The Thing with Feathers (6 page)

Andrea Cavagna and his colleagues were just getting started. Now that they had real data to play with, they could begin sorting out how flocks form in the first place. Because of their varied backgrounds in physics and statistics, the researchers could look at starlings mathematically. Cavagna had spent most of his career studying logical theories of glasses and supercooled liquids, far from the muddy realms of biology. Now he wondered: Can starling flocks be described with the same equations used in particle physics?

—

I HAVE A FRIEND
who hates starlings. The birds are not native to North America, so they’re not protected by law. A few years ago, my friend began to shoot at starlings in his yard with an air rifle, determined to eradicate them from the area. After a while, the birds caught on and would fly away whenever my friend’s silhouette appeared in the kitchen window, so he built a bunker in his garage. He proudly showed me a tally sheet of kills on a clipboard kept next to a narrow slit overlooking the yard; hash marks were divided into neat groups of five and ordered by date.
When he wings one, he runs outside, clips its flight feathers, and throws it into a large outdoor aviary. Then, whenever he has to drive somewhere, he transfers a few flightless starlings into a small cage on the backseat. Each time he passes a hawk perched on a roadside power pole, my friend tosses a hapless starling out the window and watches—with delight—in the rearview mirror as the raptor glides down for an easy meal.

If you Google “America’s most hated bird,” all of the top results refer to starlings. Such universal agreement is rare in matters of opinion, but on this everyone seems to concur: Starlings are rats with wings. When a pair of them built a nest under the eaves of my house one spring, I climbed up to it on a ladder and removed the eggs. They never came back.

“What we admire in ourselves we often abhor in our neighbors,” writes Jonathan Rosen, and he’s got a point. The starling’s only real fault is success. In the late 1800s, a pharmacist in New York released a host of European birds into the city in an aesthetic attempt to sculpt the New World according to the Old. Most of those releases, including European robins, chaffinches, common blackbirds, and Eurasian skylarks, quickly died out, unable to adapt to a new environment, but when several dozen European starlings were set free in New York in 1890, the birds thrived like cockroaches and bred like rabbits. Those few individuals quickly multiplied across the continent into a population of 120 million, distinguishing the European starling as about the seventh most abundant bird species in North America today (after the American robin, dark-eyed junco, red-winged blackbird, red-eyed vireo, white-throated sparrow, and yellow-rumped warbler, according to Partners in Flight). Few species have ever spread so fast or multiplied so quickly—except humans.

Starlings are often blamed for the population collapse of
cavity-nesting birds such as the eastern bluebird, but research shows that this probably isn’t true. They are also said to bring down airplanes, but I know of only one fatal U.S. commercial airline crash caused by a flock of starlings (geese, ducks, herons, gulls, and cranes are much more dangerous). Starlings do occasionally damage crops, but their effects are slight compared with those of red-winged blackbirds and Canada geese, which eat millions of dollars in grain each year. More bothersome is their poop; as the U.S. Department of Agriculture puts it, “excrement can create a slipping hazard on sidewalks.” In 2008, the New York City Transit Authority paid a man $6 million after he slipped on bird poop in a subway station—but that poop actually belonged to pigeons, not starlings.

Perhaps we should take more time to appreciate the starling’s merits. Up close, European starlings are feathered in beautifully iridescent layers, with black bodies that glow green or blue, depending on the light. They’re perky birds, with quirky personalities, and are excellent vocal mimics; any given starling can imitate about twenty other birds’ sounds. They make good pets when raised from chicks, as their imprinting instinct is strong. Mozart kept a starling for three years and taught it to sing bars of his music. When his bird died, the composer buried it in his backyard and wrote a commemorative poem.

