The Argentine tango is famous for being a difficult but electrifying dance. Just one look at a performance by professional dancers Mora Godoy and José Lugones shows why. Whether dancing chest to chest or obliquely angled, Godoy and Lugones whip across the floor, legs whirring like blades on a fan. When she raises a bent leg forward, he answers with a quick kick aft. The pair slip easily between the two- and four-beat phrasing of the music, perfectly matching each other's every hip swivel and toe tap, leg lick and foot volley.

Not everyone can move with the fiery grace of this expert duo, of course. But we have all felt the call to dance, which has beckoned countless participants across all cultures throughout human history. Yet dance is rare in the animal kingdom. And although a few other species can move their bodies to a beat, none of them exhibits anything like the complexity seen in human dancing.

Why should dancing be such a common human trait, and why are we so good at it? In recent years scientists have begun to identify features of the brain and body that underpin our exceptional ability. Some of these features are linked to language and upright locomotion, two traits that have contributed significantly to the success of the human lineage. Perhaps, then, dance is a happy evolutionary accident, a by-product of natural selection for those other traits that helped our ancestors thrive. Insights from psychology and archaeology hint at another intriguing possibility, however: that dancing itself evolved as an adaptive trait, one that may have strengthened human social bonds in ways that enhanced survival.

Sense the Beat

Broken down to its basic elements, dancing is the act of sensing and predicting the timing of an external beat and then matching that beat with rhythmic movements of the body. These actions require a great deal of coordination among different parts of the brain.

Over the past decade researchers in Canada, the U.S. and England have begun to identify networks of nerve cells deep within the human brain that act in concert to isolate the beat from external auditory signals. Once these networks recognize the underlying pattern, they predict the timing of subsequent beats, essentially generating a matching arrangement within the brain.

The next step is what makes dancing possible. The parts of the brain that control the muscles start to fire in conjunction with the predicted beats from the auditory networks. (Indeed, these so-called motor-planning areas of the brain kick into action even when we stand still and merely perceive a beat.) This coupling of auditory processing with rhythmic physical movement lies at the heart of our ability to tap out a beat with our fingers or to waltz across the floor. Scientists call it “entrainment.”

Barring illness, we humans come by entrainment naturally, and we can sustain rhythmic movement across a wide range of tempos for long periods. “Our synchronization abilities are incredibly flexible,” asserts Aniruddh D. Patel, a neuroscientist at Tufts University. “We can stay synchronized to a beat whether it slows down or speeds up by plus or minus 30 percent.” This capacity generally emerges between three and five years of age.

For years scientists believed that only humans had the ability to entrain their physical behavior with external sounds. Then, in 2009, studies started to emerge showing that parrots and perhaps songbirds can—to a limited extent—time their movements to music as well. Snowball, a male cockatoo famous for bobbing his head up and down in time to music from the Backstreet Boys, was among the birds studied. Indeed, in a new study published in July 2019, Patel and his colleagues report that Snowball went far beyond just bobbing his head in time to a musical beat. Over a period of 23 minutes, the researchers filmed Snowball performing 14 different dance movements using a variety of body parts to the pop songs “Another One Bites the Dust” and “Girls Just Wanna Have Fun.” This indicates that, as with humans, Snowball’s movement to music involves contributions from motor- planning regions of the brain as well as auditory ones. “Snowball developed this behavior without any training,” Patel says. “That suggests that dancing to music isn’t an arbitrary product of human culture but a response to music that arises when certain cognitive and neural capacities come together in animal brains.” And in 2013 researchers reported that a California sea lion named Ronan could move her head to a range of tempos.

Humans are, however, the only animals that can produce the closely coordinated movements required of partner or group dancing. Birds that can entrain move in spurts to music on their own, Patel says. Even when multiple parrots live together in a shelter, he says, they do not coordinate their movements or dance with one another.

Imitation Game

Dance is not the only human attribute that depends on entrainment. Speech and singing also require the ability to match sound with physical movement—specifically, of the vocal cords and muscles in the throat. Tracing the neural pathways involved in vocalization gave Patel an idea about how entrainment between nerves that process sound and those that control muscles might have evolved. His work suggests that the same neural innovations that allowed humans to learn and produce spoken language also predisposed us to be dancers.

In Patel's view, the ability to mimic sounds paved the way for predictive, flexible entrainment. Such mimicry demonstrates what researchers call “vocal learning,” in which an animal listens carefully to a sound, forms a mental model of it, aligns the motor control of its throat, tongue and mouth with that model, and then produces the modeled sound. When the animal listens to the output, it notes and corrects discrepancies between the predicted and the actual sound and tries again. Patel suggests that the coupling of auditory and motor processing required to imitate sounds laid the neurological groundwork for the later, more complex process of predictive auditory-motor entrainment.

