Conference Agenda

The Nostalgic Sentence is a specific AAAB formal structure that originated in late 19^th-century instrumental music and has since appeared in music associated with numerous nostalgic traditions such as sporting events, amusement parks, and Broadway. It shares characteristics with Tin Pan Alley’s ABAC form, Callahan’s large-scale musico-poetic sentences (2013), Baileyshea’s “Sentence with a Dissolving Third Statement” (2004) and Richards’ “Trifold Sentence” (2011). However, the author argues that the Nostalgic Sentence warrants its own category due to its broader formal level, its signature harmonic motions, and its unique combination of sentential elements (repetition and continuation) with periodic elements (a “restart” at the halfway point). This paper seeks to establish the Nostalgic Sentence’s AAAB alongside ABAC and AABA as a standard historical form in American music, highlight some of its possible text/music relationships through sample analyses, and suggest directions for further research.

Rock music’s influence on NES chiptunes is well-documented. The NES’s sound chip allowed composers to imitate the traditional four-piece rock band through its two pulse waves, triangle wave, and its noise channel. Composers often included what resembled guitar hammer-ons and pull-offs, galloping bass grooves, double-kicking bass patterns, rhythmic cycles, and virtuosic instrument solos in their music (O’Hara 2018). By coupling these with chord progressions, cadential functions, and diatonic modes commonly used in hard rock and heavy metal music, composers created the illusion of rock bands accompanying players on their virtual journeys (Mitchell 2022, Rossetti 2022, O’Hara 2018, Schartmann 2018). Formally, however, these chiptunes pose analytic problems: the absence of lyrics has thus far prevented widespread adoption of rock form terminology like intro, verse, pre-chorus, chorus, causing many to favor simple sectional labels (ABC etc.). Moreover, with the abundance of fan arrangements of—some with added lyrics—it is evident that some chiptunes lend themselves to various rocks forms better than others. The question thus arises: what causes us to hear traditionally rock-based forms within chiptunes?

Here, we address this question by conducting a mini-corpus study of over 100 rock- and pop-based NES chiptunes from 1986-1993 from over 15 different games. Rooted in contemporary theories surrounding rock music’s unique formal and harmonic syntax (Doll 2021, Nobile 2020, de Clerq 2017, Covach 2005), we address the music’s form through their purely sonic (i.e., harmonic, melodic, rhythmic, and timbral) traits, suggesting that even though the music does not contain lyrics, a clear rock form can be heard through their adherence to harmonic and melodic properties common to rock music. We also argue that ambiguity surrounding their form may arise due to their adherence to endless looping techniques coupled with their lack of lyrical text. We thus organize what we call “chip-rock form” into five distinct categories, the last two of which involve ambiguity and thus demand extra scrutiny: 1) verse-chorus, 2) chorus-verse, 3) verse-prechorus-chorus, 4) chorus-verse-prechorus, and 5) ambiguous forms. In this paper, we provide some of our findings regarding different chip-rock and their adherence to these traditional and ambiguous rock forms.

Abstract:

In the last weeks of 2022, news media buzzed over the release of ChatGPT, a text-generating AI. The buzz swirled around a burning question: how will ChatGPT influence writing, a creative human process? Some authors have since responded, claiming that the way forward requires us––developers, users, and consumers––to decide how large a role AI may play in the production and critique of writing, and what risks or benefits may follow. This call to action is not limited to text. My talk extends it to music through study of an AI called Jukebox, which generates musical audio (Dhariwal et al. 2020). Like the authors writing about ChatGPT, I ask how Jukebox and similar AIs have influenced the practice of music composition. I claim that these AIs enable their users to write music through acts of human-machine conversation and musical curation. I then demonstrate how these machine-mediated acts of conversation and curation might lead to interpretive lenses for music analysis. Finally, I raise open-ended questions about what it means to create, study, and engage with music that is co-created with deep, generative AI.