Examining the Changing Discourse of Recorded Popular Music using a Corpus-Based Approach

In this paper I present a corpus-based approach to the analysis of discourses about recorded music. The study is based on an investigation of the articles of the American Billboard magazine across eight decades from 1942 to 2012. This corpus of digitized texts is analysed using Wordsmith Tools, which aids in finding otherwise hidden, or abstruse patterns in texts, such as keywords or clusters of recurrent themes. In this way, I provide linguistic data that shows general historical trends in, and changes to, how music journalists conceptualize the aesthetics of recorded music, its meanings and values, across different decades.

Corpus-based approaches to discourse analyses are fairly new, and are made possible by the increasing number of digital books and magazines available. While a ‘traditional’ discourse analysis is usually focussed on minor collections of texts, corpus-based studies work with large data sets that can provide quantitative verifications for qualitative questions. Changes to how music critics describe the quality or aesthetics of musical sound at different points in time may, for instance, show up through the appearance of new terms, the disappearance of others, or a change in the frequency rate for certain terms or phrases. This paper shows some of the benefits of corpus studies for musicological enquiries.

This study takes a broad historical perspective on how the sound of recorded music is conceptualized, but it will also, more specifically, take into account how the “Nordic sound” is expressed in reviews of Nordic artists or artists otherwise inspired by “the Nordic”. I work from the assumption that we make sense of musical sound through the construction and deployment of certain aesthetic categories. I show how these categories are negotiated in relation to new technologies, artists and genres.