Science of Music
Jonathon Keats 乔纳森·基茨
Some 40，000 years ago， a slender bone flute was abandoned in a Central European cave. Carved with five finger holes and a tapered mouthpiece， the instrument dates from around the dawn of human settlement on the continent.
Humans have been making music for a very long time.
Even that flute is probably a recent example of our musical development. Its sophisticated design suggests knowledge of acoustics， likely drawing on long-standing musical customs. But earlier practices are elusive because the first music was most certainly made with the body and voice， dying with its creators. Charles Darwin considered our musical behaviors to be “amongst the most mysterious.” At least in terms of origins， his words still resonate.
One way of exploring musicality before Stone Age flautists crashed Europe is to study hominid anatomy. Fossils show our australopithecine ancestors had vocal structures akin to gorillas， which lack the ability to carry a tune. But Homo heidelbergensis， likely our last common ancestor with Neanderthals， had vocal physiology very similar to modern humans. Given that H. heidelbergensis evolved at least 500，000 years ago， music may have a 500-millennium history.
Of course， the capacity to make music is not proof that music was actually created. And it certainly doesn't answer the question of why. Was music important enough to drive evolution — offering selective advantages to the most musical？ Or was it just an accessory to other developments， like language？
At one extreme， Harvard University cognitive psychologist Steven Pinker has dismissed music as “auditory cheesecake.” As counterpoint， Oxford University evolutionary psychologist Robin Dunbar has suggested that among primates， singing may have been as important as grooming in fostering social cohesion. Potentially even more efficient than picking lice， this “grooming at a distance” may have facilitated the harmonization of large hominin groups.
Music certainly helps reinforce present-day cliques， especially the auditory cheesecake that dominates teens' playlists. Taylor Swift may be the pinnacle of human evolution. Just don't try to convince Steven Pinker.
Every week since Aug. 4， 1958， Billboard has compiled a list of the 100 most popular songs based on record sales， airplay and， more recently， streaming listens. For Armand Leroi， an evolutionary biologist at Imperial College London， the chart is the cultural equivalent of a fossil record. Leroi and colleagues have enlisted audio analysis techniques to sort a 50-year sample of 17，000 songs into clusters， much as field biologists might group species.
The 13 headers encompass the gamut of musical styles from 1960 to 2010. They're sorted in groups based on similarities in patterns of chord change and tone. Standard genre names such as “country” often appear in more than one category. Leroi's analysis shows that some country songs may be more similar to certain rock tunes than they are to other songs marketed as country.
Listeners who can get caught up in the beautiful sadness of a ballad or the intense anger of death metal know music carries meaning. But linguist Philippe Schlenker of France's National Center for Scientific Research thinks music isn't just a way to convey emotion. By mimicking how we experience sound in everyday life， composers embed extra nuance to help tell their stories.