We analyze 23,000 reviews from Pitchfork.com to establish a network of over 7,000 artists based on their co-presence in reviews. We use Gephi's (Louvain community detection) modularity class feature to sort the network into closely connected groups. However, because this process is non-deterministic, we improve on its results by running it 10,000 times. Using this novel "metamodularity" method we can show how often any two artists were sorted into the same group. For instance, the jazz musician Alice Coltrane was grouped with John Coltrane 10,000 times, with Sly and the Family Stone 4,939 times, and with Guns n’ Roses one time. With these data, we can visualize how the largely undifferentiated mass of artists fragments into more and more connected groups could be described as genres.

While some of the genres suggested by this method are incredibly specific, one notable exception is the group of African American musicians, which is remarkably large and stable. The artists within it range from Flying Lotus to Motown legend Marvin Gaye to trap rapper Young Thug to R&B singer-songwriter Sade to gangster rapper Tupac Shakur to funk innovators Funkadelic to cross-over hip hop star Cardi B. This is a far more capacious group than most rock clusters along aesthetic, market, and even historic grounds. This may reflect real world connections, yet it is also possible that this metamodularity difference reflects a bias at Pitchfork, where a (historically) rock-oriented hipster aesthetic and a (historically) largely white male audience may correspond to a relative lack of nuance when covering Black artists (historically). In any case, it is fair to say based on these data that Pitchfork has written about Black artists in a structurally unique way over the years relative to their coverage of white artists.