1
THE POWER OF EXPOSURE
Fame and Familiarity—in Art, Music, and Politics
On a rainy morning one fall, I was walking alone through the impressionist exhibit of the National Gallery of Art in Washington, D.C. Standing before a wall of renowned paintings, I was struck by a question that I imagine many people wonder quietly in a museum, even if it’s rude to say out loud in a company of strangers: Why is this thing so famous?
It was The Japanese Footbridge by Claude Monet, with the blue bridge arching over an emerald green pond that is gilded with patches of yellow, pink, and green—the iconic water lilies. It was impossible not to recognize. One of my favorite picture books as a kid included several of Monet’s water lily paintings. It was also impossible to ignore, on account of several kids scrambling through the geriatric crowd to get a closer look. “Yes!” a teenage girl said, holding up her phone in front of her face to take a picture. “Oh!” exclaimed the taller, curly-haired boy behind her. “It’s that famous one!” Several more high school students heard their shouts, and within seconds a group had clustered around the Monet.
Several rooms away, the gallery held a special exhibit for another impressionist painter, Gustave Caillebotte. This was a quieter, slower affair. There were no students and no ecstatic exclamations of recognition, just a lot of mmm-hmms and solemn nods. Caillebotte is not world famous like Monet, Manet, or Cézanne. The sign outside his exhibition at the National Gallery called him “perhaps the least known of the French impressionists.”
But Caillebotte’s paintings are exquisite. His style is impressionist yet exacting, as if captured with a slightly more focused camera lens. Often from a window’s view, he rendered the colorful urban geometry of nineteenth-century Paris—the yellow rhomboid blocks, the pale white sidewalks, and the iridescent grays of rain-slicked boulevards. His contemporaries considered him a phenomenon on par with Monet and Renoir. Émile Zola, the great French writer who drew attention to impressionism’s “delicate patches of color,” pronounced Caillebotte “one of the boldest of the group.” Still, 140 years later, Monet is one of the most famous painters in history, while Caillebotte is relatively anonymous.
A mystery: Two rebellious painters hang their art in the same impressionist exhibit in 1876. They are considered of similar talent and promise. But one painter’s water lilies become a global cultural hit—enshrined in picture books, studied by art historians, gawked at by high school students, and highlighted in every tour of the National Gallery of Art—and the other painter is little known among casual art fans. Why?
For many centuries, philosophers, artists, and psychologists have studied modern art to learn the truth about beauty and popularity. For understandable reasons, many focused on the paintings themselves. But studying the patches of Monet and the brushstrokes of Caillebotte won’t tell you why one is famous and the other is not. You have to see the deeper story. Famous paintings, hit songs, and blockbusters that seem to float effortlessly on the cultural consciousness have a hidden genesis; even water lilies have roots.
When a team of researchers at Cornell University studied the story of the impressionist canon, they found that something surprising set the most famous painters apart. It wasn’t their social connections or their nineteenth-century renown. It was a subtler story. And it all started with Caillebotte.
• • •
Gustave Caillebotte was born to a wealthy Parisian family in 1848. As a young man, he veered from law to engineering to the French army in the Franco-Prussian War. But in his twenties, he discovered a passion and immense talent for painting.
In 1875, he submitted The Floor Scrapers to the Academy of Fine Arts in Paris. In the painting, white light coming through a window illuminates the bare white backs of several men working on their knees, scraping the dark brown floor of an empty room, as the skinned wood curls into spirals beside their legs. But the painting was rejected. One critic later summed up the scornful response when he said, “Do nudes, but do beautiful nudes or don’t do them at all.”
The impressionists—or, as Caillebotte also called them, les Intransigents—disagreed. Several of them, including Auguste Renoir, liked his quotidian take on the floor scrapers and asked Caillebotte to exhibit with their fellow rebels. He became friends with some of the era’s most controversial young artists, like Monet and Degas, buying dozens of their works at a time when few rich European men cared for them.
Caillebotte’s self-portraits show him in middle age with short hair and a face like an arrowhead, angular and sharpened to a point, with an austere gray beard. A grave countenance colored his inner life as well. Convinced that he would die young, Caillebotte wrote a will instructing the French state to accept his art collection and hang nearly seventy of his impressionist paintings in a national museum.
His fears were prescient. Caillebotte died of a stroke in 1894 at the age of forty-five. His bequest included at least sixteen canvases by Monet, eight by Renoir, eight by Degas, five by Cézanne, and four by Manet, along with eighteen by Pissarro and nine by Sisley. It is not inconceivable that his walls would be valued at several billion dollars in a twenty-first-century Christie’s sale.
But at the time, his collection was far less coveted. In the will, Caillebotte had stipulated that all paintings hang at the Musée du Luxembourg in Paris. But even with Renoir serving as executor, the French government initially refused to accept the artworks.
