-balls, which find the deepest valleys in their energy landscapes but not before traversing many shallow valleys and saddles with unstable molecules. For instance, Poincaré describes one sleepless evening when “ideas rose in crowds; I felt them collide until pairs interlocked, so to speak, making a stable combination.”33 And according to the French philosopher Paul Souriau, ideas arranged in a creator’s mind by chance are ”shaken up and agitated…, form numerous unstable aggregates which destroy themselves and finish up by stopping on the most simple and solid combination.”34 Remarkably, both statements predate the concept of an energy landscape by decades.35

  And recall the testimony of nineteenth-century physicist Hermann von Helmholtz, who compared his progress in solving a problem to that of a mountain climber “often compelled to retrace his steps because his progress stopped” and who “hits upon traces of a fresh path, which again leads him a little further.”36

  These introspections bring us back to an already familiar question: how do creative minds overcome those valleys to get to the next higher peak?

  Chapter 8

  Not All Those Who Wander Are Lost

  Because thinking minds are different from evolving organisms and self-assembling molecules, we cannot expect them to use the same means—mechanisms like genetic drift and thermal vibrations—to overcome deep valleys in the landscapes they explore. But they must have some way to achieve the same purpose. As it turns out, they have more than just one—many more.

  Let’s begin with play.

  I don’t mean the rule-based play of a board game or the competitive play of a soccer match, but rather the kind of free-wheeling, unstructured play that children perform with a pile of LEGO blocks or with toy shovels and buckets in a sandbox. I mean playful behavior without immediate goals and benefits, without even the possibility of failure.

  Play is so important that nature invented it long before it invented us. Almost all young mammals play, as do birds like parrots and crows.1 Play has been reported in reptiles, fish, and even spiders, where sexually immature animals use it to practice copulation. But the world champion of animal play may be the bottlenose dolphin, with thirty-seven different reported types of play.2 Captive dolphins will play untiringly with balls and other toys, and wild dolphins play with objects like feathers, sponges, and “smoke rings” of air bubbles that they extrude from their blowholes.

  Such widespread play must be more than just a frivolous whim of nature. The reason: it costs. Young animals can spend up to 20 percent of their daily energy budget goofing around rather than, say, chasing dinner. And their play can cause serious problems. Playing cheetah cubs frequently scare off prey by chasing each other or by clambering over their stalking mother.3 Playing elephants get stuck in mud. Playing bighorn sheep get impaled on cactus spines. Some playful animals even get themselves killed.4 In a 1991 study, Cambridge researcher Robert Harcourt observed a colony of South American fur seals. Within a single season, 102 of the colony’s pups were attacked by sea lions, and twenty-six of them were killed. More than 80 percent of the killed pups were attacked while playing.5

  With costs this high, the benefits can’t be far behind. And indeed, where the benefits of play have been measured, they can make the difference between life and death. The more feral horses from New Zealand play, for example, the better they survive their first year.6 Likewise, Alaskan brown bear cubs that played more during their first summer not only survived the first winter better, but also had a better chance to survive subsequent winters.7

  Some purposes of such play have nothing to do with mental problem solving. When horses play, they strengthen their muscles, and that very strength can help them survive. When lion cubs play-fight, they prepare for the real fights that will help them dominate the group. When dolphins play with air bubbles, they are honing their skills at confusing and catching prey. And when male spiders play at sex, they practice how to copulate fast enough to get away from a female before other males attack them.8

  But at least in mammals, play goes beyond mere practice of a stereotypical behavior, like that of a pianist rehearsing the same passage over and over again. When mammals stalk, hunt, and escape, they find themselves in ever-new situations and environments. Marc Bekoff, a researcher at the University of Colorado and a lifelong student of animal behavior, argues that play broadens an animal’s behavioral repertoire, giving them the flexibility to adapt to changing circumstances. In other words, animal play creates diverse behaviors, regardless of whether that diversity is immediately useful. It prepares the player for the unexpected in an unpredictable world.9

  That very flexibility can also help the smartest animals solve difficult problems. A 1978 experiment demonstrated its value for young rats. In this experiment, some rats were separated from their peers for twenty days by a mesh in their cage, which prevented them from playing. After the period of isolation, the researchers taught all the rats to get a food reward by pulling a rubber ball out of the way. They then changed the task to a new one where the ball had to be pushed instead of pulled. Compared to their freely playing peers, the play-deprived rats took much longer to try new ways of getting at the food and solving this problem.10

