Switch Up the Groove: Idiosyncratic Approaches to Form and Texture in Recent Popular Music^*

Introduction

[0.1] Much of Billie Eilish’s chart-topping single “bad guy” (2019) feels like a dark-comedy parody of mainstream synth-pop songs from the late-2010s. With her murmuring voice floating above a driving, four-on-the-floor kick drum pattern and an invariant i–iv–V⁷ chord progression, Eilish mocks the “tough guy” charade of her romantic interest and, with characteristic ironic flair, asserts that she is the real “bad guy” in their relationship. The music video underscores this message using a variety of whimsical and absurd images, such as adults riding children’s bikes, a bowl of cereal being made in a man’s mouth, and animate human heads suspended in goldfish bags. This raucous tone is maintained through two verse–prechorus–chorus rotations, so when an additional chorus is included after the second rotation (2:14–2:28), with additional backing vocals to thicken the texture, a first-time listener might justifiably hear the ensuing three seconds of silence as the rather premature conclusion of “bad guy” (Example 1).⁽¹⁾ However, the silence is abruptly broken by an 808 bass grinding out a tonic pedal at less than half the tempo of the preceding music—a rhythmic and timbral alteration that, for one reviewer, proves drastic enough to shift the tone of “bad guy” from “upbeat electropop” to “a surrealist, artistic nightmare” (Gilke 2019). The music video parallels this transformation by depicting Eilish in a dimly lit room, sitting on the back of a shirtless man who is doing pushups (Example 2).⁽²⁾ The textural contrast of this final section emphasizes the song’s message by contributing a greater degree of intensity to her self-identification as the “bad guy” and serves as a foil to the rambunctious light-heartedness of the preceding music.

[0.2] Although the expressive purpose of this section is quite apparent, how might we understand its function within the formal structure of “bad guy”? I offer one interpretation in the form graph of Example 1, where I describe the final core section as a having blended “coda-verse” function, followed by a brief outro.⁽³⁾ Below the graph, I indicate the primary musical features that distinguish this formal group from the preceding sections of the song.⁽⁴⁾ In this interpretation, the compound AAB form of “bad guy,” shown at the top of my graph, might be heard as a modification of what Brad Osborn has identified as the standard “compound AAA” form of Top-40 songs influenced by electronic dance music (EDM) (2023, 48). In another interpretation, this last section might serve the contrasting function of a bridge, thus suggesting that “bad guy” is instead a truncated version of what John Covach describes as the “compound AABA” form typical of classic pop-rock music of the twentieth century (2005, 74).⁽⁵⁾ Regardless, what both of these readings share is a clear and significant formal juncture at the point of textural disruption, indicated by a dotted line in Example 1. In this article, I consider a range of popular songs written since 2015 that draw on EDM and hip hop, with the aim of uncovering the formal and expressive effects of such dramatic textural transformations.⁽⁶⁾ Building on Kofi Agawu’s (1991) “beginning–middle–end” paradigm for the “temporal functions” of formal groups, I show that the juxtaposition of starkly contrasting musical textures—or “grooves” (Butler 2006)—can play a crucial role in how we perceive large-scale temporal functions in popular music. I further demonstrate that these textural disruptions can catalyze what Janet Schmalfeldt (2011) describes as a “retrospective reinterpretation” of the formal or temporal function of some section(s) of a song. The purpose of this article is not simply to develop an analytical methodology for one of popular music’s idiosyncratic formal devices, but to illustrate how formal conventions, production technology, and an artist’s expressive aims mutually inform and mediate one another—and to outline how these valences have coalesced in a distinctive twenty-first-century poetics of form.

1. Temporal Functions in Popular Form

[1.1] Recent scholarship has shown that the formal trajectories of 21st-century EDM-influenced popular music operate according to a different set of principles than those of classic pop and rock music from the twentieth century.⁽⁷⁾ Building on Mark Butler’s influential book Unlocking the Groove (2006), much of this scholarship focuses on the degree of “sonic energy” that inheres to particular formal sections and how these energy trajectories serve to create musical momentum and cohesiveness throughout larger formal assemblages, such as a verse–prechorus–chorus rotation. Agawu’s beginning–middle–end formal paradigm (1991) is central to many of these accounts, where it serves as the basis for conceptualizing how we perceive the distinct temporal function of a given formal unit, its relationship to other units, and its participation within broader, directed formal processes.⁽⁸⁾ For Agawu, the beginning–middle–end paradigm not only reflects the listener’s “specific attitudes to a work’s beginning, its middle, and its ending,” but also provides suggestive clues as to a work’s “dramatic character” (51)—a notion that is central to the analyses in this article. Several theorists have adapted Agawu’s formal paradigm to describe energy trajectories in EDM-influenced popular music, resulting in a wide array of terminological systems, such as Nobile’s initiation–buildup–arrival functions (2022, [1.2]) or the verse–riser–drop functions used by both Alyssa Barna (2020) and Brad Osborn (2023). Most of these accounts are primarily concerned with mapping versions of the beginning–middle–end paradigm onto the individual sections of a single formal rotation—an approach that has yielded particularly valuable insights into how these sections cohere as distinctive groups in the absence of large-scale harmonic processes.⁽⁹⁾

[1.2] This prior research on EDM-influenced formal functions informs my analyses in this article, where I deploy temporal functions to describe the relationships between formal groups in several different popular songs. However, unlike many of these studies, my focus is on the insights afforded by applying the beginning–middle–end paradigm at the higher structural level of a song’s rotations, as seen in the example of “bad guy.” It is therefore important to differentiate between the temporal functions that I apply to large-scale formal groups and what are typically called the formal functions of popular music, which refer to the distinct functions of various recognizable formal sections; for instance, one often finds references to concepts like the “prechorus function” (Nobile 2022, [3.7]), or “chorus function” (Endrinal 2011, [18]) in this scholarship. In the formally idiosyncratic songs I consider, this large-scale approach provides a heuristic for the intersection of narrative and form, as well as an opportunity to investigate how contemporary approaches to songwriting have been influenced by the style conventions of EDM and hip hop.

[1.3] An important precedent for my large-scale approach to formal structure is Brad Osborn’s “Formal Functions and Rotations in Top-40 EDM,” in which he draws upon Covach 2005 and Hepokoski 2010 to develop the concept of “repetitive circularity” in popular song form (2023, 47).⁽¹⁰⁾ Osborn maintains that large-scale formal repetitions are central to how we perceive and impart meaning to musical structure and suggests that “‘repetitive circularity’ does not necessarily depend on the presence of a contrasting middle, such as the ‘B’ in a compound AABA or the development of a sonata” (47–48). Referencing a corpus of recent EDM-influenced songs, Osborn further concludes that songs written in this genre “almost never contain a viable contrasting bridge, resulting in a truly circular structure,” which he labels “compound AAA” form (48).⁽¹¹⁾ This represents a significant departure from the formal conventions of twentieth-century pop-rock music, which more often follows a compound AABA structure. In Example 3, I illustrate how the beginning–middle–end paradigm maps onto these two common large-scale structures, from which we can identify two distinctions between the formal types:⁽¹²⁾

1. There is a one-to-one correlation between temporal function and rotational unit in compound AAA form, whereas the beginning group in compound AABA form corresponds to both initial A rotations.⁽¹³⁾
2. In compound AAA form, each temporal group comprises the same thematic material (though the final rotation often omits the verse section), whereas in compound AABA form, the middle group introduces new, contrasting material.

[1.4] The absence of a contrasting B section in compound AAA form means that each A rotation fulfills a distinct temporal function within the song, but it also creates a situation in which there is no inherently contrasting material and, as Osborn notes, producers must therefore “attempt to balance sonic contrast with thematic memorability: too little sonic contrast and the song becomes repetitive; too much thematic diversity risks a lack of memorability” (2023, 48). Osborn discusses some common techniques for introducing musical contrast in compound AAA form, all of which arise from combining “previously heard thematic material [...] with the sonic characteristics of a bridge function” (48)—a possibility that is afforded by what several scholars have identified as the “instability” or “ambiguity” of formal sections in post-millennial popular music (Adams 2019; Bungert 2019; de Clercq 2017). As Kyle Adams explains, the influence of EDM on popular music has created a situation in which formal functions are typically expressed through characteristics of musical texture (e.g., the thickening texture of a prechorus that leads to the “drop” of a chorus), meaning that a particular span of music can therefore “project different formal functions from one part of a song to another, depending only on changes in texture, and does not require a consistent formal label throughout” (2019, 34). For the purposes of this article, the key takeaway from this discussion is that texture is often the defining feature of formal function in EDM-influenced popular music (and not lyrics, melody, harmonic structure, etc.), and it is consequently the primary means through which musical contrast is incorporated into an otherwise repetitive formal design. This appears to be the phenomenon that Mark Butler has in mind when he says that the “groove—far from being locked in to a single, restricted type—promotes multiple interpretations and flexible interactions, an unlocking of temporal experience into many possible directions” (2006, 6).