Why might vocal learning have evolved in select animals? Some scientists speculate that it might have enabled songbirds to master complex acoustic displays to advertise for a mate. In parrots, Patel says, it furnished an “acoustic badge—something that marks them as a member of a group.”

If Patel's hypothesis that vocal mimicry is a necessary precondition for auditory-motor entrainment is right, then the only animals that should be able to entrain are those that are already capable of imitating sounds. To date, the only animals that are known to imitate external sounds are humans, hummingbirds, parrots, songbirds, whales, certain flipper-footed marine mammals (pinnipeds), elephants and some bats. Meanwhile our nearest living relatives, bonobos and chimpanzees, are not vocal learners, and most evidence to date suggests that they do not entrain. Although one chimp in a study was apparently able to synchronize her taps with the beat at one tempo, she could not keep the beat at other tempos. Researchers also found one bonobo that seemed to be able to drum to a beat, but they caution that she might have been watching the tester for cues rather than just responding to what she was hearing.

World Beat: Children perform a classical dance in Mumbai, India (1); break-dancers in Los Angeles demonstrate moves (2); modern ballet dancers show flexibility and grace (3); Bolshoi ballet lines up perfectly (4); street parade and festive dance in Cuba (5); Tribal dancers leap in Uganda (6); teenagers strike a pose in the U.S. (7); Jewish dancers celebrate in London (8); Sufi dancers whirl in Istanbul (9); Geisha performers display fans in Japan (10); dance company rocks to the beat in Cuba (11). Credit: Arijit Sen Getty Images (1); Getty Images (2) Getty Images (3); Linda Vartoogian Getty Images (4); Christopher Pillitz Getty Images (5) George F. Mobley National Geographic (6); Getty Images (7); Dan Kitwood Getty Images (8) Gurcan Ozturk Getty Images (9); Frank Carter Getty Images (10); Nika Kramer Gallery Stock (11)

Such observations support the idea that vocal mimicry might be a necessary precursor for entrainment. But they are by no means a slam dunk. Demonstrating entrainment in nonhuman species is not easy. Think of the complicated duets between some species of songbirds. Do they take turns singing by keeping time—predicting when the other will finish—or are they merely reacting to their partner's silence? And how could you possibly test this?

The biggest problem for Patel's vocal-mimicry hypothesis, however, is Ronan, the head-bobbing sea lion. Sea lions are not known to be vocal learners, although they are related to walruses and seals, which are. Yet in 2013 researchers at the University of California, Santa Cruz, demonstrated that Ronan could move her head in time with simple beats and, later, more complex music. Further tests showed that she could correctly keep time with the beat even when it sped up or slowed down.

There are several ways to explain Ronan's apparent ability. Maybe she is just one very gifted sea lion—the exception that proves the rule. Or perhaps sea lions still possess the neural machinery for vocal mimicry and just no longer use it.

It is possible, of course, that Ronan's feat proves the vocal-mimicry hypothesis wrong. Patel and others have suggested that one way to test this hypothesis would be to determine whether horses—which are neither vocal learners nor related to them—can also be taught to entrain. Horses “should not be able to match a specific tempo, but there is widespread anecdotal evidence that they can,” says Mara Breen, an assistant professor of psychology at Mount Holyoke College, who is testing Patel's hypothesis in horses. If it turns out that these animals can entrain, then perhaps the process is not so hard after all, or it evolved in other species for different reasons than it did in humans.

A Role for Running?

Unlike dance in other creatures, human dance goes beyond head bobbing to include coordinated movement of the torso and limbs. How might the evolution of our unusual upright posture have affected our capacity for dance? One idea that has gained attention in recent years is that dance could have grown out of our ability to run—as opposed to just walk—on two legs. “Certainly we take advantage of being bipedal to dance,” says Harvard University evolutionary biologist Daniel E. Lieberman, who, in 2004, co-authored a seminal paper in Nature on the role of endurance running in human evolution. But that differs from what humans evolved to do. “We evolved to walk and to run, to throw, to dig,” Lieberman says. Natural selection for these abilities enabled our ancestors—in particular, Homo erectus—to upgrade their hunting and foraging skills.

“There are all kinds of fascinating adaptations that we think evolved for running,” Lieberman continues. The toes of modern humans are much shorter, for example, than those of our forebears. From a biomechanical point of view, this is unnecessary for walking, but it makes running more efficient. The three semicircular canals of the inner ear have grown larger over the course of millennia, allowing us to maintain our balance whenever we move our head, so that we can move with greater speed and agility. Such adaptations are also useful for dancing.

