Can it be scientifically demonstrated that music used to be better than it is now? A research group at Tuscia University in Italy carried out a large-scale data study and shows that music has become less complex over the centuries. But there is also criticism. Music scholar Julian Schaap (Erasmus School of History, Culture and Communication) was interviewed by NRC about his sociological perspective on the study.
Melody and harmony
In the research, 20,000 pieces of music converted to MIDI from an existing online database were analysed, covering different genres and time periods. The melody and harmony of these pieces were examined; the focus was on individual notes and how they relate to one another. Using this, the researchers built a kind of musical 鈥榤ap鈥, in which these relationships were visualised. The more dots and lines, the more complex the music. Unsurprisingly, jazz and classical music score highly, while hip hop, electronic, rock and pop score lower.
But what does that say about the quality of the music? According to Julian Schaap, who does not study the music itself but how people respond to it, this kind of research can reveal interesting patterns. 鈥淭he criticism is: you can鈥檛 capture music in numbers and you shouldn鈥檛 want to quantify art. But I don鈥檛 entirely agree with that. As long as you don鈥檛 claim to prove that music has become better or worse. Music is so dear to people; they cling to it as part of their life experience. If you start quantifying that, or start looking for larger patterns, you very quickly tread on people鈥檚 toes.鈥
Limitations
Schaap says the biggest limitation of the study is the MIDI format. 鈥淲hat defines different genres cannot be captured by melody and harmony alone. Rhythm, lyrics, beats and tempo are left out of consideration, but they are hugely important for genres like hip hop or rock. In addition, only music that is popular enough will end up in the dataset. As a result, the study misses a great deal of complex music. In the pop charts you find a lot of uniformity, because certain formulas are popular. But if someone makes something unusual or quirky, it is very unlikely to be included. Just look at the YouTube success of Angine de Poitrine, music that will not easily appear in the charts or MIDI datasets.鈥
The Italian researchers are themselves well aware of the limitations that come with analysing only harmony and melody. 鈥淲e only look at a small part of the music. In our article, we explain very carefully what we can and cannot say,鈥 explains data scientist and art researcher Noccol貌 Di Marco of Tuscia University.
Is complex better?
However cautious Di Marco and his colleagues may have been, many music blogs and news sites still seize on the results as confirmation of the idea that music used to be better. 鈥淭hat鈥檚 actually a strange argument,鈥 says Schaap. 鈥淲hy would complex music always be better? Is an extremely difficult book, which many people find unreadable, better than an accessible one? A normative debate about what is good cannot, by definition, be supported objectively.鈥
- Researcher
- More information