[1.5] In “bad guy” and many of the other songs I consider in this article, groove transformations have a dramatic effect on how we retroactively interpret the temporal function of a section or rotation in a song’s larger form—a phenomenon similar to what Janet Schmalfeldt describes in the context of common practice tonal music as “the special case whereby the formal function initially suggested by a musical idea, phrase, or section invites retrospective reinterpretation within the larger formal context” (2011, 9; original emphasis). Grounding her discussion in German Romanticism and the formal theory of Carl Dahlhaus, Schmalfeldt develops a processual model for analyzing formal functions that allows a single passage of music to express multiple formal functions, depending on the listener’s temporal perspective. She introduces a “becomes” operator (⇒) to describe instances of formal ambiguity, such as the much-discussed opening section of Beethoven’s “Tempest” Sonata, which she labels as Introduction ⇒ Main Theme to reflect how a passage that we initially perceived to be a slow introduction to the sonata will likely be reinterpreted as its main theme when the same material returns at the outset of the recapitulation. I implement Schmalfeldt’s “becomes” operator in several of the following analyses, where I find that it is particularly effective for representing how a given span of music can, following a textural transformation, be reinterpreted as having a different formal function or be grouped with a different large-scale temporal function. The specific effect of a transformed groove will vary based on the formal structure of the song in question, but my broader claim is that groove not only affects our perception of a song’s large-scale temporal functions, but also plays a significant role in how we select and demarcate formal boundaries. I develop this position further in the following sections by analyzing several songs that use groove transformations as a means of correlating distinct expressive characteristics—or “dramatic characters” in Agawu’s words—with specific rotations and large-scale temporal functions.

2. Disrupting the Groove: Beat Switching and Compound AAB

[2.1] Having now identified some of the distinct formal and textural characteristics of EDM-influenced popular music, we can return to the example of “bad guy” and consider how its unusual formal structure relates to the broader stylistic practices of contemporary pop. Example 4 reproduces the earlier form graph for “bad guy,” along with an abstracted version of its formal structure that I call compound AAB. The basic features of compound AAB form, as I depict it here, are two complete verse–prechorus–chorus rotations and a final “B” part consisting of a coda-verse with temporal dimensions that are approximately equivalent to a complete A rotation.⁽¹⁴⁾ As we will see in the following examples, the specific arrangement of formal sections that constitutes each of the large-scale groups will vary from song to song, particularly in response to generic conventions. This means that the final B part, for example, may comprise a core section with a blended coda-verse function (the most common), a simple verse function, or something that is more functionally ambiguous.⁽¹⁵⁾ With the exception of the last song discussed in this article, every example conforms to a large-scale ternary structure in which the initial A₁ rotation serves the temporal function of the beginning of the song, the A₂ rotation functions as the middle, and the final B section functions as the end. Despite having a tripartite structure like compound AAA form, the use of contrasting material at the end of the song creates a formal boundary between A₂ and B that is strong enough to suggest a lower-level grouping in which the two A rotations are related as the beginning and end of the first group, depicted below the graph in Example 4.

[2.2] Although songs that end with contrasting material have been acknowledged in scholarship since Walter Everett’s The Foundations of Rock (2009), most of these discussions take compound AABA as the primary point of reference. For example, Everett discusses three songs in which the bridge is not followed by a recapitulation of A material, creating a compound AAB form (2009, 149): “Soul Man” (Sam and Dave, 1967), “I Never Loved a Man (The Way I Love You)” (Aretha Franklin, 1967) and “Think” (Aretha Franklin, 1968). Responding to Everett, Osborn (2013, footnote 12) questions “whether we can distinguish between a bridge and a coda without recourse to their location within a song. Endings such as these may also be described as anti-climaxes and also occur in some modern rock songs, including Tool’s ‘Prison Sex’ (1993), Alkine Trio’s ‘Sundials’ (1997), Reel Big Fish’s ‘The Set Up’ (1998), and Stars’s ‘The Big Fight’ (2004).”⁽¹⁶⁾ While this scholarship is an important precedent for this article, I deliberately avoid positioning compound AAB form as being derived from compound AABA and instead focus on how it dialogues with more recent EDM- and hip-hop-influenced approaches to musical form.⁽¹⁷⁾ In other words, I view compound AAB as arising from the substitution of contrasting B material in place of the A₃ rotation in compound AAA, rather than the omission of the final A rotation in compound AABA—although this is an ambiguity that some songwriters deliberately exploit.

[2.3] In the case of “bad guy,” compound AAB form allows Eilish to create the two distinct performative characters that are reflected in the song’s music video: 1) the flamboyant character of the song’s first part, who Eilish performs with sing-song, staccato vocal flourishes, and 2) the more ominous character of the song’s coda, for whom Eilish uses a dead-pan, nearly whispered vocal technique. Therefore, this disruption has dramatic effects on not only our perception of the formal structure of “bad guy” but also the song’s thematic content, which is presented in a radically transformed guise in the extended coda-verse section. More generally, the contrasting coda-verse section of “bad guy” fulfills the mandate for musical contrast that Osborn identifies as an inherent aesthetic challenge for songs in compound AAA form.

[2.4] In an interview with Rolling Stone, Eilish recalls that this final section was composed nearly a year before the first part of the song and that her idea to use it at the end of “bad guy” came from two hip-hop songs, “NEVER” (JID, 2017) and “Stuck in the Mud” (Isaiah Rashad [feat. SZA], 2016), which both feature similar groove transformations (Eilish and Finneas 2019). By making this association, Eilish explicitly positions “bad guy” within a broader tradition of hip-hop production practices that originated with sampling and beat-making techniques developed in the early 1980s. As Joseph G. Schloss has argued, the hip-hop producers who pioneered these techniques ultimately paved the way for our contemporary view of music production as a process of “using digital sampling to create sonic collages” (2014, 21), a process that is now facilitated by the use of Digital Audio Workstations (DAWs).⁽¹⁸⁾ Central to this practice was the combination of looped samples from multiple source recordings, each with its own distinct metric, tonal, and timbral characteristics; collecting, altering, and assembling these diverse samples to create cohesive and aesthetically satisfying musical works was—and is still today—a highly complex process.⁽¹⁹⁾

[2.5] The techniques and technologies available to producers have changed dramatically since the 1980s, but one of the musical affordances intrinsic to sample-based production practices, then and now, is that samples with starkly contrasting musical features can easily be juxtaposed to establish distinct formal regions—a technique that came to be known as “beat switching.” Examples of beat switches can be found in early hip-hop tracks, such as The Beastie Boys’ “The New Style” (1986), Eazy E’s “Eazy-Duz-It” (1988), and Ice Cube’s “Jackin’ For Beats” (1990), and in recent songs, such as Kendrick Lamar’s “euphoria” (2024) and Drake’s “Family Matters” (2024). Both songs that Eilish mentions as inspiration for the coda-verse of “bad guy” include beat switches that disrupt the established groove and open onto new expressive regions in their respective narratives. Thus, although “bad guy” is perhaps more stylistically and commercially aligned with the genre of EDM-influenced mainstream pop, Eilish’s particular use of compound AAB might be viewed as a hip-hop solution to an EDM problem. The groove in the final B section, which resembles a trap beat, further reinforces the song’s connection to hip-hop style practices. Many of the following examples also live at this intersection of genres, exhibiting to some degree both the beat-making production techniques of hip hop and the energetic directionality of EDM. By drawing attention to the increasing number of songs in compound AAB form, this article furthers our understanding of how two major stylistic touchstones in popular music have converged to create a distinctive aesthetic and formal rhetoric.