The French elite, including conservative critics and even prominent politicians, considered the bequest presumptuous, if not downright ludicrous. Who was this scoundrel to think he could posthumously force the French government to hang dozens of blotchy atrocities on its own walls? Several art professors threatened to resign from the École des Beaux-Arts if the state accepted the impressionist paintings. Jean-Léon Gérôme, one of the most famous academic artists of his time, blasted the donation, saying, “For the government to accept such filth, there would have to be a great moral slackening.”
But what is the history of art if not one great slackening after another? After years of fighting both the French state and Caillebotte’s own family to honor the bequest, Renoir persuaded the government to accept about half the collection. By one count, the accepted paintings included eight works by Monet, seven by Degas, seven by Pissarro, six by Renoir, six by Sisley, two by Manet, and two by Cezanne.
When the artworks were finally hung in 1897, at a new wing in the Musée du Luxembourg, it represented the first ever national exhibition of impressionist art in France, or any European country. The public flooded the museum to see art they’d previously savaged or simply ignored. The long battle over Caillebotte’s estate (the press called it l’affaire Caillebotte) had the very effect he must have hoped: It brought unprecedented attention, and even a bit of respect, to his intransigents friends.
One century after the exhibition of the Caillebotte collection, James Cutting, a psychologist at Cornell University, counted more than fifteen thousand instances of impressionist paintings to appear in hundreds of books in the university library. He concluded “unequivocally” that there were seven (“and only seven”) core impressionist painters, whose names and works appeared far more often than their peers. This core consisted of Monet, Renoir, Degas, Cezanne, Manet, Pissarro, and Sisley. Without a doubt, this was the impressionist canon.
What set these seven painters apart? They didn’t share a common style. They did not receive unique praise from contemporary critics, nor did they suffer equal censure. There is no record that this group socialized exclusively, collected each other’s works exclusively, or exhibited exclusively. In fact, there would seem to be only one exclusive quality the most famous impressionists shared.
The core seven impressionist painters were the only seven impressionists in Gustave Caillebotte’s bequest.
• • •
Exactly one hundred years after Caillebotte’s death, in 1994, James Cutting stood before one of the most famous paintings at the Musée d’Orsay in Paris and had a familiar thought: Why is this thing so famous?
The painting in question was Renoir’s Bal du Moulin de la Galette. Standing about four feet high and six feet wide, the artwork shows scores of well-dressed Parisians clustered in an outdoor dance hall, waltzing, drinking, and huddling around tables in the dappled light of a Sunday afternoon in the Montmartre district of Paris.
Cutting instantly recognized the work. But he wondered what was so inherently special about the painting, apart from the fact that he recognized it. Yes, the Bal du Moulin is absorbing, he granted, but the artwork was not obviously better than its less celebrated peers in adjacent rooms.
“I really had an aha moment,” Cutting told me. “I realized that Caillebotte had owned not only the Bal du Moulin, but also many other paintings at the museum that had become extremely famous.”
He returned to Ithaca to flesh out his eureka. Cutting and a research assistant went through about one thousand books of impressionist art in the Cornell University library to make a list of the most commonly reproduced artists. He concluded that the impressionist canon focuses on a tight cluster of seven core painters: Manet, Monet, Cézanne, Degas, Renoir, Pissarro, and Sisley—the Caillebotte Seven.
Cutting had a theory: Gustave Caillebotte’s death helped to create the impressionist canon. His bequest to the French state created the frame through which contemporary and future art fans viewed impressionism. Art historians focused on the Caillebotte Seven, which bestowed prestige on their works, to the exclusion of others. The paintings of the Caillebotte Seven hung more prominently in galleries, sold for greater sums of money to private collectors, were valued more by art connoisseurs, were printed in more art anthologies, and were dissected by more art history students, who grew into the next generation’s art mavens, eager to pass on the Caillebotte Seven’s inherited fame.1
Cutting had another theory: The fact that Caillebotte’s bequest shaped the impressionist canon spoke to something deep and universal about media, entertainment, and popularity. People prefer paintings that they’ve seen before. Audiences like art that gives them the jolt of meaning that often comes from an inkling of recognition.
Back at Cornell, Cutting tested this theory. He gathered 166 people from his psychology class and presented them with paired works of impressionist art. In each pair, one of the paintings was significantly more “famous”—that is, more likely to appear in one of Cornell University’s textbooks. Six times out of ten, students said they preferred the more famous picture.
This could have meant that famous paintings are better. Or it might have meant that Cornell students preferred canonical artworks because they were familiar with those paintings. To prove the latter, Cutting had to engineer an environment where students were unwittingly but repeatedly exposed to less famous paintings the same way that art audiences are unwittingly but repeatedly exposed to the impressionist canon from a young age.
What came next was quite clever: In a separate psychology class, Cutting bombarded students with obscure artworks from the late nineteenth century. The students in this second class saw a nonfamous impressionist painting four times for every one time they glimpsed a famous artwork. This was Cutting’s attempt to reconstruct a parallel universe of art history, where Caillebotte never died prematurely, where his legendary bequest never created an impressionist wing, and where the Caillebotte Seven never benefited from a random historical accident that elevated their exposure and popularity.
