Powerful computer tools are helping musicologists study the evolution of popular music.
Musical insights from computers. I’m Bob Hirshon and this is Science Update.
Every genre of music has its own chords and tones, from the major chords of pop and rock, to the minor 7ths of disco and funk. In the journal Royal Society Open Science, musicologist Matthias Mauch at Queen Mary University of London and his colleagues report on their work using computer algorithms to listen for these signatures in 17,000 hit songs from the year 1960 to 2010. The big revolutions in music, like rock, disco, and rap, stand out clearly, along with more subtle trends. Mauch says the work is a warmup for a more ambitious goal: analyzing the past century of music worldwide.
We want to actually see how geographical patterns, how genetic patterns, how language patterns influenced the variety of music that we see around the world.
I’m Bob Hirshon, for AAAS, the science society.
Making Sense of the Research
Researchers in this study sought to explore the diversity and revolutions in pop music by tracking the range of sounds in the charts and when new musical styles came to prominence. The scientists engaged in this study included computer scientists and evolutionary biologists from Imperial College London and Queen Mary University of London. Theirs is the most substantial study of the history of popular music to date. How did they go about conducting the study and what did they learn?
Until recently, the ability to do a scientific account of the evolution of popular music was hampered by the lack of data. With the advent of digitized collections of audio recordings, however, scientists have been able to begin quantitative studies of musical evolution. In effect, the scientists involved in this study approached it in much the same way that a paleontologist might and looked at the historical data as a fossil record. They asked questions like: "Has the variety of popular music increased or decreased over time? Is evolutionary change in popular music continuous or discontinuous? And, if it is discontinuous, when did the discontinuities occur?" (Excerpted from "The evolution of popular music: USA 1960-2010")
To carry out the research, they teamed up with the music website Last.fm to access its digital archive of popular music and used signal processing and text mining to help them analyze the musical properties of songs. Signal processing is a technology that aids in the transferring of information contained in many different physical, symbolic, or abstract formats. Text mining is a process of getting high-quality information from text.
The system the researchers set up automatically grouped the thousands of songs by patterns of chord changes and tone. This allowed researchers to statistically identify trends in popular music. Some of the trends the researchers discovered include:
- The so-called "British Invasion" of U.S. pop music by groups such as The Beatles and The Rolling Stones did not start a rock revolution, but only followed existing trends.
- The greatest musical revolution in U.S. pop history was not 1964, but 1991, when hip-hop arrived in the charts.
- 1986 was the least diverse year for the charts.
- Diversity recovered after that, but it was declining again by 2010
Despite the recent apparent decline in diversity, the scientists do not believe there is evidence for a general trend towards homogenization in the charts.
Now try and answer these questions:
- What technological advance enabled the scientists to conduct this type of study?
- How did the scientists approach their study of pop music?
- What website did they team up with to conduct their research?
- Explain signal processing and text mining.
- What were some of the researchers' findings?
You can read more about this study at First evolutionary history of 50 years of music charts using big data analysis of sounds.
The Science Update, Computer Composer, looks at how computers have been used in making music for decades.
In the Science Update, Math and Music History, researchers look at how the history of Western music can be described mathematically.
You can use this Science Update and your students' interest in popular music to help engage them in lessons on data analysis and how scientists can use data to help them gain new insight on various issues.
In Perfect Pitch, students can learn how a Web-based, perfect pitch test reveals new details about musical ability.
Computer Composer helps students explore a computer program that can generate an infinite amount of original music.
What's in a Graph?
First Class First? Using Data to Explore the Tragedy of the Titanic
9-12 | Website
6-12 | Audio
Stone Age Tunes
6-12 | Audio