Research suggests that certain musical meanings may transcend cultural boundaries and become pervasive
The poet and Harvard professor Henry Wadsworth Longfellow famously said, "Music is the universal language of mankind." A new Harvard study suggests he may be right.
The study, conducted by psychology research assistant Samuel Mehr, human evolutionary biology graduate student Manvir Singh, alumni Luke Glowacki and Hunter York, and psychology associate professor Max Krasnow, found that people around the world could recognize lullabies, dance songs, and healing songs — regardless of their cultural origin — and listened only to 14 seconds of clips.
This finding suggests that not only is music deeply rooted in human nature, but that certain types of songs transcend cultural boundaries. The study was described in a Paper published Jan. 25 in Current Biology.
"It seems like everybody makes music in some way," Mel said. "But there is no good empirical evidence of whether the different types of music they make have common characteristics across cultures." One way to test this is to experiment with this naïve listener... The results show that, in some cases, the answer is yes. ”
The findings are based on a wide-impact experiment in which 750 online participants in 60 countries listened to short excerpts of songs collected from nearly 90 small societies around the world, including hunter-gatherers, herders and subsistence farmers.
Participants then answered six questions, rating each clip on a six-point scale based on whether they thought the song was being used to dance, appease babies, cure illness, or express love. Two additional uses — mourning the dead and storytelling — are included as contrasts.
Mehr, a postdoctoral fellow in data science at Harvard's Data Science Program, said the data showed — despite participants' unfamiliarity with the culture, random sampling of each song, and short sample duration — that people were able to reliably infer songs' features, and that their ratings were consistent across the globe.
The findings run counter to the expectations of experts.
Mehr, Glowacki, and Krasnow also surveyed scholars, including ethnomusicologists, music theorists, performers, composers, psychologists, and cognitive scientists, to see if they believe people can identify song genres.
"We gave them an idealized version of the experiment," Mel said. "Imagine you have unlimited time and resources and the ability to record every song from a variety of cultures and play those songs to people around the world.
"The question we asked was if we played these recordings for people, would they be able to tell the difference ... Is it a lullaby or is it for dancing? He continued." Mainly among ethnomusicologists, the answer is no. Not only that, but they also predicted that people's reactions would be inconsistent with each other. This is not what we found. ”
Singh also wondered if listeners recognized certain non-musical features of songs — for example, lullabies are usually sung by a woman, while dance music is more about a group.
"The question at the time was, if people could do that, how exactly would they do it?" Singer said. "How could a man from Tallahassee recognize a dance from a hunter-gatherer tribe in Southeast Asia who knew nothing about the culture of this tribe?"
To test this, the team conducted a second study. This time, they asked the listener about some background and musical characteristics, from the number and gender of the singers to the rhythm and melodic complexity of the song.
"From all of this, we can do a very simple and basic analysis of each song," Mel said. "It turns out that when you ask people these very simple questions about songs, the level of agreement between them is very high. Even on truly subjective musical characteristics, such as melodic complexity, they tend to give each other a consistent rating. ”
When combined with data from the two studies, the results show that songs with the same function have similar characteristics — for example, lullabies tend to be slower and simpler melodies than dance music — suggesting that something of musical characteristics crosses cultural boundaries.
"It seems like everyone is making music in some way. But there is no good empirical evidence as to whether the different types of music they produce have common characteristics across cultures. One way to test this is to experiment with this naïve listener... The results show that, in some cases, the answer is yes. —Samuel Meier
Mehr said the researchers were able to draw wide-ranging conclusions because the songs used in the study came from the record catalog of The History of Natural Songs, a Harvard-based project that created a rigorously constructed database of ethnographic texts about music and recordings of music.
"We bring together all the musical examples in a systematic way, so the inferences drawn from the entire recording can be generalized to humans, not just the culture studied," said Mehr, who directed the project with Singh and Glowacki and is now a researcher at the Toulouse Institute for Advanced Study. "This has always been a problem in music research. Studies that are billed as the study of musical universality often include only a few cultures, or do not systematically sample different types of music in a principled manner. ”
Going forward, the team hopes to conduct a deeper analysis of the music collected for Natural History and conduct more research to improve inferences about the music's ability to cross cultural boundaries.
"One of the weaknesses of this study is that the audience we sampled was people on the internet, so they all had access to things like YouTube, and they might all be familiar with, say, Taylor Swift," Mel said. "Does the result tell us about the design of the human mind, or do we tell us what modern listeners hear in world music?"
To address this, the team is working to translate the studies into more than two dozen languages and conduct online experiments in more countries. Singh and Glowacki are also working to bring this research into the field by playing song clips for small-scale members of society in Indonesia, Ethiopia and elsewhere.
"That's the most exciting part," Mel said. "Because these people have very little exposure to the Internet, radio or Western culture. The only music they know is their own music. We will find out if they share the same musical form and functional concept as our English-speaking internet users. ”
Finally, Meier said, this study and others like it will allow scientists to lay the groundwork for answering many long-term questions about music and its evolution.
"This is one of the most important contributions we hope to make to the field," he said. "This basic, cross-cultural fact-finding of human behavior is the first step in developing a new music science."