At the end of the second course, Cutting asked the 151 students to choose their favorite paintings among fifty-one pairs. The results of the popularity contest turned the canon upside down. In forty-one of fifty-one pairs, the students’ preference for the most famous impressionist works disappeared. The emerald magnetism of Monet’s gardens, the electric polychrome of Renoir, and the genius of Manet were nearly nullified by something else—the power of repeated exposure.
It’s extraordinary that Caillebotte’s bequest helped to shape the canon of impressionism because he purposefully bought his friends’ least popular paintings. Caillebotte made it a principle to buy “especially those works of his friends which seemed particularly unsaleable,” the art historian John Rewald wrote. For example, Caillebotte served as a buyer of last resort when he purchased Renoir’s Bal du Moulin de la Galette. Today, the painting that Caillebotte rescued from obscurity and that inspired Cutting’s famous study of art psychology is considered a masterpiece. When it sold at auction for $78 million in 1990, it was the second most expensive artwork ever purchased. You may find Renoir’s painting to be inherently beautiful—I do—but its canonical fame is inseparable from its absurd good fortune to be among the Caillebotte collection.
Mary Morton, the curator of French paintings at the National Gallery of Art, organized the museum’s 2015 Caillebotte exhibit. She told me that a lack of exposure might account for Caillebotte’s anonymity for another reason: Impressionism’s most important collector didn’t try to sell his art.
One of the most important behind-the-scenes figures in impressionist history is Paul Durand-Ruel, a French collector and dealer who served as a one-man clearinghouse for impressionist paintings before they became world famous. His exhaustive efforts to sell work by Monet and others created and sustained the movement while the French salons and European aristocracy considered their patched style a heinous affront to French romanticism. Durand-Ruel found more success among American collectors. “As the industrial revolution and income growth cranked up, newly wealthy people inhabited big new apartments in Paris and New York City,” Morton told me. “They needed decoration that was affordable, beautiful, and widely available, and impressionist paintings were all three.” New wealth created the space for new tastes. Impressionism filled the void.
But Caillebotte does not fit into this story of impressionism’s popularity among the nouveau riche. He was a millionaire, as the heir to a large fortune in textiles, and he had no need to make money from a painting hobby. There are more than 2,500 paintings, drawings, and pastels attributed to Monet. Despite his severe arthritis, Renoir produced an astonishing 4,000 works. Caillebotte produced about 400 paintings and made little effort to distribute them to collectors or museums. He faded into obscurity in the early twentieth century while his peers hung in crowded galleries and private collections, as the echoing power of Caillebotte’s gift rolled through history.
When today’s high school students recognize Monet’s water lilies, they’re seeing more than a century’s worth of exposure and fame. Caillebotte is the least known of the French impressionists. But it’s not because he’s the worst. It’s because he offered his friends a gift that he was willing to withhold from himself: the gift of exposure.
• • •
For centuries, philosophers and scientists have tried to reduce the vast complexity of beauty into a pat theory.
Some argued for forms and formula. Going back to ancient Greece, philosophers have proposed that beauty is quantifiable, hidden in the fabric of the observable universe. Others inclined to mystical explanations proposed that a precise number—1.61803398875 . . . , otherwise known as “the golden ratio”—could explain the visual perfection of objects like Greek flowers, Roman temples, and modern devices from Apple. They have suspected that the world is ripe with such secrets and equations. Plato proposed that the physical world was an imperfect replica of an ideal realm. Even the most ingenious art or the most dazzling sunset was merely striving toward the unachievably perfect form of Beauty itself. In the 1930s, the mathematician George David Birkhoff went so far as to propose a formula for writing poetry: O = aa + 2r + 2m − 2ae − 2ce.2 (It is unlikely that any person has ever used the formula to write a poem worth reading.)
Is there really an equation lurking in the calculus of the universe that explains why we like what we like? Many weren’t so sure. There were the skeptics, and they argued that beauty is always subjective, residing in individuals rather than in math. The philosopher David Hume said that “to seek the real beauty, or real deformity, is as fruitless an enquiry, as to pretend to ascertain the real sweet or real bitter.” The philosopher Immanuel Kant agreed that beauty was subjective, but he emphasized that people have aesthetic “judgment.” Imagine listening to a beautiful song or standing before an exquisite painting. Losing oneself to wonder is the opposite of brainlessness. Pleasure is a kind of thinking.
The long debate between the formula hunters and the skeptics was missing an important voice: the scientists. Hard data did not enter this discussion until a nearly blind German physicist named Gustav Theodor Fechner came along in the middle of the nineteenth century and, in the process of investigating artistic taste, helped to invent modern psychology.