[2.6] Two songs that clearly use compound AAB form to introduce musical contrast are H.E.R.’s “Still Down” (2017) and Travis Scott’s “STARGAZING” (2018). Both begin with two rotations through recognizable formal sections that establish distinctive musical grooves, then conclude with extended B sections that continue the songs’ narrative while departing markedly in texture, tempo, and harmony. Despite these shared formal characteristics, “Still Down” and “STARGAZING” effect their groove shifts in different ways and with different expressive results. In “Still Down” (see Example 5), the A₂ rotation is followed by a striking moment in which the tempo of the song is “tuned down” by nearly 10 beats per minute and the pitch center is lowered a whole step from D to C—a technique reminiscent of the “chopped and screwed” production practices that developed in the 1990s in Houston’s hip-hop scene.⁽²⁰⁾ Following an interlude with ad lib lyrics from the chorus, H.E.R. transitions into a final coda-verse section with new melodic and lyrical content that culminates in repeated fragments of the chorus lyric “Are you still down.” Although the narrative effect of this final section is not as dramatic as that of “bad guy,” the B material contributes a welcome period of musical contrast following the restrained A rotations, and functions unequivocally as the final large-scale formal group of “Still Down?”

[2.7] Travis Scott’s “STARGAZING” follows a formal arrangement that has become increasingly common in hip-hop music, in which each rotation is initiated by the chorus, or “hook,” rather than the verse (see Example 6).⁽²¹⁾ That is, following a rather long intro section, the A₁ rotation unfolds with chorus–postchorus–verse units, in which the chorus occupies a prominent, early position and the verse serves an ending function for the rotation, rather than a beginning function as would be typical in a pop-rock song. However, Scott’s formal organization becomes rather ambiguous when a second verse fails to appear at the end of the A₂ rotation and an instrumental passage that fades to near-silence is included in its place. Because this instrumental passage is equal in duration to the intro of “STARGAZING” and features nearly identical musical material, a first-time listener might be inclined to hear this section as an outro that interrupts the A₂ rotation before the second verse can be performed.⁽²²⁾ However, the ensuing silence is interrupted by a new beat that shifts the tonal center from G to B and increases the tempo by 15 BPM, over which Scott raps a final verse that also functions as the large-scale conclusion to “STARGAZING.”⁽²³⁾ This final section not only accommodates (albeit belatedly) the unfulfilled expectation for a verse in the A₂ rotation but also meets the need to incorporate some kind of musical contrast within the broader song.⁽²⁴⁾ Importantly, while the B group provides the “missing” verse, it cannot credibly be grouped with the A₂ rotation due to the dramatic beat switch and the two instrumental interludes that separate this final verse from the chorus–postchorus sections of the A₂ rotation. This section is, decisively, an independent and conclusive formal group. The textural disruption in “STARGAZING” has the additional effect of catalyzing a retrospective reinterpretation of the “outro” in the A₂ rotation, which must be viewed as either an interlude or perhaps a small-scale outro that effects closure only at the level of the song’s A rotations (i.e., “end-of-the-middle”).

[2.9] In addition to fulfilling the need for musical contrast, compound AAB form also offers substantial expressive opportunities for songwriters. For example, consider Nicki Minaj’s “Barbie Dreams” (2018), in which two verse–chorus rotations are followed by a final B section with distinct rhythmic and textural material (see Example 7). The title of “Barbie Dreams” is a reference to a 1994 song by The Notorious B.I.G., “Just Playing (Dreams),” which Minaj not only samples and incorporates as the foundation of the groove in the A rotations but also uses as a thematic starting point for her own song. “Barbie Dreams” is part of a broader assemblage of songs and public appearances in which Minaj assumes a character named Harajuku Barbie—a figure who deliberately transgresses the default whiteness and heteronormativity that Mattel’s Barbie Doll projects in popular culture. According to Philip Auslander, this Barbie character allows Minaj to use “her excessive performance of femininity and race as a means of addressing the gender politics of the musical genre in which she is situated and the gendered racial politics of the society at large” (2021, 212). The Harajuku Barbie character, as a representation of both “hyperfemininity” and “hyperblackness,” is an integral part of the “Barbie Dreams” narrative, where she works to rebuff the hypermasculinity of B.I.G.’s original track. Brad Osborn describes the relationship between these two songs:

Minaj’s use of the sample works on a number of levels. First, both tracks contain the word “Dreams,” with Minaj singing the rest of B.I.G.’s song title (“I’m just playin’/but I’m sayin’”) in each of the two choruses. More importantly, Minaj flips the script on B.I.G.’s original track, in which he brags about several imagined sexual conquests. Each verse of Minaj’s version imagines the poor sexual performance of several rappers, which are portrayed as puppets in the video. (Osborn 2021, 20)

Thus, although “Just Playing (Dreams)” is both the musical and thematic starting point for “Barbie Dreams,” Minaj’s song provocatively subverts B.I.G.’s message by countering with the symbol of “Black Barbie,” described by Margaret Hunter and Alhelí Cuenca as “a commodity who has gone rogue by upending gender norms, detouring from the politics of respectability for black women performers, and taunting the audience with queer performances while denying a queer identity” (2017, 33).

[2.10] The subversion that is central to the narrative of “Barbie Dreams” is mirrored and, in fact, strengthened by the song’s formal structure. B.I.G.’s “Just Playing (Dreams)” comprises two verse–chorus rotations and an outro that fades to silence as the rapper ad libs to the lyrics “I’m just playin’ / but I’m sayin’.” Minaj’s song similarly features a fade-out and ad lib after the second chorus, thus establishing a formal correspondence with B.I.G.’s track and introducing the possibility that this is the conclusion of the song. However, just moments after the B.I.G. sample completely fades out, a new groove consisting predominantly of drum samples is introduced and Minaj returns for a closing verse in which her rapping “reaches an impressive fever pitch” (Osborn 2021, 20). This final section not only formally exceeds B.I.G.’s original song but also serves as the moment at which Minaj becomes most emphatic in her critique of the prominent male rappers who are the targets of her verses. As Osborn notes in the quote above, several of these rappers are depicted as puppets in the music video for “Barbie Dreams,” perhaps to playfully visualize the contempt that Minaj has for them. However, while these puppets appear consistently throughout the A rotations of the song, they are entirely absent from this last verse. The final B group therefore represents the moment at which Minaj formally and thematically surpasses The Notorious B.I.G.’s braggadocio and presents herself as being in a class all her own.⁽²⁵⁾

3. Recapitulatory B Sections

[3.1] All of the examples discussed so far incorporate contrasting, thematically independent material for the ending temporal group. That is, these groove disruptions, once effected, mark decisive junctures in the formal structure of each song, opening onto entirely new expressive regions characterized by distinctive lyrical and textural content. In this section, I discuss two songs that recapitulate thematic material after the point of groove disruption, but that nevertheless avoid a complete expression of compound AABA form (or compound AAA form) by recasting this repeated material in a new form-functional guise.⁽²⁶⁾ As stated earlier, one effect of EDM’s influence on contemporary production practices is that songwriters often use the same musical material to different functional ends by modifying its texture or formal location. The songs considered in this section exploit this possibility, often in ways that seem to deliberately play on listeners’ expectations about formal structure and musical development.

[3.2] Maggie Rogers’s “Fallingwater” (2019) provides a relatively straightforward example of a recapitulatory B section, reusing lyrics from the chorus in the song’s closing group (see Example 8). The two A rotations of “Fallingwater” consist of verse–chorus–postchorus units, with only a slight textural thickening in the second rotation (an additional synth layer) serving to create a sense of dynamism throughout the song’s first three minutes. After the A₂ rotation, the tempo decreases from 102 BPM to 87 BPM and a new groove accumulates, consisting of a clapped back-beat rhythm, hummed vocals, syncopated block chords on the keyboard, and a looped vocal tag (“And I’m like falling water, set me free / You were like falling water coming down on me”). Once this new groove is established, Rogers recapitulates the melodic content of the chorus twice in full, first with the chorus’s original lyrics and then with a modified text, both of which are shown below the graph in Example 8. A brief outro concludes “Fallingwater” with fragments of lyrics from the postchorus, thus creating a loose parallel between the formal structure of the B group and the chorus–postchorus sections of the preceding A rotations.

[3.3] According to a voice recording that Rogers posted to Twitter (2018), the first two A rotations of “Fallingwater” had been finished for about a year when she had the idea to add the song’s final section, which she conceived as “this, like, outro chorus, this alternative chorus that happens over all of it.” This account of the song’s genesis reveals two distinctive features of how Rogers conceptualizes musical form. First, it suggests that Rogers considered the two A rotations of “Fallingwater” to be insufficient, on their own, to constitute a complete song—they existed independently for quite some time, but she still felt that some kind of contrasting material was necessary. Second, it shows that Rogers recognizes an inherent fluidity to formal functions that is contingent on musical characteristics such as tempo, texture, and formal location.⁽²⁷⁾ The recapitulated chorus material in the final section of “Fallingwater” is, for Rogers, so dramatically transformed by its musical context that it assumes a new formal function, what she calls an “outro chorus.” I preserve this blended function in my “coda-chorus” label in the form graph of Example 8, only substituting the term “coda” for Rogers’s “outro” to better reflect that this section introduces new musical material and assumes broader proportions.

[3.4] To my ears, this section is not a third rotation through A material but a distinct B section that both fulfills the mandate for large-scale musical contrast and provides closure by retooling preexisting thematic material so that it conforms to listeners’ rhetorical expectations for an ending.⁽²⁸⁾ This analysis is further strengthened by the fact that the formal structure of “Fallingwater” is so minimal. Although the two A rotations together span almost three minutes—an adequate duration for a contemporary pop song—even a first-time listener would likely anticipate some kind of contrasting material after the A₂ rotation. If “Fallingwater” were primarily in dialogue with standard verse–chorus form, this would probably entail a bridge followed by a final chorus; if it followed EDM’s compound AAA form, one would expect a drop in sonic energy and a climactic build that arrived at a final, celebratory chorus repetition.⁽²⁹⁾ That neither of these expectations is fulfilled in a predictable way is certainly notable, but it is not inherently surprising that the chorus is recapitulated for the final coda-chorus section, as such recapitulations are a feature of both the compound AABA and compound AAA models. By contrast, the next example, Childish Gambino’s “Little Foot Big Foot” (2024, feat. Young Nudy), draws on the stylistic and formal conventions of both EDM and hip hop to present a highly idiosyncratic formal structure with multiple interpretive possibilities.

[3.5] In March 2020, Childish Gambino—the songwriting alter-ego of actor and creative savant Donald Glover—released an album simply titled 3.15.20. That date was in the first weeks of the global Covid-19 pandemic, and this album drop seemed, understandably, a bit rushed: there was virtually no publicity prior to the release, nearly all the tracks had only timestamps for their titles, and the album cover was a simple, white square with the text “Donald Glover Presents” in a barely visible, off-white font color. Four years later, in May 2024, Glover released a “finished” version of the album with the title Atavista and removed the original album from streaming services.⁽³⁰⁾ Although the re-release was similar to the original album, with the exception of some production changes and the addition of song titles, a few significant alterations were made, including to the track originally released as “35.31.” Examples 9 and 10 include formal graphs for this song in both its original version and its re-released version, the latter of which is titled “Little Foot Big Foot” and includes an additional feature by Atlanta rapper Young Nudy. Considering how this song was modified for its “final” version will allow us to understand not only how Glover consciously manipulates formal conventions to achieve the song’s desired expressive ends, but also how compound AAB form creates a unique scenario in which thematic recapitulations can be recast with different formal functions.

[3.6] Both versions of “Little Foot Big Foot” open with a statement of the catchy, upbeat chorus—a structure that would seem to indicate that the song adheres to a chorus–verse paradigm, like Scott’s “STARGAZING.” However, although the verse’s first half is harmonically supported by alternating tonic and submediant chords, the harmonic progression of the second half is more clearly functional and culminates in a strongly articulated V⁷. This dominant chord resolves to tonic on the downbeat of the following chorus, thus clarifying a verse–chorus structure for the song’s rotations and retrospectively showing the first chorus to have an introductory function (indicated as “intro-chorus” in Examples 9 and 10). This is further corroborated by the comparatively thin texture of the opening chorus statement (which omits percussion) and by the presence of a brief instrumental interlude before the verse of the A₂ rotation in “35.31.” In the song’s original version, Glover modifies the A₂ rotation by interpolating an additional chorus section that exchanges the chorus’s familiar groove for ambient vocal harmonies. The unexpected drop in energy for this chorus statement is representative of what Nobile calls an “anti-telos chorus”—a common feature of EDM-influenced popular music from the 2010s in which the climactic arrival of a chorus is withheld by a sudden reduction in energy (2022, [4.4]–[4.11]). As is often the case, the anticipated climax in “35.31” is simply deferred to the following section of the song and the anti-telos chorus therefore serves to create a heightened sense of expectation and a stronger sense of arrival when the climax eventually does occur. The final chorus of the A₂ rotation is the moment of thickest texture and highest energy in the entire song—an arrival that is carefully prepared in the anti-telos chorus by a sustained V⁷ chord and a sung count-in by Glover. After this climactic chorus statement, a sudden groove shift occurs that foregrounds vocal harmonies like those heard in the anti-telos chorus but transposed to D♭ Lydian and accompanied by sampled hi-hat and synthesizer layers. Much like in “Fallingwater,” chorus material is recapitulated twice in this large-scale ending group and transformed to assume a rhetorical closing function, which in this case involves a shift to half-time feel, a long-range diminuendo, and a thinner texture.

[3.7] As shown in the form graphs of Examples 9 and 10, the “Little Foot Big Foot” re-release is nearly identical to “35.31” up to the moment when the second chorus of the A₂ rotation is most anticipated, even including the V⁷ in F major and the rhythmic count-in. However, rather than fulfilling the expectation for a climactic statement of the chorus, Glover pivots directly to a new groove over which featured artist Young Nudy raps a final, lengthy verse. This moment is striking not only because it departs from the structure of the original song, but also because it produces a deceptive resolution from the V⁷ at the end of the anti-telos chorus to a D♭M harmony, ♭VI in the key of F major. A truncated version of the original coda-chorus is still included in the re-released version, but with the percussion omitted and Glover’s voice transposed higher so that its rhetorical function as a closing section is more contextually apparent, despite being a recapitulation of chorus material. Unlike the coda-choruses of “35.31,” this outro-chorus is auxiliary to the core verse section that precedes it and is thus properly “after-the-end” of the song.

[3.8] If the groove disruptions in both versions of “Little Foot Big Foot” provide clear examples of thematic recapitulation in compound AAB form, they nevertheless introduce a strong ambiguity to the song’s grouping structure. The A₂ rotation in the original song is communicated unequivocally by the buildup–arrival relationship between the anti-telos chorus and the following climactic chorus. But this relationship is destabilized in the re-release, when a significant groove disruption prevents the ensuing section from making good on the expectation for a climax—or, at least, prevents it from doing so in the expected way. Consequently, there are compelling reasons for grouping the anti-telos chorus with both the A₂ rotation as an end-of-the-middle and, alternately, with the B group as a beginning-of-the-end. On the one hand, it is possible to hear the anti-telos chorus as anticipatory material that leads to the song’s final section, an interpretation that is supported by the strong (if deceptive) harmonic arrival at the onset of Young Nudy’s verse. Because the re-released version omits the instrumental interlude before the A₂ rotation, we might viably hear the song as having a chorus–verse structure that completes three rotations through the core sections, albeit with significant groove modifications for the B rotation—an interpretation I depict in Example 11. However, this glosses over the jarring textural shift that separates Young Nudy’s verse from the preceding music and fails to account for the strong harmonic trajectory from verse to chorus that supports the interpretation shown in Example 10. On the other hand, the anti-telos chorus might be heard as the end of the A₂ rotation, with its reduced energy serving to create a sense of closure that loosely parallels the outro-chorus of the B group. This interpretation is also unsatisfactory, however, as it ignores the accumulation of energy that occurs throughout the section, as well as the strong harmonic and rhythmic anticipation created by the V⁷ and count-in.