  University of Cambridge ethologist Patrick Bateson linked observations like this more directly to the landscapes of creation when he argued that play can “fulfill a probing role that enables the individual to escape from false endpoints, or local optima” and that “when stuck on a metaphorical lower peak, it can be beneficial to have active mechanisms for getting off it and onto a higher one.”11 In this view, play is to creativity what genetic drift is to evolution and what heat is to self-assembling molecules.

  If that is the case, it is hardly surprising that creative people often describe their work as playful. Alexander Fleming, who would discover penicillin, was reproved by his boss for his playful attitude. He said, “I play with microbes.… It is very pleasant to break the rules and to find something that nobody had thought of.”12 Andre Geim, 2010 Nobel laureate in physics, declared that “a playful attitude has always been the hallmark of my research.… Unless you happen to be in the right place and the right time, or you have facilities no one else has, the only way is to be more adventurous.”13 When James Watson and Francis Crick discovered the double helix, they had help in the form of colored balls they could stick together—LEGO-like—to build a model. In Watson’s words, all they had to do was “begin to play.” And C.G. Jung, one of the fathers of psychoanalysis, said it best: “The debt we owe to the play of imagination is incalculable.”14

  One hallmark of play is that it suspends judgment so that we are no longer focused on selecting good ideas and discarding bad ones. That’s what allows us to descend into the valleys of imperfection to later climb the peaks of perfection. But play is only one means to get there.

  Less deliberate but just as powerful are the dreams that we experience in our sleep. It is no coincidence that the psychologist Jean Piaget, whose trailblazing research helped us understand how children develop, likened dreaming to play.15 It is in dreams that our minds are at their freest to combine the most bizarre fragments of thoughts and images into novel characters and plotlines. Paul McCartney famously first heard his song “Yesterday” in a dream and did not believe that it was an original song, asking people in the music business for weeks afterward whether they knew it. They didn’t. “Yesterday” would become one of the twentieth century’s most successful songs, with seven million performances and more than two thousand cover versions. Another dream whispered to the German physiologist Otto Loewi the idea for a crucial experiment, which proved that nerves communicate through chemicals that we now call neurotransmitters. It would win him a Nobel Prize.

  Even in the state of half-sleep—psychologists call i
t hypnagogia—our minds are sufficiently loose to descend from those lowly hills. In this state, August Kekulé saw the chemical structure of benzene, Mary Shelley found the idea for her iconic novel Frankenstein, and Dmitri Mendeleev discovered the periodic table of the chemical elements.16

  Similar to playing and dreaming is the wandering of our minds. Ninety-six percent of adult Americans report that it happens to them daily—the other 4 percent may be too absentminded to notice. To quantify how often any one mind wanders during a task is simple: ask. Interrupt people who work on the task and ask what’s on their mind. Or let mobile phones do the work for you. Program them to send study participants a text asking what they are thinking about at random times of the day.17 When psychologists do that, they find that mind-wandering is staggeringly frequent. The typical mind is absent between a third and half the time.18

  Mind-wandering is often considered a harmless quirk, as in the cliché of the scatterbrained professor. But it has real consequences. Let’s begin with the bad ones. Absentminded people perform less well on tests that require focused attention, such as reading comprehension tests. More worrisome, they also perform more poorly on tests that you better not flunk if you have any career aspirations. Among them is the Scholastic Aptitude Test (SAT) that many colleges require for admission.19

  But mind-wandering also has an upside—at least for well-trained minds. Indeed, many anecdotes of creators like Einstein, Newton, and Poincaré report that these scientists solved important problems while not actually working on anything. The common wisdom that the best ideas arrive in the shower is exemplified by Archimedes’s discovery of how to measure an object’s volume. (Ok, he was in a bathtub.) But while Archimedes’s discovery was triggered by the rising water as he entered the tub, other breakthroughs surface apropos of nothing. Take this well-known quote from the eminent mathematician Henri Poincaré describing a period in his life when he had worked without success on a mathematical problem: