For those who can’t make it to my session, you can find some materials for my SMT 2017 talk below. You can find the full program here.
“Genre is dead!” The sentiment resounds throughout current popular-music critic-fan and artist discourses, as developments like predictive algorithms, professional playlist curators, and ubiquitous access all throw wrenches into traditional machines of musical categorization. Recent sociological work on the increasing eclecticism of musical tastes appears to support this perspective, flattening conventional boundaries between kinds of music and classes of people.
But in this talk, I argue that such omnivorousness of musical proclivities doesn’t obviate popular music genres; instead, it hints at a deeper shift in genre ordering. To address this change, I explicitly theorize the work genre does in the smooth and striated spaces of popular music with a new concept I call “#genre.” Essentially, #genre captures the adjectival quality and in-between-ness of the seemingly-flattened stylistic world of popular music categories by exploring clusters of related artists, genre tags, and playlist constituency.
My methodology approaches this topological change by excavating the kinds of linear genre-fabric that Spotify weaves through its platform, investigating relational algorithms and proprietary metadata. To do so, I use original Python scripts to access and parse Spotify’s metadata, quantitatively assessing the various kinds of connections that the streaming service generates. I compare these results to demographic biases to problematize notions of a “post-genre” musical landscape, nudging genre discourses away from conventional phylogenetic cartographies or nested hierarchies and towards lateral and multiple models. My methodology and conception of #genre show how classification continues to guide all parts of the 21st-century popular music machine, demanding a renewed investigation.