[3.9] Rather that suggesting that there is a “correct” interpretation of the temporal functions in “Little Foot Big Foot,” I find it more worthwhile to consider how this song is positioned at the intersection of formal and stylistic conventions drawn from mainstream popular and hip-hop music, and how this cross-pollination creates particular expressive affordances. The A rotations in “Little Foot Big Foot” are jaunty and driving, with highly repetitive rhythmic and melodic content and a formal structure that foregrounds the energetic chorus material. However, this musical exuberance masks the gravity of the underlying narrative, which centers on a child who is unable to escape a life of drug dealing, encountered through his father. The incongruity between music and narrative produces an expressive dissonance highlighting the grotesqueness of a social structure in which Americans with limited socioeconomic opportunities can become victims of generational trauma and inequality.⁽³¹⁾

[3.10] Notably, the two versions of this song grapple with this dissonance in distinctive ways. As we have seen, the A₂ rotation in “35.31” provides a satisfactory resolution of the musical tension that accumulates throughout the anti-telos chorus; however, the final chorus fails to contribute anything new that could address the expressive tension that Glover has developed by this point in the song. By contrast, “Little Foot Big Foot” does seem to provide some expressive resolution in the B group, when the groove switches to a more conventional trap beat.⁽³²⁾ Early in his verse, Young Nudy draws attention to how the musical transformation has changed the tenor of “Little Foot Big Foot,” claiming that he can “turn a pop song into a real scrap song” immediately before an 808 bass enters the texture and he begins to rap about his wealth, sexual conquests, and cultural prestige. The narrative that Glover developed in the preceding music fades to the background of the B group, displaced by an array of hip-hop tropes that abandon all the subtlety and double-entendre of Glover’s verses and conform more stereotypically to the braggadocious self-stylings of a hip-hop rapper. However, given the strong contrast between Young Nudy’s verse and the preceding music, we might hear this as a tongue-in-cheek response to the expressive dissonance of the A rotations, rather than a capitulation to any stereotypical connection between drug dealing and hip-hop music. It is as if Glover (and Young Nudy) are toying with the listener’s expectations by highlighting the absurdity of this expressive resolution, using the stark juxtaposition of trap music as a means of exaggerating and thereby subverting what Tricia Rose (2008, 3) calls the “explicitly exploitative and increasingly one-dimensional narratives of black ghetto life” that have dominated twenty-first-century hip hop.⁽³³⁾ We may therefore hear the formal and expressive functions of the groove disruption in “Little Foot Big Foot” as supporting a broader commentary on the societal forces that draw boundaries around musical styles, establish their generic expectations, and reinforce cultural stereotypes for the songwriters working within these spheres.

4. Beyond Compound AAB

[4.1] The preceding analyses have illustrated how compound AAB form serves as an expressive tool for songwriters who want to conclude their songs with distinct formal and expressive regions. However, contrasting closing groups show up in a variety of ways and not every example can be accommodated by compound AAB form. In the appendix of this article, I provide abbreviated form graphs for several additional songs with contrasting material in their ending temporal group, some of which are in modified strophic (e.g., A₁–A₂–B₁–B₂–Coda–Verse, where each letter corresponds to a single verse section) or large-scale binary form (typically compound AB or AABB, where each letter comprises a verse–chorus rotation or the like).⁽³⁴⁾ I believe it is also significant that many of the artists represented in the analyses of this article and in the appendix have multiple songs that explore this formal device in different ways, suggesting that it is both a well-defined technique in their songwriting toolkit and something of a personal signature in their approach to musical form. For example, following the success of “bad guy,” Billie Eilish released several more songs with contrasting ending groups, including “Happier Than Ever” (2021), “L’AMOUR DE MA VIE” (2024), and “BLUE” (2024). To illustrate how this methodology can be applied to songs that extend beyond compound AAB form, I turn to a representative case study: Daniel Caesar’s “TOO DEEP TO TURN BACK” (2019). This song features a contrasting bridge section before the closing group, resulting in a rather idiosyncratic compound AABC form.

[4.2] Caesar grew up in a devoutly religious household and much of his music consequently grapples with themes of spirituality, psychology, and mortality. In “TOO DEEP TO TURN BACK,” Caesar reflects on organized religion and its social function as an arbiter of morality, questioning whether such a worldview can allow actions that reflect “good intentions” but do not necessarily conform to orthodoxy. As shown in Example 12, each rotation comprises a verse and chorus section, in which the verses take the perspective of an individual who is struggling to reconcile their Judeo-Christian faith with persistent feelings of anxiety and inadequacy (“So it’s a quick fix / What else does life have to give but the pain / Generated inside our own brains?”), while the choruses depict a religion that requires absolute devotion (“Submit to me, come get down on one knee / I’m your savior and I’ll give you everything that you need”). After the A₂ rotation, the B group introduces a contrasting bridge-like section that continues the harmonic and textural content of the chorus but features a different melody and lyrics that bring closure to the song’s thematic content by asking the listener to turn inward (“The answers it seems / Are inside your own dreams”). The B group concludes with two further repetitions of the chorus’s harmonic progression, thickened by the addition of textless vocals and muted guitar arpeggiations and followed by a brief but decisive pause at the 4:00 mark.

[4.3] If “TOO DEEP TO TURN BACK” ended at this point, one might reinterpret the bridge-like section of the B group as a coda-verse, as it offers narrative closure, introduces new musical material, and is followed by an instrumental section that could serve as a viable outro. However, the formal function of this section becomes even more ambiguous when, after the pause, Caesar redirects into a lengthy closing section that is distinguished from the prior music by a change in the harmonic progression, a declamatory vocal delivery, and the introduction of piano and percussion to the texture. The addition of this closing group yields a compound AABC form, in which the first two A rotations are followed by two additional contrasting groups—an admittedly unusual formal structure but one that allows us to hear the first section of the B group as serving a bridge function, which aligns more intuitively with its fragmented melody and harmonically open conclusion on an unstable dominant chord. However, because this bridge-like section is followed by an instrumental closing section and a textural disruption, it is formally separated from the C part and thus the song’s temporal end—a departure from the standard temporal function of a bridge in compound AABA, where it typically functions as part of the middle group to prepare a recapitulation of the A material (see Example 3).

[4.4] There is thus a palpable tension between how this bridge-like section sounds and how it functions. Although it exhibits many sonic characteristics of a transitional bridge section, it ultimately serves as a closing unit for the song’s middle group, a formal function that would typically be filled by a postchorus or an instrumental tag in the A₂ rotation.⁽³⁵⁾ I have tried to convey my own experience of how this tension transforms the formal function of this section during the listening experience by using two “becoming” operators and a blended section label in Example 12: “bridge ⇒ coda ⇒ bridge-coda.” This admittedly unwieldy label is intended to reflect how this section bears the sonic characteristics of a bridge while serving an ending function within the middle temporal group. But even this complex designation is rather unsatisfactory, not least because codas are invariably positioned at the very end of a song’s formal structure. Nevertheless, as Trevor de Clercq (2017) argues, it is not necessarily possible or even desirable to reconcile such tensions in our analyses of popular song form, and I follow him in viewing the formal ambiguity of “TOO DEEP TO TURN BACK” as an intrinsic feature of its formal structure rather than a problem to be solved with a definite analysis. Indeed, as we have seen in several examples above, formal ambiguity is a common byproduct of the idiosyncratic structural dynamics of groove shifts. That is, dramatically upsetting the groove of a song creates such a strong formal boundary that it becomes perceptually unintuitive to group across the moment of transformation, which in turn creates the potential for tensions to arise between a section’s rhetorical function, formal function, and location within the broader structure of the song. The confluence of these competing semantics is what ultimately results in polysemic sections like the bridge-coda in “TOO DEEP TO TURN BACK.”

[4.5] As with many of the songs considered in this article, the groove disruption at the end of “TOO DEEP TO TURN BACK” corresponds to a pivotal moment in the song’s narrative. As already discussed, the bridge-coda section provides closure to some of the main themes of the song with its invocation to seek answers not in organized religion but within one’s subconscious. Caesar takes this assertion a step further in the final coda-verse section by immersing us in what seems to be one of his own dreams—a liminal place in which he comes face-to-face with his own mortality.⁽³⁶⁾ Rather than succumbing to anxiety as he does in the preceding verses, Caesar instead compares himself to Jacob and Elijah, Biblical figures with extraordinary connections to the heavenly realm (“I’ve slept like Jacob, a rock for a pillow / Run swift like Elijah, away from the middle”).⁽³⁷⁾ By plumbing the depths of his subconscious, Caesar achieves a state of spiritual enlightenment and encourages the listener to accompany him on this journey (“Follow me to salvation”), suggesting that his transcendence is something that can be experienced by anyone who overcomes the feeling that they are “too deep to turn back” from external arbiters of ethical and spiritual action. The musical transformations that accompany this narrative shift—the rapid rhythmic layer, thicker instrumentation, chromatic melodic movement, and harmonic instability—serve to establish this C group as a distinct expressive region that is less predictable and more “dreamlike” than what we encountered in the preceding music.

