PrintWHEN Indian sitar virtuoso Ravi Shankar and his ensemble played at Madison Square Garden, New York, in 1971, the audience broke into rapturous applause at the first short pause. "Thank you," said Shankar. "If you appreciate the tuning so much, I hope you will enjoy the playing more."
When we listen to music from another culture, it's easy to get it badly wrong. Even if our misunderstanding isn't quite as embarrassing as the Shankar faux pas, we are likely to miss most of the nuances and allusions, think it all sounds the same or even dismiss it as a racket. Most 20th-century ethnomusicologists who compared the music of different cultures argued that this was because the way we make music and respond to it is learned, and therefore culture-specific.
Yet there does seem to be something universal about music. All cultures make music, though no one knows why; it's not obviously useful in the way cooking or language are. A number of musicians, including some notable composers, claim that music is a universal form of human communication which transcends barriers of culture and language. Now psychologists are putting this universality back on the agenda, and are investigating whether certain elements of music are hard-wired into the brain.
The issue of universality is significant, because it points to a deeper question. If there are universals in how we perceive music and respond to it, our musical sense might have some adaptive value - evidence for the widely held, but still unproven, belief that music is an evolutionary adaptation, rather than a mere parasite on other auditory processes. As Josh McDermott of the Center for Neural Science at New York University puts it: "Any universal features of music are potentially the result of adaptations enabling musical behaviour."
The ancient Greeks thought music must be universal because it seems rooted in mathematics. Pythagoras pointed out that two notes played simultaneously sound harmonious if the length of the strings that produce them - and hence, in our modern understanding of sound, their acoustic frequencies - are related in simple ratios. Notes an octave apart, for example, have frequencies in the ratio 1:2. Notes spaced by the harmonious interval called a perfect fifth - a C and the G above it, say - have a frequency ratio of 2:3 . Most music theorists maintain that such combinations of notes with simple frequency ratios are in some sense natural and will sound harmonious to everyone, while combinations with a more complex frequency ratio sound jarring and dissonant.
There is a plausible biological explanation for some of this. All natural sounds are composed of many frequencies: a "fundamental", plus overtones that are generally simple multiples of that frequency. The lower overtones are typically the strongest. When you pluck a string, the pitches of the first and third overtones are respectively one and two octaves above the fundamental, while the second overtone is an octave plus a perfect fifth above the fundamental. An auditory system that blends these tones harmoniously into a single note provides a more accurate representation of our sonic environment, and so is likely to be selected for by evolution.
There are also conceivable reasons why dissonant harmonies should sound grating. "Unlike the overtones of harmonic intervals, those of dissonant intervals do not overlap," says Sandra Trehub of the University of Toronto in Mississauga, Ontario, Canada. Interference between these frequencies produces rapid periodic changes in loudness, called acoustic roughness, which induces unpleasant sensations in many listeners. "You'll get the general idea if you imagine the sound of nails scratching on a blackboard," Trehub says.
These neat acoustical theories are all very well, but how do they relate to the music people play and listen to? Unfortunately, the answer is far from clear. Just about every culture bases its musical scales around the octave, and most also use the perfect fifth, but there is at least one that does not use octaves and some that omit fifths. In Indonesian gamelan music, for example, the only simple frequency ratio is the octave; one of its scales simply divides the octave into five equally spaced notes.
That's where music psychologists come in. They have devised all kinds of experiments to figure out if people really do have an innate preference for simple-ratio harmonies over complex ones.
To avoid the influence of learned preferences, psychologists have focused their tests on animals and babies. Animals such as monkeys can reveal their preferences simply by moving away if they don't like what they hear. In 2004, McDermott and Marc Hauser of Harvard University put tamarin monkeys in a V-shaped cage, offering them the choice of sitting in one branch with harmonious, or "consonant", musical chords playing, or the other, with dissonant chords. The monkeys were clearly aware of the sounds, showing a marked preference for low over high volume - but they didn't care whether they heard consonant or dissonant chords (
So if monkeys cannot tell us very much, what about human infants? Though it's not easy to figure out what they prefer, you can at least find out what they notice by following their head movements. Babies show greater attentiveness to novel stimuli, a fact we can exploit to test what grabs their attention in music. For example, you can play them subtly different patterns of notes and see if they spot the changes.
Trehub and her colleague Glenn Schellenberg did just that, testing how well 6 and 9-month-old infants could detect small changes to consonant versus dissonant intervals. They reasoned that if the human brain evolved to prefer consonant intervals, then babies should be more alert to changes in these - and that is what they found. The more consonant the interval, the more reliably the babies could detect the change. They discerned changes to fifths (C plus G) best of all, fourths (C plus F) less well, and tritones (C plus F sharp) least well (
Even this, however, doesn't prove that we are innately disposed to like consonance. What we notice and what we prefer are not necessarily the same; besides, even 8-month-old babies may have learned a musical culture. Babies can hear in the womb during the third trimester and are astonishingly receptive to sensory information. They can recognise their mother's voice virtually from the moment of birth. This complicates all attempts to deduce innate preferences from infant studies. "It's very difficult to fully rule out learning effects," says McDermott.
