As well as being one of the oldest, the voice is probably one of the most versatile instruments of the human being. It is used for both speaking and singing, but which of the two activities came first, what characteristics distinguish one from the other, and whether they are related or not, has been the subject of reflection and research for centuries. Just as the reason why singing and music, as well as language, are present in all known human societies is a matter of study.
One of the most accredited hypotheses is that the human ability to produce and appreciate melodies – considered by Charles Darwin, in 1871, “one of the most mysterious” – is the evolutionary result of behaviors advantageous for the survival of the species, useful for example in courtship or before the clashes between human groups. Another hypothesis is that, despite being present in all human societies, singing and music are cultural inventions completely independent of natural selection. And that they lack objective physical characteristics found in all cultures of the world.
A study recently published in the scientific journal Science Advances provided a series of evidence in support of the evolutionary hypothesis, because it identified some specific acoustic properties that in various cultures characterize singing and music compared to speech. Conducted by an international group of 75 researchers from 46 countries on all continents, the study analyzed dozens of traditional popular songs from different cultures around the world: folk songs, lullabies, religious music or other. And he found that some acoustic differences between the sung or played version of each song and spoken word, in terms of timing and frequency, are generally shared across cultures.
The analyzed sample, made up of 300 recordings, was produced directly by the authors of the study: musicologists, psychologists, linguists, neuroscientists and biologists, but also professional musicians, each of whom recorded at least one traditional song from their culture of origin . For example: the neuroscientist from La Sapienza University and the Max Planck Institute for Psycholinguistics Andrea Ravignani, Italian co-author of the study, played “Bella ciao” on the saxophone. The languages represented in the study are 55 in total: in addition to the most widespread and those spoken by very large linguistic communities, such as Basque, the languages of small ethnic groups such as the Rikbaktsa in Brazil and the Ainu in Japan are also included.
The investigation method involved recording a brief description of each song, a sung version, an instrumental version and one in which the researcher pronounces the words of the lyrics. Among the various musical instruments used were clubs, bamboo flutes, and traditional Central and South Asian lutes such as the tar and sitar. In the intentions of the research group, directly involving in the recordings people who are native to the cultures examined is useful to limit the risk of prejudices and errors of interpretation. Prejudices that are instead frequent in cases where – as in many songs recorded in existing databases – musicologists interpret the structure and lyrics of songs from cultures they do not know.
A traditional Brazilian song, “Petara”, sung and played on the bamboo flute by the researcher Tutushamum Puri Teyxokawa (zenodo.org)
Musicologist and psychologist Patrick Savage, co-author of the study and researcher at the University of Auckland, New Zealand, told the New York Times that he recruited the other participants by drawing on a network of contacts that he has developed over the past ten years during conferences and business meetings. Together they focused on the acoustic characteristics of speech and music regardless of differences between languages and musical styles. And they found that in all cultures, with rare exceptions, the sung and instrumental versions of songs tend to be slower, higher pitched and more stable in pitch than spoken words (both the one used to describe the song and the one used to pronounce the words of the text).
It’s unclear why the songs and instrumental melodies analyzed in the study were generally slower and higher-pitched than speech. But according to Savage, music is more predictable and regular than speech because it may have evolved to form and strengthen social bonds between groups of people. “The characteristics we discovered that distinguish music from speech fit well with this theory,” added psychologist Peter Pfordresher in a press release published on the website of the University at Buffalo, which was involved in the study.
– Read also: New musical instruments no longer appear
Pfordresher said that when people make music they tend to make it collectively, and therefore have to synchronize and tune in with each other. Since being in sync becomes more difficult as the tempo increases, when the tempo slows down the pace becomes more predictable and easier to follow. More or less the same goes for tuning, considering that it is much easier to match someone else’s tone if that tone is stable. In speech, however, the need to synchronize is lost, because generally in conversations people alternate with each other. And taking breaks or speaking slowly, according to Pfordresher, could actually provide the false signal that one’s shift is over.
According to Yuto Ozaki, co-author of the study and researcher at Keio University in Tokyo, music could result from signals capable of producing an evolutionary advantage. «Maybe it was necessary to improve the cohesion of the group», he told al New York Timeshypothesizing that singing in a choir and sharing rhythms and melodies could be a way of preparing the people of the same community for battle.
– Read also: Is music really universal?
Daniela Sammler, a neuroscientist at the Max Planck Institute for Empirical Aesthetics in Frankfurt, told the New York Times that the study may have avoided some prejudices but introduced others, given that the singers were largely scholars. “At the end of the day these are academics singing songs that may not be representative,” Sammler said. Furthermore, some of them had significant musical training, such as the Hindustani-speaking Pakistani neuroscientist and classical music singer Shantala Hegde, or the Senegalese percussionist Latyr Sy. Their music may have different characteristics than that produced by a random sample of participants.
Reiterating the need for more studies on this topic, Sammler however pointed out another one in format preprint (i.e. not yet peer-reviewed) which analyzed the singing and speaking of 369 people from 21 different urban and rural societies around the world. From the study, which would confirm the conclusions of the one published above Science Advancesit emerges that singing and music on the one hand and speech on the other differ in the same way in all cultures, on the basis of acoustic characteristics independent of cultural characteristics.
The hypothesis that singing and music have universal non-cultural characteristics does not in any case exclude the possibility of evolutionary origins other than the need for social cohesion before battles between groups. It could also arise, for example, from parents’ need to create a bond with their children, Sammler said, citing a 2022 study on “motherese”, that is, the sounds and sounds we make to communicate with newborns and very young children.
– Read also: The way we talk to newborns is a kind of lingua franca
