How is music changing over time? On one scale, it’s easy to see how music evolved through the ages as baroque music was displaced by classical, romantic and then modern music. In the 20th century alone, jazz, country and western, rock’n’roll all emerged along with pop, rap, hip hop and so on.
Much of this rapid rate of change was driven by new ways of composing, experiencing and sharing music via vinyl, cassettes, CDs and mp3s. The emergence of digital streaming platforms, social networks and composing technologies like GarageBand have accelerated these changes by breaking down barriers and enabling new forms of musical expertise.
So an important question is how this new cultural landscape is changing the nature of music. And today we get an answer thanks to the work of Niccolò Di Marco at Sapienza University of Rome, and colleagues, who have developed a way to think about musical compositions as networks and then to use the tools of network science to study how it is changing over time. They say this network approach clearly distinguishes between musical genres and that it reveals clear changes in music composition over timescales stretching from months to centuries.
Soundscape Network
Revealing the network properties of a piece of music is straightforward. The trick is to think of each note as a node on the network. These nodes are connected by an edge if they appear consecutively in the composition. The edges become thicker depending on the number of times one note transitions into another.
This approach immediately reveals patterns associated with distinctive styles. For example, classical music often contains complex sequences of notes that repeat throughout the composition. The network structure then clearly represents these themes. But pop music by comparison tends to consist of simpler sequences repeated far more often. This leads to smaller structures within the network that are more tightly linked. Jazz has a more diffuse structure of links, and so on.
A key advantage of this approach is that scientists have developed highly sophisticated mathematical tools to study the properties of networks. These have come from areas as diverse as the study of the spread of disease, of the structure of the internet and of food networks in ecosystems. Di Marco and co have been able to bring all this to bear on the structure of music.
The team analyzed approximately 20,000 MIDI files spanning six genres of music from the last four centuries. This expansive dataset provided a broad base to compare genres such as classical, jazz, rock, and pop, across different time periods.
Notable differences immediately became apparent. “Our results show that classical and jazz compositions have higher complexity and melodic diversity than recently developed genres,” say Di Marco and co.
Simple Minds
But curiously, the network analysis reveals that these forms of music are becoming simpler. “A temporal analysis reveals a trend toward simplification, with even classical and jazz nearing the complexity levels of modern genres,” they say.
This simplification trend raises intriguing questions about the factors driving the change. The researchers speculate that technological advancements, changes in listening habits promoted by digital music platforms, and shifts in cultural production and consumption are an important influence. “Our study highlights that the democratization of the composition process and the advent of new technologies and platforms have fostered the development of genres characterized by reduced complexity relative to earlier eras,” they say.
This kind of change is not unique to music. Di Marco and co point out that similar simplification trends have been observed for other forms of cultural expression, such as language and literature. “The observed trend of musical simplification reflects broader societal changes, including the influence of global interconnectedness, rapid content dissemination, and the algorithmic curation of music consumption.”
Ref: Decoding Musical Evolution Through Network Science : arxiv.org/abs/2501.07557
