The universality of music

A new study prompts us to ask what in music deserves an evolutionary or physiological explanation, and what is better explained by the individual cultures we live in

December 24, 2019
Slade performing live in concert in 1985. Photo: Hardy Schiffler/DPA/PA Images
Slade performing live in concert in 1985. Photo: Hardy Schiffler/DPA/PA Images

Christmas, as everyone knows, officially begins the first time—often now in mid-November—you hear the ancient refrain: “So here it is, Merry Christmas…”

It may be guaranteed this year that plenty of British people of one political persuasion or another (maybe all) will certainly not be having fun at all. But you might still muster up the spirit to sing a few carols, because Christmas is arguably the most musically codified of all western festivals. It is a demonstration of one of music’s universal functions as an essential aspect of ritual.

That music possesses a global universality is often asserted (“Music is a universal language”), yet proving it is challenging. The predicate is rather ill-defined in any case. Yes, every culture that we know of has some kind of music—but do these diverse musics share anything in common, for example in terms of function, affect and content? An extensive ethnographic study published in Science in November presented the case that they do.

Exercises like this are nothing new, but the latest study, conducted by an international team led by data scientist Samuel Mehr of Harvard University, is notable for both the breadth of its cultural coverage and the depth of the analysis that looked for shared properties. Using two databases—audio recordings of songs from 315 societies, and ethnographic texts from 60 that document the uses of songs in that culture—the researchers were able to mine rich cross-cultural resources to compare both the musical characteristics and the social contexts and functions of music. They found that the variations are greater within than between societies—for example, hip hop might be more different from western sacred music than the latter is from Tibetan sacred music. The study found that most music has a “tonal centre”—in western terms, a key, or a sense of a “home note” that the melody will return to—on which naïve listeners can generally agree. And the acoustic features of music—the pitches and tempos, say—are rather systematically linked to the emotional goals and responses of the performers and audiences across cultures: “sad” music, say, tends always to be softer and slower.

The team found that musical behaviour can be attributed three “dimensions”: degree of formality (from ceremonies to spontaneous singing or intimate infant care), degree of arousal (from calm and meditative to excited), and religiosity (from shamanic or sacred rites to communal sing-alongs). Songs are somewhat clustered in this “musical space”: across many cultures, for example, dance songs are high in arousal and formality, while lullabies are low in both. Western listeners proved pretty adept at identifying these categories: guessing, say, whether a song from an unfamiliar culture is used for dancing or as a lullaby.

Some of these conclusions echo previous work. For example, a study in 2009 found that people from a remote tribe in the Cameroon isolated from western music interpreted the “emotions” of such music—whether it was happy or sad, say—in ways that were consistent with the judgments of western listeners more often than by chance alone. It’s not fully understood why this consistency exists, but a very plausible explanation invokes a kind of pathetic fallacy: we instinctively read into music the emotional features we associate with people. Sad people move and speak slowly, angry people have more harsh, jagged intonation and rhythm.

Some universality in music, meanwhile, seems to be determined by our physiological and cognitive limitations for producing and processing sound. We divide up a continuum of musical pitch into discrete steps—scales—to avoid ambiguities about which pitch is being sung and to create an agreed, communicable “alphabet” for music-making. These scales or modes typically have around four to 12 notes per octave. If there were more, it gets hard to tell notes apart (and to sing them accurately), and to process the information they convey. If there were fewer, melodies would be too monotonous. The use of the octave itself as an organising principle seems to feature in nearly all cultures (a possible, so far unique exception was reported recently in the Tismané people of the Bolivian Amazon), probably because it is a salient property of natural acoustics. Most sounds contain prominent octave “overtones” above the fundamental acoustic frequency, so our auditory system may have evolved to award them a quality of “sameness” to help them blend in our awareness.

By the same token, the finding of the new study that the preferred melodies and rhythms across cultures tend to find a sweet point between monotony and chaos probably also reflects the constraints of our brain’s information-processing capacity: we want stimulation but not confusion. Of course there is plenty of individual variation—some folk prefer the predictable tunes of country’n’western, others crave Ornette Coleman’s freeform improvisation—but most of us get tired of nursery rhymes and will be challenged by the rapid-fire note-blizzards of New Complexity.