In the 1860s, Fechner was determined to discover the laws of beauty for himself. His methods were unique, because few had thought to do the simplest thing when approaching a question about people’s preferences: Just ask people what they like. His most famous experiment involved shapes. He had subjects of various ages and backgrounds point out which rectangles they considered most beautiful. (It was early days in science.) He noticed a pattern: People enjoyed rectangles that had the proportions of the golden ratio, with the long sides about 1.6 times longer than the short sides.
It would be lovely to report that the first study in psychology’s history was a triumph. Alas, science is a long journey out of wrongness, and Fechner’s conclusion was fabulously wrong. Subsequent scientists repeatedly failed to replicate it. Not all founding fathers have worshipful ideas.
Fechner’s finding was a dud, but his first instinct was brilliant: Scientists should study people by asking them about their lives and ideas. Over time, this principle yielded all sorts of fruitful conclusions. In the 1960s, the psychologist Robert Zajonc conducted a series of experiments where he showed subjects nonsense words, random shapes, and Chinese-like characters and asked them which they preferred. In study after study, people reliably chose the words and funny shapes that they’d seen the most. It wasn’t that some rectangles were perfectly rectangular. It wasn’t that some Chinese-like characters were perfectly Chinese-like. People simply liked whatever shapes and words they saw the most. Their preference was for familiarity.
This discovery is known as the “mere exposure effect,” or just the “exposure effect,” and it is one of the sturdiest findings in modern psychology. People don’t just prefer friends over strangers or familiar smells over unfamiliar odors. Across hundreds of studies and metastudies, subjects around the world prefer familiar shapes, landscapes, consumer goods, songs, and human voices. People are even partial to the most familiar version of the thing they should know best in the world: their own faces. The human face is slightly asymmetric, which means that a photograph captures a slightly different face than a mirror. People sometimes wince when they see photographs of themselves, and several studies show people prefer the face that they glimpse in a reflection. Does a glassy surface reveal your countenance at its objectively most beautiful? Probably not. It’s just the face you like, because you’re used to seeing it that way. The preference for familiarity is so universal that some think it must be written into our genetic code from back when our ancestors trawled the savanna. The evolutionary explanation for the exposure effect is simple: If you recognize an animal or plant, then it hasn’t killed you yet.
• • •
The philosopher Martin Heidegger once said, “Every man is born as many men and dies a single one.” There are a handful of preferences shared by almost all infants—for example, for sweet foods and harmonies without dissonance. But adult tastes are diverse, in large part because they’re shaped by the experience of life, and each person enjoys and suffers life in a different way. People are born average and die unique.
There is nothing more important to the preservation of the hunter-gatherer groups than having sex and moving from place to place safely. So let’s consider these two pillars of their psychology—What makes a face beautiful? What makes a landscape desirable?—to see the potential origins of a bias toward the familiar.
It’s commonly said that people like faces that are symmetrical. But horizontal equivalence alone isn’t the best predictor of beauty. Think about it: Can’t you tell how attractive somebody is by just looking at one side of their face? And making an unappealing face perfectly symmetrical doesn’t suddenly create a supermodel. The more scientifically rigorous explanation for beauty is that people are attracted to faces that look like lots of other faces.
When it comes to looks, average is truly beautiful. Several studies using computer simulations have shown that blending many faces of the same gender creates a countenance more attractive than its individuals. If you blend a lot of extremely good-looking people together, the composite is even more bewitching. What’s so beautiful about an average face? Scientists aren’t quite sure. Perhaps it’s evolutionary, and a face-of-many-faces suggests genetic diversity. In any case, the appeal is universal and perhaps even innate. In studies of adults and children, based in China and throughout Europe and the United States, the most average faces are judged to be the most attractive.3
Beyond averageness, however, tastes diverge wildly. There is no universal attraction to lip plates, lipstick, or bangs, although you can find thousands of people around the world who find each to be seductive. Many people think glasses are sexy, but this is quite evolutionarily backward. Requiring precisely calibrated glassware technology to go about your day signals bad genes for vision. Ancient hunter-gatherers probably wouldn’t have been smitten by wires resting on the nose and ears to balance optical lenses in front of one’s eyes, but this doesn’t diminish the popularity of the sexy-librarian fantasy. If biological preferences for faces exist, they are soft clay, and culture can mold it into myriad shapes.
Another place to see the branching of adult preferences from a common origin is in scenery. A global study of landscape pictures—like rain forests, savannas, and deserts—found that children around the world seem to prefer the same topography. It looks like a savanna with tree cover, which happens to resemble the East African landscape where the Homo sapiens species may have originated. It would appear that humans are born with what the professor of philosophy of art Denis Dutton calls a “pervasive Pleistocene taste in landscape.”