Conclusion

[5.1] The analyses in this article illustrate not only the importance of texture in our perception of formal structure, but also the idiosyncratic formal and expressive effects that may result when textural disruptions signal the onset of a distinct closing group. These insights are afforded by close attention to popular music’s conventional large-scale temporal groupings, as well as the ambiguities and reinterpretations that arise when standard temporal and form-functional structures are subverted. Furthermore, by considering examples from a broad cross-section of popular music, I have suggested that the increased use of contrasting ending groups reflects the convergence of formal conventions from two specific stylistic touchstones, EDM and hip hop. This study contributes to our understanding of how contemporary songwriters work in dialogue with multiple stylistic practices throughout the compositional process and highlights the ongoing interplay among genres—such as EDM and hip hop—within the broader landscape of popular music.

[5.2] Although the scope of this article is deliberately limited to recent developments in popular music, the methodology I have developed for analyzing temporal functions is adaptable to a range of stylistic contexts beyond music influenced by EDM and hip hop. In particular, there is a long tradition in the rock and folk genres of writing “song pairs” that are connected by shared thematic and musical content, such as Pink Floyd’s “Brain Damage” and “Eclipse” (1973) or Green Day’s “Brain Stew” and “Jaded” (1995).⁽³⁸⁾ In some cases, these songs can be productively analyzed according to a single beginning–middle–end paradigm at their highest structural level. For example, because “Brain Damage” lacks a bridge section, we might hear “Eclipse” as a contrasting ending group that provides closure for the entire two-song unit. The song pair technique has manifested more recently as multipart songs that are recorded as a single track, such as The Decemberists’s “Rusalka, Rusalka / Wild Rushes” (2018) or Fleet Foxes’s “Third of May / Ōdaigahara” (2017), both of which can be found in the appendix.⁽³⁹⁾ Like many of the examples in this article, these multipart songs establish discrete formal groups through transformations of the musical texture, often in order to signal shifts in their expressive register or narrative arch. This is the case in “Rusalka, Rusalka / Wild Rushes,” where a shift from D Aeolian to D Mixolydian and an increase in tempo serve to distinguish the two parts of the song, which are both in strophic form with three verses each. Although multipart songs are composed in dialogue with a different set of formal conventions than those under consideration in this article—typically the verse–chorus form of rock or strophic form of folk—further study might outline how multipart formal structures developed in parallel to the groove disruptions that I have discussed here, with ramifications for how we understand the stylistic lineage of several major currents in today’s popular music landscape that are operating just outside the EDM- and hip-hop-influenced mainstream. Such research would further elaborate the premise of this article by showing how the broad stylistic palette of recent popular music has led songwriters to develop inventive approaches to form and texture, resulting in a novel poetics of form that is uniquely suited to the diverse expressive requirements of the twenty-first-century songwriter. 

Appendix

Nathan Cobb
Department of Music
Emory University
1804 N Decatur Rd
Atlanta, GA 30322
nathan.cobb@emory.edu

Works Cited

Adams, Kyle. 2019. “Musical Texture and Formal Instability in Post-Millennial Popular Music: Two Case Studies.” Intégral 33: 33–45.

Agawu, Kofi. 1991. Playing with Signs: A Semiotic Interpretation of Classical Music. Princeton University Press. https://doi.org/10.1515/9781400861835.

Auslander, Philip. 2021. In Concert: Performing Musical Persona. University of Michigan Press. https://doi.org/10.3998/mpub.10182371.

Barna, Alyssa. 2020. “The Dance Chorus in Recent Top-40 Music.” SMT-V 6.4. https://doi.org/10.30535/smtv.6.4.

Bungert, James. 2019. “‘I Got a Bone to Pick’: Formal Ambivalence and Double Consciousness in Kendrick Lamar’s ‘King Kunta.’” Music Theory Online 25 (1). https://doi.org/10.30535/mto.25.1.12.

Butler, Mark. 2006. Unlocking the Groove: Rhythm, Meter, and Musical Design in Electronic Dance Music. Indiana University Press.

Caplin, William. 1998. Classical Form: A Theory of Formal Functions for the Instrumental Music of Haydn, Mozart, and Beethoven. Oxford University Press. https://doi.org/10.1093/oso/9780195104806.001.0001.

—————. 2009. “What are Formal Functions?” In Musical Form, Forms, and Formenlehre, ed. Pieter Bergé, 21–40. Leuven University Press. https://doi.org/10.2307/j.ctt9qf01v.5.

de Clercq, Trevor. 2017. “Embracing Ambiguity in the Analysis of Form in Pop/Rock Music, 1982–1991.” Music Theory Online 23 (3). https://doi.org/10.30535/mto.23.3.4.

Covach, John. 2005. “Form in Rock Music: A Primer.” In Engaging Music: Essays in Music Analysis, ed. Deborah Stein, 65–76. Oxford University Press.

Duinker, Ben. 2019. “Good Things Come in Threes: Triplet Flow in Recent Hip-Hop Music.” Popular Music 38 (3): 423–56. https://doi.org/10.1017/S026114301900028X.

—————. 2020. “Song Form and the Mainstreaming of Hip-Hop Music.” Current Musicology 107: 93–135. https://doi.org/10.52214/cm.v107i.7177.

Eilish, Billie, and Finneas O’Connell. 2019. Interview by Rolling Stone. December 16, 2019. YouTube video, 11:41. “Billie Eilish and Finneas Break Down Her Hit Song ‘Bad Guy.’” Accessed April 28, 2023. https://www.youtube.com/watch?v=kpx2-EMfdbg.

Endrinal, Christopher. 2011. “Burning Bridges: Defining the Interverse in the Music of U2.” Music Theory Online 17 (3). https://doi.org/10.30535/mto.17.3.6.

Everett, Walter. 2009. The Foundations of Rock: From “Blue Suede Shoes” to “Suite: Judy Blue Eyes.” Oxford University Press. https://doi.org/10.1093/oso/9780195310238.001.0001.

Geary, David. 2024. “Formal Functions of Drum Patterns in Post-Millennial Pop Songs, 2012–2021.” Music Theory Online 30 (2). https://doi.org/10.30535/mto.30.2.3.

Gilke, Chloe. 2019. “Billie Eilish’s ‘Bad Guy’ Video is a Surrealist, Artistic Nightmare.” Uproxx. Accessed November 10, 2022. https://uproxx.com/music/billie-eilish-bad-guy-video/.

Heetderks, David. 2020. “Play with Closing Markers: Cadential Multivalence in 1960s Prechoruses and Related Schemas.” Music Theory Spectrum 42 (1): 1–3. https://doi.org/10.1093/mts/mtz023.

Hepokoski, James. 2010. “Clouds and Circles: Rotational Form in Debussy’s ‘Nuages.’” Dutch Journal of Music Theory 15: 1–17.

Hepokoski, James, and Warren Darcy. 2006. Elements of Sonata Theory: Norms, Types, and De-formations in the Late Eighteenth-Century Sonata. Oxford University Press. https://doi.org/10.1093/acprof:oso/9780195146400.001.0001.

Hudson, Stephen S. 2021. “Compound AABA Form and Style Distinction in Heavy Metal.” Music Theory Online 27 (1). https://doi.org/10.30535/mto.27.1.5.

Hunter, Margaret, and Alhelí Cuenca. 2017. “Nicki Minaj and the Changing Politics of Hip-Hop: Real Blackness, Real Bodies, Real Feminism?” Feminist Formations 29 (2): 26–46. https://doi.org/10.1353/ff.2017.0015.

Nobile, Drew. 2016. “Harmonic Function in Rock Music: A Syntactical Approach.” Journal of Music Theory 60 (2): 149–80. https://doi.org/10.1215/00222909-3651838.

—————. 2020. Form as Harmony in Rock Music. Oxford University Press. https://doi.org/10.1093/oso/9780190948351.001.0001.

—————. 2022. “Teleology in Verse–Prechorus–Chorus Form, 1965–2020.” Music Theory Online 28 (3). https://doi.org/10.30535/mto.28.3.8.

Osborn, Brad. 2013. “Subverting the Verse-Chorus Paradigm: Terminally Climactic Forms in Recent Rock Music.” Music Theory Spectrum 35 (1): 23–47. https://doi.org/10.1525/mts.2013.35.1.23.

—————. 2021. Interpreting Music Video: Popular Music in the Post-MTV Era. Routledge. https://doi.org/10.4324/9781003037576.

—————. 2023. “Formal Functions and Rotations in Top-40 EDM.” Intégral 36: 35–54. https://www.doi.org/10.17613/6wk7e-76341.