In an effort to overcome this problem, Nobuo Masataka of Kyoto University in Japan tested a group of 2-day-old babies who were unlikely to have heard much music while in the womb - they were babies with normal hearing born to deaf parents. Masataka tested whether their eye movements showed greater attentiveness to a Mozart minuet or the same minuet modified to introduce dissonance (
Is there a way of answering the question once and for all? "One could conceivably rear infants in an abnormal musical environment to see whether that alters their preferences," says McDermott. "But how many parents would really want to do that to their kids?"
Despite the uncertainties, Trehub thinks that the evidence "is consistent with inherent processing biases and preferences for consonance". McDermott, however, is not persuaded. "I think the idea that there are universal preferences for particular harmonic or melodic intervals reflects a pervasive western-centric bias in the science of music," he says. "It's pretty clear that intervals that are considered dissonant by westerners are sometimes prevalent in other cultures."
While the universality of consonance and dissonance remains open to debate, what of another element of music that might be universally understood? For many people, music is all about emotion. Is our emotional response to music hard-wired, or merely a cultural convention?
In one of the first detailed studies to address this question, Laura-Lee Balkwill and William Forde Thompson, both then at York University in Toronto, found that Canadian listeners were pretty good at identifying whether pieces of Kyrgyzstani, Hindustani and Navajo music were meant to convey joy or sadness (
Similarly, listeners from other musical cultures seem to be able to pick up the emotional intent of western classical music. A team led by Stefan Koelsch of the Max Planck Institute for Human Cognitive and Brain Sciences in Leipzig, Germany, played piano music to people from a remote region of Cameroon, some of whom had never heard western music before. The music had been specially composed to express happiness, sadness and fear in ways that were easily recognised by western listeners. The Cameroonians correctly identified the right emotion significantly more frequently than chance, although less reliably than westerners (
What is it that conveys emotion in music? One possibility is the scale it is played in. To western listeners, the major scale seems so obviously happy and the minor scale sad that it is tempting to imagine everyone will hear them that way. Yet theories to account for these perceived qualities tend to be speculative and based on tenuous evidence.
Take the recent claim by neuroscientist Daniel Bowling at Duke University in Durham, North Carolina, that major and minor scales have intrinsic emotional associations because their sound spectra are akin to those of happy and sad speech (
Tempo is another emotional signpost. Relatively fast, jaunty music tends to be interpreted as happy, while slower music is seen as sad or fearful.
Other musical parameters also furnish emotional cues. Balkwill and Thompson found that complex melodies with large, irregular or unusual changes in pitch tend to be associated with negative emotions, while tunes that seem happy, such as Beethoven's
Music psychologists agree that these associations probably do come from the way music mimics emotional speech and behaviour. Happy people the world over speak moderately loudly, with animated voices and gestures, while sad people speak and move in slower, softer ways. We can judge music's emotional state just as we can often tell when someone speaking an unfamiliar language is joyous, woeful or angry.
However, the emotional quality of music is more complex than just conveying a basic emotion such as happiness and sadness. Most music aims to represent not a single, uniform mood but one that is constantly changing. Many music psychologists believe that the key to picking up on this emotional flux rests on our ability to discern patterns in the notes and rhythms and use them to make predictions about what will come next. When our anticipations are violated, we experience tension; when the expectation is met we have a pleasurable sense of release.
If this model is correct, our emotional response depends on identifying patterns and regularities in the first place. So the question then arises: can we experience this moment-by-moment emotional landscape in the music of unfamiliar cultures?
Music psychologist David Huron of Ohio State University in Columbus and his collaborators have found that we are pretty good at deducing the underlying patterns in unfamiliar music. They played Balinese and American musicians a Balinese tune, and asked them to bet with poker chips on what the next note of the melody would be. Unsurprisingly, the Balinese subjects made better guesses, but the Americans did better than pure chance. It seems the American listeners were basing their predictions mostly on how often they'd heard the different notes already. This suggests we can quite quickly acquire the basic tools that enable the violation-of-expectation model of musical emotion to do its stuff.
So where does this leave the idea of the universality of music? McDermott thinks the only universal properties we are likely to find in the structure of music are those imposed by the limitations of our auditory perception. For instance, he says, "We can't hold too many pitches in mind at once, which constrains the number of notes that can be used in a scale. And intervals also need to be of a certain size in order to be easily heard."
But beyond the notes themselves, there may be something universal in the uses to which music is put. "Every culture that we know of has a genre of music geared towards infants, probably indicative of the importance of music in parent-children interactions," he says. "Music tends to always be used to signify important events within a culture, such as weddings or funerals, and always seems to accompany religious rituals."
Observations like this have led some evolutionary biologists to argue that musicality evolved as a kind of social glue. But to McDermott, music's role might primarily "reflect the deep connection between music and emotion". Just how our brains make that connection is something we are only just beginning to explore.
Philip Ball is a freelance writer based in London. His latest book is The Music Instinct (Bodley Head)