Shakespeare, while writing
Henry IV, Part I
in the 1590s, penned a line that may have unwittingly changed the course of avian history: “I’ll have a starling shall be taught to speak nothing but ‘Mortimer,’” suggests a character named Hotspur, apparently plotting to remind King Henry of Mortimer’s imprisonment by training a starling to constantly repeat the name. Three hundred years later, an eccentric pharmacist in New York—one Eugene Schieffelin, president of the American Acclimatization Society, and a big fan of Shakespeare’s—is
said to have used the line to justify his introduction of starlings to the United States. According to popular lore, Schieffelin tried to import every bird mentioned in the Bard’s plays. Direct evidence for this is shaky at best—the druggist would have had to be pretty sharp to spot such an obscure starling reference (much less every reference to the forty-five-plus species of birds mentioned within Shakespeare’s plays), and crazy to act solely on it—but there’s no doubt that Schieffelin gave America the starling, for one reason or another.

And now they literally flock by the millions. Though a few tens of thousands is normal, particularly large starling flocks have been documented to contain more than 1.5 million birds. One and a half million! Very few animals ever form groups that big. By comparison, the largest known army ant swarms have been counted at around a million individuals, and the world’s largest-ever single-artist concert crowd, a 1994 Rod Stewart performance on Copacabana Beach in Rio de Janeiro, Brazil, drew 3.5 million partiers. Yet those totals wilt in comparison to flocks of passenger pigeons, which used to fly around in groups exceeding a billion birds. Congregations of North Atlantic herring have been documented to contain several billion fish in schools measuring 5 cubic kilometers. Even herring have nothing on the Rocky Mountain locust, a type of grasshopper that used to range across the western United States. A single locust swarm was once estimated to weigh 27 million tons, including 12
trillion
insects. That particular plague, known as Albert’s swarm for the Nebraska physician who documented the event in 1875, apparently covered an area the size of California a quarter-mile deep; that’s a locust for every grain of sand in 1,800 loaded dump trucks, every second in 400,000 years—the entire evolutionary history of
Homo sapiens
, or every dollar in the U.S. national debt, all in one group. If
starlings ever formed a flock that big, it would weigh twice as much as the living human race. (Alas, the Rocky Mountain locust went extinct about thirty years later, and today only North America and Antarctica don’t have locusts. Passenger pigeons also went extinct shortly after forming flocks of billions. Maybe we should be worried.)

Still, a million starlings present an impressive sight. A flock of a mere ten thousand is spellbinding as it twists and turns across the evening sky. And starlings, for all the flak they receive, are generally beautiful birds with funky personalities. Even if they have been unnaturally introduced in many places, starlings deserve more respect.

While unwanted starling populations in the United States are booming, the species is crashing in its native haunts across Europe. The Royal Society for the Protection of Birds recently reported that numbers of starlings in the United Kingdom have fallen between 80 and 90 percent in the past thirty years, the largest decline of any British farmland bird. Nobody knows why; perhaps modern farming methods have eliminated insects from the landscape, or the birds have changed their migration patterns to winter elsewhere. Starlings were red-listed in the United Kingdom in 2002 as a species of severe conservation concern, and recent evidence indicates that their population has declined by 40 million there—from an average of fifteen per household garden to just three—since 1979. The flocks that once floated over Manchester have almost all gone.

—

ITALIAN PHYSICIST ANDREA CAVAGNA
became engrossed early in his career with mathematical theories of supercooled liquids. When the temperature of a liquid is decreased to a certain point, it crystallizes, as water turns to ice. Given certain conditions,
though, it’s possible to maintain a liquid below that critical temperature, and odd things happen to supercooled liquids; for instance, if you cool a liquid fast enough, it becomes a viscous glass that responds extremely slowly even when returned to normal temperatures. Cavagna spent years delving deep into the physics behind these events.

For his Ph.D. research, he studied theoretical physics under the supervision of Giorgio Parisi, one of Italy’s most eminent physicists. Parisi is best known for his investigations of spin glasses, disordered magnets with characteristics similar to chemical glasses, such as windows; he’s also made significant contributions to particle physics and quantum field theory, and won a range of prestigious awards, including the Boltzmann and Max Planck medals. If anything, Parisi likes to focus on disorder—in magnets, glasses, pure statistical theory, and whatever else he can find.

