Music baffled Charles Darwin. Humanity’s ability to produce and enjoy melodies, he wrote in 1874, “must be ranked amongst the most mysterious with which he
is endowed.”

All human societies made music, and yet, for Darwin, it seemed to offer no advantage to our survival. He speculated that music evolved as a way to win over potential mates. Our “half-human ancestors”, as he called them, “aroused each other’s ardent pass-
ions during their courtship and rivalry”.

Other Victorian scientists were sceptical. William James brushed off Darwin’s idea, arguing that music is simply a byproduct of how our minds work — a “mere incidental peculiarity of the nervous system”.

That debate continues to this day. Some researchers are developing new evolutionary explanations for music. Others maintain that music is a cultural invention, like writing, that did not need natural selection to come into existence.

In recent years, scientists have investigated these ideas with big data. They have analysed the acoustic properties of thousands of songs recorded in dozens of cultures. A few days ago, a team of 75 researchers published a more personal investigation of music. For the study, all the researchers sang songs from their own cultures.

The team — which comprised musicologists, psychologists, linguists, evolutionary biologists and professional musicians — recorded songs in 55 languages, including Arabic, Balinese, Basque, Cherokee, Maori, Ukrainian and Yoruba. Across cultures,
the researchers found, songs share certain features not found in speech, suggesting that Darwin might have been right: despite its diversity today, music might have evolved in our distant ancestors.

“It shows us that there may be really something that is universal to all humans that cannot simply be explained by culture,” said Daniela Sammler, a neuroscientist at the Max Planck Institute for Empirical Aesthetics in Frankfurt, Germany, who was not involved in the study.

Databases of songs collected by ethnomusicologists sometimes lack important details. It can also be hard for researchers to make sense of the structure and lyrics of songs from other cultures. Computers, likewise, are not very good at recognising many features of music.

“We thought we should involve the insiders,” said Yuto Ozaki, who earned his doctorate at Keio University in Japan by helping to lead the project.

Ozaki’s colleague, Patrick Savage, took on the job of recruiting the singers. “It was a combination of the network I’d already built up through the first decade of my career along with going to conferences and making small talk and meeting people,” said Savage, now a musicologist at the University of Auckland, Australia.

All the team members picked traditional songs from their cultures for the purpose of the study.

In addition to singing, they recited the lyrics of the songs without a melody so that the team could later compare the music and speech. And for a further point of comparison, the researchers played their songs on a wide range of instruments, including sitars
and melodicas.

In each recording, the researchers measured six features, such as pitch and tempo. Despite their variety, all of the songs shared a number of features that set them apart from speech. The pitch was higher and more stable, for example, and the tempo was slower.

Sammler cautioned that the singers in the new study were mostly academics and that the songs they chose might have introduced some bias into the research. “It’s essentially academics singing material that may not be representative,” she said.

But she also noted that another study, not yet published in a scientific journal, came to a similar conclusion. In that study, researchers analysed songs from 18 languages and pinpointed many of the same features.

It’s possible that songs have distinct features because they have a special role in human communication separate from speech, said Aniruddh Patel, a psychologist at Tufts University, US, who was not involved in the study. What’s more, our brains appear to be sensitive to those features. In 2022, Patel pointed out, researchers discovered human neurons that only responded to singing — not speech or music played on instruments.

“There is something distinctive about songs all around the world as an acoustic signal that perhaps our brains have become attuned to over evolutionary time,” Patel said.

What sort of evolutionary benefit would come from that signal is still a matter
of debate.

“Maybe music was needed to improve group cohesion,” Ozaki said. Singing in choruses, sharing rhythms and melodies could have brought people together whether as a community or in preparation for a battle.

But Sammler didn’t think that the new study ruled out other roles for music, such as helping parents bond with their children. “It could support a lot of theories,” she said.

NYTNS