But adult tastes for scenery are neither pervasive nor Pleistocene. They’re all over the place. Some people prefer the sharp canine tooth of the Matterhorn peak, some love a Maine pond painted pink by the sunset, and others are partial to Morocco’s burnt orange ergs. Some landscape details seem to be universally appealing. For example, people around the world are drawn to the presence of clean-looking water, an ancient and eternal necessity for life. There is some evidence that onlookers from various backgrounds and cultures are attracted to mountains divided by snaking rivers and forests cut by trails that slink away toward a vanishing point. These details signify something human ancestors would have loved to see: a navigable path through the chaos of nature. But as adults see different movies, calendars, magazines, photographs, and vistas, their impression of the “perfect” landscape branches off in a million directions.
In the final analysis, beauty does not reside in forms, or cosmic ratios, or even in the standard-issue wiring of humans’ minds, hearts, and guts. It exists in the interplay between the world and people—which is to say, in life. People adapt. To paraphrase Tennyson, they are the sum of all they have met. They are born average and die unique.
• • •
Before there were social media feeds, before there were cable networks or television broadcast stations, and even before there were modern national newspapers, there were public museums. Not counting the amphitheater, the public museum was arguably the first technology to distribute artistic works—what is now collectively known as “content”—to a mass audience. It is strange, perhaps, to think of the museum as anything like a modern innovation, since to many it brings to mind antiquity, mothballs, and young children announcing that they have to pee. But like so much technology, from steam engines to smartphones, the public museum democratized a market—artworks and artifacts—making available to the masses that which was previously accessible only to the rich.
Although crowds have gawked at publicly displayed art for millennia, most art collections throughout history have been private and closely guarded by royals. The modern public museum was an invention of the Enlightenment and its radical notion that ordinary plebes deserved an education. The first national public museum was the British Museum, which opened in 1759 as a “cabinet of curiosities,” including artifacts from ancient Egypt and flora from Jamaica. National museums flowered throughout Europe and across the Atlantic in the next few decades. The American polymath Charles Willson Peale founded America’s first modern public museum in Philadelphia in 1786, with thousands of species of plants and paintings of animals from his collection.4 The Louvre opened in Paris in 1793, and the Prado, in Madrid, followed in 1819. The Caillebotte bequest graced the walls of the Musée du Luxembourg at the height of this craze for new public museums in Europe. In the second half of the nineteenth century, one hundred museums opened in Britain alone.
If public museums have been, for several hundred years, the most important real estate in art, then radio has been the public museum of pop music, the great hallway of mass exposure. Airplay was so critical to building popularity for new music in the mid-twentieth century that music labels developed elaborate “payola” schemes to directly pay radio stations to play their songs. Even into this century, omnipresent airplay is critical to make a hit. “Every bit of consumer research we’ve ever done shows only one consistent thing: Radio is the number one driver of sales and the biggest predictor of a song’s success,” says Dave Bakula, senior vice president of analytics at Nielsen, which tracks music sales and airplay. “You almost invariably see the biggest songs hit radio first, then pick up [in other platforms].” Public exposure on radio can be even more powerful than “mere” exposure, because a song’s presence on a Top 40 station offers other cues about its quality, like the sense that tastemakers and other listeners have already heard and endorsed it.
Even at the dawn of the American music business, to make a song a hit, a memorable melody was secondary to an ingenious marketing campaign. “In Tin Pan Alley, what publishers understood was that no matter how clever, how catchy, how timely a song, its [success] depended on its system of distribution,” music historian David Suisman wrote in Selling Sounds: The Commercial Revolution in American Music. In New York City near the end of the nineteenth century, writers and publishers near a part of Union Square nicknamed “Tin Pan Alley” developed an elaborate process of plugging new music. They would pass out song sheets to local musicians, who would play each tune in different neighborhoods, from the Lower East Side to the Upper West, and report back on which songs clicked. The American standards that came out of this period—such as “The Band Played On,” “Take Me Out to the Ball Game,” and “God Bless America”—were the products of an elaborate testing and distribution strategy that ran on sheet music and shoe leather.
Tin Pan Alley’s pluggers gave way to radio, and now radio is giving way to new forms of distribution that are more open, equal, and unpredictable. Today’s hits break off TV commercials, Facebook posts, and online videos. A Spotify playlist by Napster cofounder Sean Parker is widely credited with launching Lorde’s “Royals,” the surprise hit of 2013. Two years earlier, a Canadian singer-songwriter, Carly Rae Jepsen, released a peppy song, “Call Me Maybe,” that debuted at ninety-seven in the Canadian Hot 100. By the end of the year it still wasn’t in the top twenty. But another Canadian pop singer, Justin Bieber, heard the song on the radio and then praised it on Twitter. In early 2012, Bieber made a YouTube video of him and several friends with fake mustaches, including pop star Selena Gomez, dancing to it. That video now has more than seventy million views, and it helped launch “Call Me Maybe” (which itself has more than eight hundred million views on YouTube) to become one of the biggest pop songs of the decade.
Music—and, for that matter, all culture—attaches itself to such moments, and now that moment can come from anywhere. Terrestrial radio still holds great distribution power—after all, that’s how Bieber first heard “Call Me Maybe”—but it no longer has a monopoly on exposure. Every social media account, every blogger, every website, and every promiscuously shared video is essentially a radio station.