In Lieberman's view, dance could be a coincidental outgrowth of the evolution of running that proved so useful it conferred its own additional selective advantage. “It doesn't have to be an all-or-nothing thing,” he says. “It can be partial. It could be that dancing was selected for, or it could be that dancing was never selected for, or it could be that certain elements of dancing were selected for.” He pauses. “Testing those hypotheses—boy, that is hard.”

Group Effort

Observations of modern-day dancers offer some tantalizing clues to the kinds of advantages dancing might have conferred in our evolutionary past. A notable feature of human dance is that we tend to do it together. As we feel and predict one another's movements, there is a physical and emotional give-and-take between individuals, whether they are tango partners or throngs of millennials rocking out to Bruno Mars.

This group capability represents what can be called social entrainment, and it confers what Émile Durkheim, who helped to create the field of sociology in the late 1800s, termed “collective effervescence,” or the feeling of being part of something larger than oneself. That kind of social cohesion could be valuable for life-sustaining activities such as food gathering or predator avoidance.

Anthropologist Edward Hagen of Washington State University Vancouver takes that idea a step further. He hypothesizes that music and dance might have evolved as a way for groups to appraise one another when seeking to form alliances that reached beyond the bonds of kinship. How well a group danced together, for instance, might give an indication of how well its members would perform as part of a larger coalition.

Greater social cohesion imparts physiological benefits as well. A 2010 study by scientists at the University of Oxford shows that synchronized physical activity driven by a unified goal—in this case, rowing in the university's boat club—significantly increased participants' pain thresholds compared with solo training. The authors attributed the increase to the release of endorphins, natural opioids in areas of the brain associated with mood. Robin I. M. Dunbar, an anthropologist and evolutionary psychologist at Oxford, argues that these endorphins strengthen social bonds when people engage in group musical activities as well.

“You could imagine two societies, one that didn't dance and one that did, and the one that did would have much stronger social bonds,” says archaeologist Clive Gamble, a professor at the University of Southampton in England. In a competitive situation between the two, he says, the society that danced “would have an evolutionary advantage.”

Given the dearth of direct evidence for the origins of dance, scientists in varying fields have turned to the behavior of today's few remaining hunter-gatherer societies for clues about our ancestral past. Their way of life probably offers the closest approximation that anthropologists have of what human societies were like before the widespread adoption of agriculture 10,000 years ago.

Evolutionary anthropologist Camilla Power of the University of East London studies the Hadza people of northern Tanzania, who typically live in “camps” of 20 to 30 people, in which men and women are social equals. Over the generations, dance has emotionally bound the Hadza and other groups, including the Bayaka people in central Africa and the San people in the Kalahari Desert, together in “shared fictions.” Participants enact initiations, healing rituals and gender relationships, among other things, Power says. Among the Hadza, key dance rituals include feigned “sex wars” in which women taunt men and the men return the favor. “This dynamic is what underlies the egalitarianism,” she says. Women consolidate their power, even playing male roles, goading the men to hunt in return for later “cuddles.”

There is indirect evidence that large group dances have taken place for thousands of years. So-called aggregation sites—large, heavily trampled areas where prehistoric musical instruments have been recovered—provide hints of such activities having taken place among Upper Paleolithic peoples. Among them is Isturitz, a cave in the French Pyrenees, where bone pipes dating to 35,000 to 20,000 years ago were found.

“It's clear from the other archaeological evidence that lots of different groups were gathering at these sites at particular times of the year,” says Oxford paleoanthropologist Iain Morley, author of the 2013 book The Prehistory of Music: Human Evolution, Archaeology, and the Origins of Musicality. “When we see that kind of big group activity in hunter-gatherer societies today, music and dance occur.” Thus, Morley believes, humanity's ancestors were likely making music and dancing for tens of thousands of years—plenty of time for evolution to influence the outcome.

There is one absolute about this most elusive of art forms. Dancing is about communicating, whether it is between the participants themselves or the participants and the observers. Dancers are, in essence, sharing a world of their own invention.

In doing so, they are also changing their brain. Clinicians and researchers alike have acknowledged the benefits of dance for people with movement disorders such as Parkinson's disease. Indeed, many who suffer from the tremors, stiffness and difficulty initiating movements that characterize Parkinson's can, by taking dance classes, regain some of their ability to entrain. As an added benefit, the classes help to form social bonds that may have been diminished by the disease.

Dance classes for people with Parkinson's do not, of course, aim to turn out the next Mora Godoy. But they offer their own transformations. This most ancient of human activities unites body and mind in ways we are only beginning to grasp.