Peres, Asaf. 2016. “The Sonic Dimension as Dramatic Driver in 21st-Century Pop Music.” PhD diss., University of Michigan.

—————. 2018. “Sonic Functions: The Producer’s Alternative to Harmonic Functions in Modern Music.” Top40 Theory. Accessed July 15, 2024. https://www.top40theory.com/blog/sonic-functions-the-alternative-to-harmonic-functions-in-modern-music.

Reuter, Anders. 2022. “Who let the DAWs Out? The Digital in a New Generation of the Digital Audio Workstation.” Popular Music and Society 45 (2): 113–28. https://doi.org/10.1080/03007766.2021.1972701.

Rogers, Maggie. 2018. Twitter post, June 5, 2018, 8:36am. Accessed July 8, 2024. https://x.com/maggierogers/status/1004024124438560768?lang=en.

Rose, Tricia. 2008. The Hip Hop Wars: What We Talk About When We Talk About Hip Hop—And Why It Matters. Basic Books.

Schloss, Joseph G. 2014. Making Beats: The Art of Sample-Based Hip-Hop. Wesleyan University Press.

Schmalfeldt, Janet. 2011. In the Process of Becoming: Analytic and Philosophical Perspectives on Form in Early Nineteenth-Century Music. Oxford University Press.

Stephenson, Ken. 2002. What to Listen for in Rock: A Stylistic Analysis. Yale University Press. https://doi.org/10.12987/yale/9780300092394.001.0001.

Stroud, Cara. 2022. “Codetta and Anthem Postchorus Types in Top-40 Pop from 2010 to 2015.” Music Theory Online 28 (2). https://doi.org/10.30535/mto.28.2.9.

Summach, Jason. 2012. “Form in Top-20 Rock Music, 1955–89.” PhD diss., Yale University.

Tatar, Jeremy. 2024. “Meter, Phrase, and Form in Sample-Based Hip Hop.” PhD diss., McGill University.

Temperley, David. 2018. The Musical Language of Rock. Oxford University Press. https://doi.org/10.1093/oso/9780190653774.001.0001.

Footnotes

* I extend thanks to the many individuals who offered their perspectives and feedback at various stages of this project, including Janet Bourne, Benjamin Levy, and the editors and anonymous reviewers of this journal. I also thank Emma Keller for her outstanding work animating the examples in this article. An earlier version of this project was presented at the Annual Meeting of the Society for Music Theory in 2022.
Return to text

1. I follow Brad Osborn (2023) by using the term “rotation” to refer to sections of a song that are structurally connected and, typically, repeated together. Note that Drew Nobile (2022, [4.16]) cites “bad guy” as an example of ambiguous chorus function, due to the lack of lyrics in the “arrival” section at the end of each rotation. David Geary (2024, [3.7]), by contrast, locates the chorus immediately after the verse section, thus presumably reinterpreting the final section of the rotation as a postchorus. In an interview with Rolling Stone, Eilish acknowledges that this final section is somewhat unusual but nevertheless refers to it as the chorus of “bad guy,” a terminology that I follow in my analysis (Eilish and Finneas 2019).
Return to text

2. Note that the music video also reinforces the moment after the repeated chorus as a point of closure by reversing the video’s opening footage of Eilish kicking through a wall of paper (Example 2).
Return to text

3. Blended functions are discussed at length in de Clercq 2017, Adams 2019, and Osborn 2023. The distinction between core sections (verses, choruses, bridges, etc.) and auxiliary sections (intros, outros, interludes, etc.) is outlined in Summach 2012 (40–57).
Return to text

4. I use a similar graphing technology throughout this article, depicting the duration of formal sections by the relative width of their “cell” in the graph and outlining major changes to the groove on either side of a dotted line. Auxiliary sections are indicated in italics, while core sections are in standard roman font.
Return to text

5. My thanks go to an anonymous reader for suggesting this alternative interpretation, in which the overall form of “bad guy” is heard as unexpectedly contracted when the final A rotation fails to return after the contrasting section. The combination of verse–chorus rotations to create large-scale compound forms was first proposed by John Covach (2005), but this idea has been developed in greater detail in Temperley 2018 (Chapter 8), Hudson 2021 ([2.0.0–2]), and Nobile 2020.
Return to text

6. See the appendix for additional examples of songs that feature this formal device, both within and beyond the EDM- and hip-hop-influenced practices that are the focus of this article.
Return to text

7. The most systematic account of pre-1991 popular form to date is provided in Drew Nobile’s Form as Harmony in Rock Music (2020), where he suggests that rock harmony follows a syntactical structure similar to the tonic-predominant-dominant-tonic phrase model of classical music, referred to as a functional circuit (see also Nobile 2016).
Return to text

8. William Caplin also extensively develops the concept of “temporal function” through association with his theory of formal functions in Western tonal music (1998, 2009).
Return to text

9. A representative approach can be found in Nobile’s “Teleology in Verse–Prechorus–Chorus Form, 1965–2020,” in which he uses the initiation–buildup–arrival functions to differentiate between three rotation types: verse–prechorus–chorus form (in which each section correlates to a single function), continuous verse–chorus form (in which the verse serves an initiation function and the chorus provides both the buildup and arrival functions), and sectional verse–chorus form (in which the verse and chorus each comprise a full initiation–buildup–arrival cycle) (2022, [1.2]).
Return to text

10. The work of Asaf Peres (2016; 2018) and David Geary (2024) are further precedents for the methodology outlined in this article, as they also apply versions of the beginning–middle–end paradigm to the large-scale rotations of entire songs.
Return to text

11. See also Nobile’s description of “compound strophic form” in Form as Harmony in Rock Music (2020, 195–98).
Return to text

12. Note that Osborn uses the terms “riser” and “drop” in place of “chorus” and “postchorus” to describe the buildup and arrival sections of each rotation. Although these alternative terminologies are appropriate for the clearly EDM-influenced songs that he discusses, not all of the songs that I analyze include unequivocal “drop” sections, so I retain the more familiar “chorus” terminology in this figure and subsequent analyses.
Return to text

13. See Peres 2016 (Chapter 5, especially 134–38) for a further discuss of how temporal groups and thematic material relate in compound AABA form. Note that he uses the terms setup–buildup–peak.
Return to text

14. My application of compound AAB form is distinct from that outlined in Peres (2018), where it is used as a higher-level temporal grouping for standard verse–chorus form (with the B section correlating to the final bridge and chorus units that serve as the song’s “climax”). Also note that because the B section does not participate in the “repetitive circularity” of the preceding two A sections, it cannot usually be considered a “rotation” in the same sense.
Return to text

15. In this analysis, the specific formal structure of an individual rotation is ultimately of secondary importance to the overall ternary grouping. This explains, in part, why the coda-verse of “bad guy” has an end function rather than the after-the-end function that is typical of a coda–the textural contrast is so dramatic and the section’s duration so extended that its perceptual weight is roughly equivalent to the preceding two rotations. There are countless songs that feature textural disruptions in their closing material but lack the proportions for this material to assume large-scale significance, such as “This is America” (2018) by Childish Gambino, “love language” (2020) by Ariana Grande, or “brutal” (2021) by Olivia Rodrigo. On similar grounds, I also deliberately exclude what Walter Everett identifies as the “unrelated coda” (2009, 154) in classic pop-rock music, as this material is typically either too brief (e.g., “Strawberry Fields Forever” [The Beatles, 1967] or “Cry Baby Cry” [The Beatles, 1968]) or comprises a loosely organized instrumental breakdown that is more properly “after-the-end” (e.g., “Bold as Love” [Jimi Hendrix, 1967], “Spinning Wheel” [Blood, Sweat and Tears, 1968], or “Thank You” [Led Zepplin, 1969]).
Return to text

16. It is worth noting at this point that Osborn’s (2013) concept of a “terminally climactic form” (TCF) is not mutually exclusive with the idiosyncratic formal structures that I am considering in this article and his two-part model for a TCF is particularly likely to overlap with compound AAB form as I present it here. However, although nothing precludes a terminal climax from occurring in the songs I am focusing on, climax and memorability–two important criteria for TCFs–do not play a significant role in my methodology. On the contrary, most songs that I have encountered feature a decrease in energy in their final contrasting sections that is more akin to what Osborn identifies here as the “anti-climax” of a compound AABA form that omits the final A rotation. Furthermore, the contrasting B group in compound AAB is not entirely precluded from recapitulating thematic material from the A rotations, as I discuss in the next section.
Return to text

