The Distinctive Signal Found in All Music Across the World

Latyr Sy, a Senegalese percussionist and co-author of the study, performing a traditional song. (image credit: Max Planck Institute for Psycholinguistics)

Post Author - Erik Wells

“You’ve got to make your own kind of music”... “If you want to sing out, sing out”... “Music here, music there, music music everywhere!” From the Mamas and the Papas to Jake Gyllenhaal, artists across time, space, and genre have used their music to rhapsodize about the importance of making music. But why do humans make music? Darwin called it one of our most “mysterious” abilities developed through natural selection, while pioneering psychologist William James dismissed it as a “mere incidental peculiarity of our nervous system.” In recent years, a number of theories have cropped up that categorize music as either an evolutionary or cultural development. But a new study has helped to synthesize the two schools of thought by compiling recordings from 75 researchers in 55 languages ranging from Arabic to Yoruba.

Previous studies have allowed ethnomusicologists to theorize about the role of music in social bonding and attracting mates, but while those studies analyzed the mechanisms used to produce music and speech, this study instead examined the acoustic attributes of those verbalizations. Another attribute that set this study apart was that the authors were all participants, meaning that they each recorded a song in their chosen language and manually annotated it. This approach helped account for subtle cultural differences in music structure that an algorithm would miss. 

A word cloud showing the most common words across the 75 songs recorded for the study

Each scientist/performer chose a traditional song from their culture’s repertoire and submitted four recordings: a sung performance of the song, a spoken recitation of the lyrics, a spoken description of the song’s subject matter, and a purely instrumental performance of the song. Across the data set, singing consistently produces a higher and more stable pitch and a slower tempo than speech. This study provides the strongest evidence so far that there are statistically universal differences between speech and song, suggesting that music contains a distinctive acoustic signal that our brain has become attuned to over time. In other words, while some aspects of music vary from culture to culture, there are some basic elements that developed universally over time.

What conclusions can we draw from this? On the whole, song was shown to be more consistent and predictable in its rhythms than speech, which suggests that it developed as a way to facilitate social bonding, synchronization, and memorization. Just think of how we use songs to teach children about the alphabet, states, and more. This theory coincides with the findings of a 2022 study, which explored the universality of “parentese” (aka baby talk). That study shows that the more musical rhythms of parentese allow babies to create an association between sounds and mouth shapes, which facilitates memory and bonding.

Evolutionarily, music’s ability to create a social bond may have been less important than its ability to signal a bond to outsiders. The appearance of group cohesion, developed through shared songs, might have aided in fending off attacks. While we still cannot say for certain, we do know that even the act of studying music can improve our social cohesion across language barriers.  Some languages featured in the study, such as Ngarigu (an Australian Aboriginal dialect) and Aynu (a Western Chinese mixed language featuring Turkic grammar and Iranian vocabulary), are near extinction. Including them in the study may shine new light on the languages and aid in their linguistic and musical revival efforts.

Works Cited

Hilton, C.B., Moser, C.J., Bertolo, M. et al. Acoustic regularities in infant-directed speech and song across cultures. Nat Hum Behav 6, 1545–1556 (2022).

Sammler, Daniela. Signatures of speech and song: “Universal” links despite cultural diversity, Science Advances, 10, 20, (2024).