It has long been noted too that music in many cultures is associated with particular activities: with sacred ritual, dance, child-rearing and group bonding for example. All these uses have been invoked in theories of music’s origin. Some believe it shares a root with language, and that a primal “musilanguage” became specialised into language for conveying semantic meaning and music as a vehicle of emotion. If so, some vestige of the emotive function might remain in language in how the rise and fall, or “prosody,” of speech signals emotional valence. (Its absence in text messaging, where the need for brevity reduces the ability to compensate with words, necessitates emojis.) Charles Darwin thought that music acted as a vehicle for sexual selection: it showed off your dexterity, stamina or imagination, qualities that a prospective mate might desire. But evidence for that has failed to get much beyond anecdotes of rock-star sex gods and goddesses. Assessing theories of music’s origins is hampered by a lack of empirical evidence to test them, as well as by emotive convictions about the value of music—can it really be as important is it feels, if we can’t prove it once had some evolutionarily adaptive function? The new study seems unlikely to resolve those arguments, but it might well fuel them.

Regardless of what they might say about origins, the existence of some shared characteristics of music across cultures is widely accepted. So it’s surprising to see the authors of the new study claim that “the common view among music scholars today is summarised by the ethnomusicologist George List [in 1971]: ‘The only universal aspect of music seems to be that most people make it.’” They support that assertion with data, however. “In response to the question ‘Do you think that music is mostly shaped by culture, or do you think that music is mostly shaped by a universal human nature?’”, they say, “the majority of music scholars responded in the “’Music is mostly shaped by culture’ half of the scale (ethnomusicologists, 71 per cent; music theorists, 68 per cent; other musical disciplines, 62 per cent).”

Of course, that question has plenty of scope for interpretation. And it’s certainly true that culture can create some significant divergences in how we respond to and understand music, such as when a doting crowd at Madison Square Garden in 1971 rapturously applauded Ravi Shankar, on one of his first US visits, after he had merely been tuning up. Affective misinterpretation was not uncommon either. In one set of tests, western listeners mistakenly attributed a mournful quality to Javanese music that the performers themselves intended as joyous, because it happened to use a scale close to the western minor. Even the popular notion that major and minor scales are somehow fundamentally happy and sad seems to arise from nothing more than acculturation. The same is true of the associations we have with the “rightness” of intervals: the supposed dissonance of the “devil’s interval” of a tritone (C-F# for example) owes more to its long association with the demonic, from Liszt and Saint-Saens to Black Sabbath, than to its acoustic properties.

But anyone paying attention to the field of music cognition in the past few decades should have little reason to doubt that there are universal characteristics in what we might call humankind’s musicality: the biological and social endowment that we have for creating, perceiving and using music. Part of the value of the new study lies with its potential to clear away some of this confusion: to clarify what in music demands a general, perhaps physiological or evolutionary explanation, and what is specific to a given culture. Perhaps some of the apparent discrepancy between the “universality” revealed here and the cultural specificity ascribed by many of the experts polled at the outset lies with the power of music itself. The features that seem to be universal might be called low-level: an ability to hear music as such at all, to decode the complex acoustic signal into notes, to gauge a general mood, as well as general social functions. What often most moves and excites us, however, are details that depend on a lifetime of acculturation and experience: a sudden change in key, violation of learned expectation, the way an exquisite lyric was sung, a poignant recollection of the context in which we first heard the track.

The new study by Mehr and colleagues might also help to challenge cultural bias and prejudice—it has long been apparent to any observant musicologist that, say, Indian classical music and gamelan are every bit as sophisticated and complex as western art music. This leveling of judgment—of social function at least, if not “quality”—needs to happen within western music too. If we see that the dance imperative of techno is no less a part of music’s universal heritage than Giselle, or that the theatrics of Whitney Houston might serve a similar role to that in Senegalese griot tradition as a socially sanctioned delegation of emotion, or even that Slade’s tired Christmas anthem has a ritualistic purpose after all, we can admit that it might be not just permissible but important to include them in the discussion. (No one says we have to like them.)

In that way, this impressive cross-cultural comparative study continues the journey of anthropology since its emergence as a discipline in the late 19th century. No longer is there an insistence on the otherness (and its implication of inferiority) of non-western cultures. In our spiritual, oral, mythological, supernatural and musical traditions and beliefs, humankind has many streams, all flowing through the same broad channel.