It would be nice to think that in a cultural market like music, quality is everything, and each number-one hit is also the best-in-class. Plus, it seems awfully hard to prove otherwise. How do you demonstrate that a song nobody has heard of is “better” than the most popular song in the country? You would need something crazy: a parallel universe to compare where thousands of people have listened to the same songs and come to different conclusions without the power of marketing.
In fact, that parallel universe exists. Music labels consult it all the time. It is the universe of HitPredictor and SoundOut, online song-testing companies that ask thousands of people to evaluate the catchiness of new songs before the general population has formed an opinion about them.
As the name suggests, HitPredictor (which is owned by iHeart Media, the largest owner of FM and AM radio stations in the United States) “predicts” what songs will be hits by playing a hook from a new song to an online audience three times without telling them too much else about it. The point is to capture the song’s pure “catchiness” in a vacuum. Audiences give the song a numerical rating: A song can score high into the 100s, but any score above 65.00 is considered eligible to be a breakout hit. Sixty-five is the threshold: Above that level, a track has the intrinsic appeal to be one of the top songs in the country.
Here are the HitPredictor ratings of several extremely popular songs that entered the top five on the Billboard Hot 100 in the fall of 2015:
“Hotline Bling,” Drake: 70.25
“The Hills,” The Weeknd: 71.39
“Stitches,” Shawn Mendes: 71.55
“Sorry,” Justin Bieber: 77.14
“What Do You Mean?,” Justin Bieber: 79.12
“Hello,” Adele: 105.00
Study those numbers for a second. Do you notice something weird? A song can theoretically score in the 100s, but most of these massive hits—and they were all massive hits—are just a few notches above the 65.00 threshold. There is nothing at 80.00 or above, except for the incredible outlier of “Hello.” Keep that mystery in the back of your head for a few paragraphs, because there is something deceptively powerful about all these hits scoring “only” in the 70s.
SoundOut is a similar company based out of the UK, which tests about ten thousand tracks online each month. Each new song is randomly streamed to more than one hundred people, who rate it after at least ninety seconds. At SoundOut, the magic number is 80, and any song scoring above that threshold—about 5 percent of tested music—is deemed sufficiently catchy to be a hit. The best-performing recording in SoundOut history was Adele’s sophomore album, 21, which included three worldwide number one hits and won the 2012 Grammy Award for Album of the Year. “Every song on the 21 album scored above an 80,” SoundOut founder and CEO David Courtier-Dutton told me in late 2015. “We’ve never seen that before, and we’ve never seen it since.”
Both HitPredictor and SoundOut find that, yes, there is such a thing as quality or catchiness. Melodies that fail to hit their magic numbers tend to fail in the real world as well.
But look back at the top hits of late 2015: Songs in the 70s routinely beat dozens, if not hundreds, of songs that scored in the 80s and 90s. Above a certain level, catchiness doesn’t make a song a monster hit. Exposure does.
“For every great song that makes it into the charts and has months of airplay, there are a hundred other songs that are just as good, if not better, which, if sung by the right artist with the right marketing, would be a smash hit,” SoundOut’s Courtier-Dutton said. “It is absolutely, categorically true that there are thousands of songs out there that will never see the light of day because they will never get the distribution they require to catch in the market, even though they scored above 80.”
What’s holding back the success of thousands of catchy-enough songs? Sometimes they simply lack the marketing might of a label, or the luck of a viral online video, or the support of a celebrity like Justin Bieber. Sometimes DJs don’t care about the artist or the song doesn’t fit their playlist. Perhaps the band is recalcitrant and a total pain to market. Perhaps it’s several of these things at once. But the point is that every year hundreds of songs won’t become hits, and it will have very little to do with the fact that they weren’t “catchy enough.”
It is the Caillebotte effect all over again: Two pop songs come out. Independent surveys determine they are equally catchy. But one song becomes a massive hit—ubiquitous in coffee shops, praised by mainstream music sites, adored by high school students, and even parodied on YouTube—while the second song is widely ignored and ultimately forgotten because, for some reason, it never received that crucial moment of public consecration. There are simply too many “good-enough” songs for every worthy hook to become a bona fide hit. Quality, it seems, is a necessary but insufficient attribute for success.
Critics and audiences might prefer to think that markets are perfectly meritocratic and the most popular products and ideas are self-evidently the best. But the universes of HitPredictor and SoundOut prove that for every hit song you’ve ever heard there are hundreds of equally catchy, but relatively anonymous, melodies that you haven’t. Beyond a certain level of songwriting genius, how many times audiences have heard a melody matters more for its popularity than how inherently catchy it is.