17. In this way, I am building directly on an observation made by Osborn in a later publication (2021, 21), in which he acknowledges that codas with unrelated musical material “are relatively rare, though they do seem to be occurring with some frequency in recent hip-hop.”
Return to text

18. In the Rolling Stone interview, Finneas O’Connell, Eilish’s brother and the producer of “bad guy,” demonstrates several of his production techniques for the song in Apple’s Logic Pro X, one of the major DAWs in use today (Eilish and Finneas 2019). See Reuter 2022 for an analysis of how the technological affordances of DAWs have influenced recent popular music production practice.
Return to text

19. Consequently, scholars like Jeremy Tatar have recently framed hip-hop production as both a creative and an “analytic act:” “Producers are expert listeners possessing many forms of expert knowledge. The beats that they create are the products of multiple interpretive choices; they represent, in other words, the way that a producer hears their source(s)” (2024, 30).
Return to text

20. There is a clear connection here to the song’s title, as both the tempo and pitch center shift “down” for this final large-scale group. My thanks to one of the anonymous reviewers of this article for noting this correspondence.
Return to text

21. See Duinker 2020 (114–18) for a corpus analysis that aims to determine when and why the hook–verse paradigm became a common practice. Note that, for the sake of consistency, in this example and the following I employ the term “chorus” to refer to recurring sections of a song with invariant lyrics, although “hook” is more typical for hip hop, as Duinker explains in the same publication (119–20).
Return to text

22. Determining whether a song uses chorus–verse or verse–chorus arrangement is not always straightforward and can occasionally lead to similarly idiosyncratic or ambiguous formal structures like we find in “STARGAZING.” See Bungert 2019 for an analysis of “formal ambivalence” in Kendrick Lamar’s “King Kunta,” which arises from this very ambiguity.
Return to text

23. Unlike many of the other analyses in this article, I do not interpret the final section of “STARGAZING” as having a blended coda-verse function, primarily due to the increase in tempo. The same reasoning also applies for the next analysis, of Nicki Minaj’s “Barbie Dreams.”
Return to text

24. It is worth noting that the need for contrasting material is especially pressing in “STARGAZING,” because not only is the same groove sustained throughout both A rotations (with the exception of the intro/interlude sections), but Scott deliberately blurs the formal functions of the first verse and chorus sections by opening them with the same lyrics, “Rollin’ rollin’ rollin’, got me stargazing,” which makes them initially indistinguishable.
Return to text

25. Note that Osborn (2021, 21) identifies the B section of “Barbie Dreams” as an “unrelated coda.”
Return to text

26. For a thorough account of “thematic recapitulation” in popular music, see Osborn 2013.
Return to text

27. I am not suggesting that Rogers’s views are representative of other songwriters, but rather offering one example of an artist whose compositional approach resonates with recent scholarship on the inherent ambiguity of popular formal functions.
Return to text

28. See Heetderks 2020 and Stroud 2022 for definitions of what constitutes “closing rhetoric” in the context of popular music.
Return to text

29. I use the term “dialogue” here and in the remainder of my analyses in the sense developed by James Hepokoski and Warren Darcy, according to which the compositional act involves entering “into a dialogue with an intricate web of interrelated norms as an ongoing action in time” (Hepokoski and Darcy 2006, 10).
Return to text

30. The original album remains available for streaming on Donald Glover’s YouTube account: Donald Glover, “3.15.20,” YouTube, March 22, 2020, https://www.youtube.com/watch?v=pTGXfF05M0s&t=2357s. Accessed July 8, 2024.
Return to text

31. Glover also uses this technique in an earlier song, the Grammy Award-winning “This is America” (2018), which centers on the exploitation and appropriation of Black culture in America. The music videos for these two songs further emphasize the disconnect between their upbeat music and grave thematic content by using drastic tonal shifts in the filmed narratives and by situating absurdly animated performances in bleak surroundings.
Return to text

32. “Trap” is a hip-hop subgenre that originated in the early 2000s in close association with Atlanta’s drug trade and has become nearly ubiquitous in recent hip hop. Ben Duinker (2019) characterizes trap music as having “deep sub-bass lines and 808 beats (drum beats produced by the Roland TR-808 machine) [...] supplemented by synthesiser-driven string lines, minor and diminished harmonic sonorities, slower tempos and inconsistent hi-hat patterns” (435). By the late 2010s, trap shed most of its association with drug dealing as several popular musicians, such as Ariana Grande and Miley Cyrus, imported characteristics of the genre into mainstream popular music.
Return to text

33. This interpretation is further corroborated by the music video, in which Glover performs the music of the A rotations in a Jim Crow–Era roadside bar but zooms out for the B group to show a larger-than-life Young Nudy towering over the bar for the delivery of his verse. The exaggerated contrast between these parts of the song is possibly intended to contribute an additional degree of surrealism to the already striking expressive incongruity of Glover’s narrative and the accompanying music.
Return to text

34. See Stephenson 2002 for an account of compound binary form in classic pop-rock music.
Return to text

35. Related to this subject is Cara Stroud’s identification of an “ambiguous” blend of postchorus and bridge functions in Maroon 5’s “Animals” (2022, [6.5–7]).
Return to text

36. Caesar explores the subject of dreams extensively in his music, even naming his 2017 breakthrough album Freudian.
Return to text

37. Accounts of Jacob and Elijah are both provided in the Hebrew Bible, in the books of Genesis and 2 Kings, respectively. In “TOO DEEP TO TURN BACK,” Caesar references a dream of Jacob’s in which he sees a ladder stretching from heaven to earth, which in many Judeo-Christian interpretations symbolizes a connection between the two realms. The prophet Elijah is notable in this context for the story of his ascension, according to which he is carried to heaven in a chariot of fire and thus never experiences death.
Return to text

38. A “song pair” is a set of songs that are conventionally played back-to-back when aired on the radio. While the degree of musical continuity between the songs can vary significantly, they are typically sequential tracks on an album or the A- and B-sides of a 45-rpm single. To my knowledge, song pairs have not been discussed in scholarship at any length, but several notable examples can be found in pre-1990 pop-rock music, including the Beatles’ “Sgt. Pepper’s Lonely Hearts Club Band” and “With A Little Help from My Friends” (1967), Led Zeppelin’s “Friends” and “Celebration Day” (1970), ZZ Top’s “Waitin’ for the Bus” and “Jesus Just Left Chicago” (1973), Queen’s “We Will Rock You” and “We Are the Champions” (1977), Journey’s “Feeling That Way” and “Anytime” (1978), Van Halen’s “Eruption” and “You Really Got Me” (1978), Pink Floyd’s “Empty Spaces” and “Young Lust” (1979), Genesis’s “Home by the Sea” and “Second Home by the Sea” (1983), and INXS’s “Need You Tonight” and “Mediate” (1987). I am grateful to one of the anonymous reviewers for pointing out the connection between the compound song forms I discuss and this older practice of writing song pairs.
Return to text

39. Pre-1990 examples of multipart songs include Elton John’s “Funeral For A Friend / Love Lies Bleeding” (1973), Boston’s “Foreplay / Longtime” (1976), and Chicago’s “Hard to Say I’m Sorry / Get Away” (1982).
Return to text

Copyright Statement

Copyright © 2025 by the Society for Music Theory. All rights reserved.

[1] Copyrights for individual items published in Music Theory Online (MTO) are held by their authors. Items appearing in MTO may be saved and stored in electronic or paper form, and may be shared among individuals for purposes of scholarly research or discussion, but may not be republished in any form, electronic or print, without prior, written permission from the author(s), and advance notification of the editors of MTO.

[2] Any redistributed form of items published in MTO must include the following information in a form appropriate to the medium in which the items are to appear:

This item appeared in Music Theory Online in [VOLUME #, ISSUE #] on [DAY/MONTH/YEAR]. It was authored by [FULL NAME, EMAIL ADDRESS], with whose written permission it is reprinted here.

[3] Libraries may archive issues of MTO in electronic or paper form for public access so long as each issue is stored in its entirety, and no access fee is charged. Exceptions to these requirements must be approved in writing by the editors of MTO, who will act in accordance with the decisions of the Society for Music Theory.

This document and all portions thereof are protected by U.S. and international copyright laws. Material contained herein may be copied and/or distributed for research purposes only.