Waura musicians play the Urua flute during the Quarup festival of the dead.Sergio Moraes / Reuters

Imagine that you’re a researcher who has unlimited time and resources, and a time machine that can travel anywhere in the world. You use these wondrous gifts to get a recording of every song that has ever been sung, whether by people in big cities or those in small hunter-gatherer groups. You play these recordings to random volunteers, and ask them to guess the behaviors that were associated with each tune. Was the song used for dancing? For soothing a baby? For healing illness? Could people guess what songs are for by their sound alone, without any knowledge about their cultural context?

When Samuel Mehr and Manvir Singh posed this scenario to 302 cognitive psychologists, who study how people think, around 73 percent predicted that the listeners would make accurate guesses. But when the duo surveyed 206 ethnomusicologists, who study the music of different cultures, just 29 percent felt the same. The two groups of scholars largely disagreed, and Mehr and Singh think the ethnomusicologists are wrong. Music, they say, does have certain universal features that allow even untrained ears to predict its function.

As cognitive psychologists themselves, they’re hardly impartial, but they’ve backed up their claim with an experiment that’s similar to their hypothetical one. Even without unlimited resources and an omniversal traveling machine, they managed to amass songs from 86 cultures around the world—all small-scale societies, like hunter-gatherers and subsistence farmers. Their collection—the Natural History of Song discography—represents music of four types: dances; lullabies; expressions of love; and healing songs intended to cure the sick in ceremonies.

The team then played these songs to 750 volunteers recruited through the internet—a third from the United States, a third from India, and another third from a mix of 58 other countries. Every participant listened to 36 recordings, and rated how likely each one was to be, say, a dance song or healing song.

They were surprisingly accurate, for every category except love songs. Dance songs and lullabies, in particular, share enough features around the world that naïve listeners can identify them with no experience of the cultures from which they arise. The three groups of volunteers were also surprisingly consistent. “Some random person from Texas who’s doing our survey is expected to have a similar conception of what a healing song should be to someone at her computer in India,” says Singh.

In many ways, the Natural History of Song is a 21st-century take on a grandiose project from the 1960s called Cantometrics. Led by ethnomusicologist Alan Lomax, the Cantometrics team systematically analyzed 4,000 songs from 400 cultures around the world according to 37 factors—everything from group cohesion to breathiness to rasp. It was a massive attempt to statistically link the traits of the songs to the characteristics of the cultures that produced them.

The problem is that the Cantometrics collection “wasn’t built in any systematic way,” says Mehr. “It was just a collection of lots of interesting music from all over.” To create something more representative, he and his colleagues deliberately sought out song recordings from 30 regions that cover the whole globe. On a shoestring budget, they contacted anthropologists and ethnomusicologists for any unpublished recordings, and scoured libraries for published ones. “I know we annoyed the hell out of the librarians in [Harvard’s] Loeb Music Library,” Mehr says.

They also annoyed a lot of ethnomusicologists. When the team announced the Natural History of Song project on the Society for Ethnomusicology listserv in late 2016, some members accused them of “denying human agency and geopolitics in [the] very title” while others worried that the project sounded like a colonial “search for the pristine.” “It’s like because we’ve taken up a question that hasn’t been asked for a long time, we were portrayed as mid-century armchair anthropologists with many assumptions, some of them racist,” says Singh.

This backlash has its roots in the reaction to Cantometrics. Many ethnomusicologists felt that the team’s quantitative analysis was insensitive to the cultures that their 4,000 songs came from. The project created a backlash against the search for universal qualities, in favor of focusing on subjective, individual experiences.

“I am skeptical of this sort of attempt to impose order on humanity’s music making by scholars with relatively little on-the-ground ethnographic experience,” says David Locke, an ethnomusicologist at Tufts University. In the West African songs he studies, songs of very different styles can be repurposed for all kinds of functions. “Songs associated with war or death can be sung to soothe an infant—but there would not be a thunderous drum ensemble and full dance ensemble present,” he says. “When I teach courses that ask students to listen to unfamiliar music, they usually make wrong associations between the singing style of the selection and its use in human life.”

“While music is universal, its meanings are not,” adds Anne Rasmussen, an ethnomusicologist at the College of William and Mary. And those meanings are created both by the people making and hearing the music, and by the entire cultural package that surrounds it. A Bach cantata that was composed to celebrate God, for example, means something very different when played in a 21st-century concert hall or in a New York deli. The meaning of music, in other words, “is not something you can perceive while listening through a pair of headphones,” says Rasmussen.

She adds that most of the volunteers who rated the songs used in Mehr and Singh’s study would struggle to name, say, lullabies or healing songs in their own cultures. “These categories of song are ones that the subjects likely barely sing themselves if at all,” she says. “The assumption that we can recognize and name the intention of the expressive culture of small-scale societies without, ourselves, participating in the same kinds of activities seems extraordinarily imperialist, and essentialist.”

Mehr denies the charge of imperialism. “If we took songs from Central Africa and had orchestral musicians reproduce them with violins, that would be weird,” he says. “But we’re just taking the songs and playing them to people around the world, and asking them what they hear.” He’s not privileging the Western perspective on those songs. Indeed, he’s arguing that this perspective doesn’t matter. According to his results, people from around the world, for all their cultural differences, hear something in these songs that is similar to what the original singers intended.

That’s even the case for healing songs—a genre that, as Rasmussen suggests, is largely unfamiliar to people from the United States. “It’s kind of nuts,” says Mehr. “They’re not part of popular music, and yet, people are reliably rating songs that are actually healing songs.”

“It’s a very thorough and important [study],” says Pat Savage, a musicologist at Keio University Shonan Fujisawa Campus, who has also looked at common traits in the world’s music. “It gets us a little closer to answering the really important and controversial questions of whether there’s anything universal about beauty or meaning in music, and why music evolved—a question that has intrigued scientists since Darwin.”

Sandra Trehub from the University of Toronto notes that Mehr and Singh haven’t quite pinpointed what those universal traits might be. “They’ve found some similarities in the ideas that people have about what a lullaby should sound like,” she says. “Now they need to know what makes something a lullaby.” They’ve made a start. In a rough analysis, the team showed (perhaps unsurprisingly) that dance songs have faster tempos, steadier beats, more instruments, more melodic complexity, and more singers. Lullabies are the opposite.

The study’s main weakness, Trehub says, is that the people who listened to the recordings, though hailing from different cultures, were all internet-savvy Anglophiles. “That means they would have had exposure to Western music and Western ideas about music,” she says. Mehr acknowledges this problem. He and his colleagues are now repeating their study with larger sets of recordings, and a bigger sample of internet users. They’re also going to travel around the world to test people from small-scale societies, “who have mostly only ever heard the music of their own culture,” he says.

Whatever they find, Mehr suspects that the answers will be intriguing. “Maybe there is some prototype in the mind of what a dancy song is, or maybe people have heard a lot of music that’s associated with dance and that just happens to correlate with the 86 cultures we have,” he says. “That would be still interesting. It would be crazy if cultural evolution has turned pop music into something that reflects music in small-scale societies.”