• • •
In a world of scarce media—just one public French museum, or just three local radio stations—popularity is more top-down. Hits are easier to control and easier to predict. But today there are more than eighty museums in New York City alone. On streaming sites, like Pandora, Spotify, and Apple Music, there are millions of public and personalized radio stations. The power of the press belongs to anybody with a smartphone. In this bottom-up world, where cultural authority shatters into a million channels of exposure, the hits are harder to foresee—and authority is harder to protect.
Consider the nation’s most solemn arena for popularity contests: political elections. “Politics as entertainment” is a common phrase in the press, but the truth might be one letter off; for better or worse, politics is entertainment.
Every political campaign is a media organization. Political campaigns spend half their money on advertising. Elected representatives spent 70 percent of their time engaged in what any sane person would recognize as telemarketing—directly asking for money, asking other people to ask for money, or building relationships with wealthy people, which is a politely indirect way to achieve the same goal.
Even governance is showbiz: One third of the White House staff works in some aspect of public relations to promote the president and his policies, according to political scientists Matthew Baum and Samuel Kernell. The White House is a studio, and the president is its star.
But the star power of the presidency has shrunk in the last few decades at the same time that the channels of exposure have grown.
The most successful way for a president to shape public opinion is to speak directly to voters. Owning the public’s attention used to be a simpler task. In the 1960s and 1970s, CBS, NBC, and ABC accounted for more than 90 percent of the TV audience. It was a heyday for the bully pulpit: In 1970 alone, President Richard Nixon delivered nine prime-time addresses to the nation. The typical address of both Nixon and his successor, Gerald Ford, reached half of all television-owning households.
As the television channels grew, however, the American president became easier to ignore. Ronald Reagan, whose telegenic skills were legendary, reached less than 40 percent of households on average, and Bill Clinton’s loquacious charm got him only to 30 percent. Meanwhile, the average presidential sound bite on the news shrank from forty seconds in 1968 to less than seven seconds in the 1990s. Cable created the golden age of television, but it ended the golden age of presidential communication.
The president is shrinking, and so is the political party. For the past half century, the best predictor of a political candidate’s electoral success was the so-called invisible primary of endorsements from politicians, party leaders, and donors. According to one theory called “the party decides,” it is Democratic and Republican elites, not voters, who decide on their favorite candidates, and these authority figures send signals through the media to the obedient rank and file. It is quite like the old information flow of the music industry: Authority figures (labels and DJs) blasted their preferred products (songs) through scarce and powerful channels of exposure (radio stations) and consumers typically obeyed (bought albums).
But in the 2016 primaries, the apparent power of advertising all but vanished. The GOP candidates with the most elite support, Jeb Bush and Marco Rubio, spent about $140 million on television ads through early 2016, but they both flamed out. The GOP candidate with the least elite support, Donald Trump, spent less than $20 million on advertising. But he still won the exposure primary in a landslide, because his outrageous statements and improbable candidacy were such irresistible fodder for networks and publishers desperate for audiences. Through the summer of 2016, Trump had earned $3 billion in “free media,” which was more than the rest of his rivals combined. With the rise of alternative media sources, party elites lost their ability to control the flow of political information to voters, and an unprecedented figure of single-minded celebrity, authoritarian bluster, and nativist bombast ran away with the Republican primary.5 The party didn’t decide. Rather, like the concentration of political media itself, it seemed to dissolve.
In politics, as in any industry, there is a product, a marketing strategy, and a buying opportunity—a politician, a campaign, and a vote. In both politics and business, research shows that advertising is most powerful when consumers are clueless. Political advertising, for example, is most potent when voters are ignorant of politics in general or with the choices in one particular election (that’s one reason why the influence of money tends to be greater in local elections that voters don’t follow as closely).
Similarly, corporate brands are most powerful in markets where consumers have little information, according to Itamar Simonson, a marketing professor at Stanford, and Emanuel Rosen, a former software executive. It could be because the product is somewhat technical (e.g., toothpaste, since most consumers don’t actually know what gel is best for their enamel) or because the product is refined (e.g., wine, where studies have found that consumers prefer any vintage they think is expensive).
But just as cable and the Internet have washed away the power of political authorities, challenging the theory that the party decides, the Internet’s flood of information is also diluting the brand power of several consumer products. Consider the market for flat-screen televisions: There are only a few relevant details about a big screen that projects images, like width and resolution. Anybody can find those details online, so who needs to consult the plastic name at the bottom of the screen? No wonder, then, that the business of selling flat-screen televisions has been a disaster: The price of TVs declined 95 percent between 1994 and 2014. In that same period, Sony’s television unit lost money every single year.
When consumers don’t know the true value of the products they’re looking for, they rely on corporate iconography to guide them. But when they can figure out the absolute value of a product on their own, they ignore advertisements and brands. That’s why Simonson and Rosen have named their theory “absolute value.” The Internet, they say, will be a brand-assassinating technology, flooding the world with information and drowning out the signal of advertising for many products.
In the 1890s, a single museum had the power to set an artistic canon. In the 1950s, a handful of television channels had the power to brighten every living room with a vision of the president.