Cavagna picked up some of his famous adviser’s research tastes, including an interest in disordered systems, and continued to study supercooled liquids and glasses after finishing his Ph.D. In 2006, he signed on to a project called StarFLAG, which would change the course of his career.

The ambitious project, overseen by Parisi, was designed to probe the mechanics of starling flocks in order to understand other swarming systems. Teams of scientists from France, Germany, Hungary, the Netherlands, and Italy worked together, each group tackling a different aspect of the issue—computational models, wind tunnel experiments, social theories, and so forth. Cavagna led the group that used stereoscopic photography to record starling flocks in Rome; when they managed to describe a flock of thousands of birds for the first time, he was hooked. Using his training as a theoretical physicist, Cavagna, with his team, cracked a biological phenomenon
that had mystified onlookers for generations—even though he had no experience in biology, wasn’t what you’d call a birdwatcher, and had hardly conducted a single tangible experiment in his life.

The objectives of StarFLAG went beyond starling flocks. In a sweeping mission statement, the project aimed not only to capture data on bird flocks, but also to use those data to construct new models of collective behavior. Then, the scientists hoped, they would be able to apply their knowledge to other fields, the way that Cavagna applied his understanding of physics to biology.

“Collective movements are a common phenomenon also in human behaviour,” the project announced. “We think that [it] is worthwhile to explore the possibility of using the models built for the description of flocking, to describe economic herding behaviours. In this way we hope to get new tools to understand the reasons of social events, e.g. fashions, social dominance.” One researcher on the project decided to focus on how people’s friends influenced the music they downloaded and the way that they voted. Cavagna didn’t know much about the mechanics of fashion trends and market bubbles, but he could contribute his knowledge of physics to try to describe, mathematically, how starlings form cohesive groups. His team of physicists and statisticians got down to work analyzing the data they’d collected from the terrace of the Palazzo Massimo.

They compared the behavior of individual starlings within a flock to the three basic rules used by models dating back to Boids in the 1980s: separation, cohesion, and alignment. Cavagna’s group found that starlings avoid collisions, stay at least a wing’s length away from one another, and seldom stray far enough from one another to break up the flock—just as the models assumed. Starlings also align with one another, but not
quite in the way that flocking models traditionally predicted: Instead of basing directional decisions on birds within a certain distance, each starling uses its nearest seven neighbors to decide which direction to fly in, no matter how far away they are.

This is an important difference. Topological distance—a comparative measurement, like the number of stops on a subway line—appears to be more important than absolute metric distance within a flock. Future models made this adjustment, with good results: When a certain number of nearest neighbors are used instead of those within a certain distance, flocks become less likely to break up, and can expand and contract more easily in response to predators and other fluctuations. And the number seven is particularly interesting. Cavagna mused that even though each starling could probably see more than a dozen individual members of its flock around it, the birds’ brainpower is limited to processing seven at a time.

This is a trait that humans may share with starlings. In 1956, scientist George Miller published a fascinating paper, “The Magical Number Seven, Plus or Minus Two,” which has become something of a legend; the paper has been cited, at last count, more than 16,000 times in other publications. Miller discussed a variety of experiments that showed an odd psychological convergence on the number seven, not in a black-cat-and-mirrors way, but much more logically.

He described one experiment where people were presented with a screen on which random patterns of dots were flashed for one-fifth of a second. When fewer than seven dots were shown, people were almost always able to correctly count the dots, but they often resorted to imprecise estimates when more than seven dots were flashed. In another test, a psychologist read aloud lists of random items at a rate of one per second, and then asked people to repeat what they’d just heard. No
matter what items were being read—words, letters, numbers—people could store about seven unrelated items at a time in their immediate memory, like the seven digits of a phone number. Although these results have been generalized at times past their scientific usefulness—for instance, self-help resources that advise that PowerPoint presentations should have seven main points as a sort of subconscious trick—we do seem to reach certain cognitive limitations near seven items, and starlings may do the same.