No more. Cable television drowned the bully pulpit. Social media is eroding the parties. The Internet drowns out corporate branding. In all sorts of markets—music, film, art, and politics—the future of popularity will be harder to predict as the broadcast power of radio and television democratizes and the channels of exposure grow. Today, there are so many platforms that nobody—not the president, not the Republican Party, not Coca-Cola—can hope to own them all at once. The gatekeepers had their day. Now there are simply too many gates to keep.
• • •
The power of exposure is pervasive—from art and music to politics and brands—but its origin is elusive. How does familiarity shape-shift into pleasure, so that “I liken this to something else” becomes simply “I like this”?
In his 1790 treatise Critique of Judgment, Immanuel Kant proposed that pleasure can arise from a “free play” of the mind. When a person discovers an attractive idea or story, it triggers a dialogue between imagination and understanding, each quickening the other. According to free play, beautiful art, music, and ideas offer a kind of cognitive tease: They dangle the promise of comprehension but never provide the full satisfaction of getting it.
Free play is a lovely idea: How pretty to think that our thoughts and feelings have dancing partners. And perhaps it is more than philosophical poetry.
Two centuries after Kant’s treatise, psychologists developed the idea of “metacognition.” They proposed that there is a level of thinking above thinking. People have thoughts about their thoughts, and feelings about their feelings.
Have you ever heard somebody say, “This is hurting my brain”? It’s often a joke, but it acknowledges something quite real. In a way, we can feel our thoughts. Some of them feel easy: like imagining words that rhyme with “hat,” listening to simple repetitive music, or hearing an eloquent argument for a political position one already agrees with. But sometimes thinking feels like work: like imagining words that rhyme with “strategy,” listening to avant-garde electronica without a time signature, or processing a complex argument for a political position one considers abhorrently wrong.
There is a psychological term for thinking that feels easy, and fortunately it’s easy to remember, too. It’s called “fluency.” Fluent ideas and products are processed faster and they make us feel better—not just about ideas and products we confront, but also about ourselves. Most people generally prefer ideas that they already agree with, images that are easy to discern, stories that are easy to relate to, and puzzles that are easy to solve.6
One of the most important sources of fluency is familiarity. A familiar idea is simpler to process and place in the mental map. When people see an artwork that reminds them of something they’ve been taught is famous, they feel the thrill of recognition and attribute the thrill to the painting itself. When they read a political argument that reflects their biases, it fits snugly into their story of how the world works. Thus, familiarity, fluency, and fact are inextricably linked. “That idea sounds familiar,” “That idea feels right,” and “That idea is good and true” spill into each other in one mental mush.
But not all thinking feels so easy. Some ideas, images, and symbols are more difficult to process, and the term for hard thinking is called “disfluency.” Just as the mental mush conflates fluency with good ideas, people tend to consider disfluency a sign that something is wrong. There is a play-it-at-home game to explain this effect. Follow these four steps:
- Think of the last movie, play, or TV show that you finished: Whisper it to yourself.
- Between 1 (awful) and 10 (perfect), imagine how you might rate it.
- Now think of seven specific things you liked about the movie or show. Count them on your fingers and don’t stop until you hit seven.
- Finalize your rating of the show.
This sort of game is famous because something curious often happens: Between steps 2 and 4, the rating typically goes down.
Why should one’s opinion of a show decline as you think of more reasons to like it? After a few easy examples come to mind, the effort to dig out more examples becomes palpably difficult. People experience disfluency. And sometimes, they misattribute the feeling of disfluency to the quality of the show itself.
This is the “less is more” or “less is better” effect. It means that less thinking leads to more liking. A cheeky UK experiment found that British students’ opinion of former prime minister Tony Blair sank as they listed more of his good qualities. Spouses offer higher appraisals of their partners when asked to name fewer charming characteristics. When something becomes hard to think about, people transfer the discomfort of the thought to the object of their thinking.
Almost every piece of media people consume, every purchase they make, every design they confront lives on a continuum between fluency and disfluency—ease of thinking and difficulty of thinking. Most people lead lives of quiet fluency. They listen to music that sounds like the music they’ve already heard. They look forward to movies with characters, actors, and plots that they recognize. They don’t heed political ideas from opposing parties, particularly if these ideas also seem painfully complicated. As we’ll see in the next chapter, this is a shame, because the greatest joys often come from discovering fluency in places you didn’t expect.
Fluency’s attraction is obvious. But there is a quieter truth: People need a bit of its opposite. They want to be challenged, shocked, scandalized, forced to think—just a bit. They enjoy what Kant called free play—not just a monologue of fluency, but a dialogue between “I get it” and “I don’t” and “I want to know more.” People are complicated: curious and conservative, hungry for new things and biased toward the familiar. Familiarity is not the end. It’s just the beginning.
This might be the most important question for every creator and maker in the world: How do you make something new, if most people just like what they know? Is it possible to surprise with familiarity?