Conference Agenda

Cardiacs were an English cult rock band known for their blend of progressive and punk genres and for their bizarre aesthetic. Over the band’s career (1977-2008), their primary songwriter Tim Smith developed an idiosyncratic harmonic language, jokingly described by composer Craig Fortnam as “loads of major chords randomly [arranged] on a page.” Previous scholarship has observed the pervasive use of major triads in various contexts, including the “pentatonic systems” (Biamonte 2010, Everett 2004) and “riff schemes” (Easley 2015) of hard rock, metal, and punk. Since these practices often weaken diatonicism, suspend harmonic function, and subvert tonal centricity, they are fundamentally at odds with the functional understanding of rock harmony. What theoretical tools might help us understand this seemingly random harmonic syntax?

This presentation explores the idea that Cardiacs’ triadic successions behave as motives within a broadly developmental approach. To describe harmonic motives, I use Murphy’s (2014) "MnM" labeling system, which indicates the triad type and the ordered intervallic distance between chord roots (ex. M2M). These motives are tonally flexible, making them adaptable to both surface and global developmental processes. To support this hypothesis, I reference three songs that demonstrate this motivic approach, in which we find procedures including (1) varied repetition by transposition, inversion, and interval cycle, (2) the large-scale projection of a motive’s interval span, and (3) transformations of a chord series, such as by addition/subtraction and reordering. Beyond these surface techniques, this presentation primarily considers the larger strategy behind these procedures, loosely viewing them through the prism of Schoenberg’s developing variation, whereby a basic unit – often the first chord series of a song – provides the impetus for further motive forms.

Though my analyses largely counter the charge of randomness, substantial challenges remain for pop listeners experiencing Cardiacs’ chord progressions in real time. My presentation concludes by pondering how listeners might navigate these highly contextual developmental processes with an approach focused on surprise and reaction. As such, this research can serve as a model for future investigations of harmonic approaches that both exploit and defy genre conventions.

Chord loops in popular music challenge traditional formulations of tonal syntax in two main ways. Firstly, they often do not project a single clear tonal center. Secondly, the contextual behavior of chords in loops is poorly explained by teleological models of harmonic function developed for Western common-practice music. Yet behavioral studies have shown that listeners ascribe different ratings of stability to different chords in a loop, suggesting that they engender a flow of tension and release. How might we model this flow?

We propose a two-dimensional account of syntactic tension whose axes embed two different temporal orientations. Present-oriented tension (POT) captures how surprised we are to hear the current chord, while future-oriented tension (FOT) captures how strongly we expect a particular chord to follow. We first quantify these dimensions using concepts from information theory. Then, we train a machine-learning model to estimate POT and FOT in the McGill-Billboard corpus (Burgoyne, Wild, and Fujinaga 2011), a dataset of 735 harmonic analyses of popular songs from 1958-1991 that we represent using a la-based minor system. This system was chosen to avoid making any assumptions about a song’s tonal center.

We find that two-chord shuttles consistently move from a chord with low POT and FOT (which we define as the “head”) to a chord with high POT and FOT. Moreover, we find that three- and four-chord loops combine the expected syntactic tension dynamics of their constituent pairs of chords, hence providing consistent functional interpretations of those chords. Finally, we analyze “Just Like Jesse James” by Cher (1989) using our model. We show that the model automatically encodes the phrase structure of the song by highlighting the entry and exit points of the do-sol-fa-sol loop while also implicitly learning the song’s hypermeter. Ultimately, our computational approach shows that a model that is given no information about tonality beyond a set of arbitrary tokens corresponding to solfège chord roots nevertheless develops implicit knowledge about tonal syntax in loop-based popular music.

Few musical expressions have as explicit a relationship with microrhythmic phrasing as hip-hop. Emcees will reference their stylistic rhythmic deliveries in their lyrics, the online discourse amongst hip-hop-heads will routinely reference “pocket” or “laidbackness”, and late legendary producer J Dilla/Jay Dee has been credited with inventing an entire tradition of rhythmic feel – “Dilla Time” (Charnas, 2022).

The field of rap analysis has grown into a discipline of its own, with theoretical and methodological frameworks imported and remixed from musicology/music theory, literature, linguistics, in addition to a rich body of interdisciplinary scholarship on Black aesthetics. Microrhythmic phrasing or “microtiming” or “expressive timing” is an area of interest in rap analysis which has seen a remarkable rise in attention in latter years.

There are several challenges relating to the analysis of microrhythmic phrasing in rap. First, there are questions of measurement tools, precision, granularity and representation. Further, one can ask to which degree different listeners and different listenings yield different interpretations of the same musical passage. To tackle questions like this, this paper will emphasize the role of categorical perception, and discuss the advantages and challenges that framework affords microrhythmic analysis. The paper will argue that a musical expression like rap allows for great categorical malleability. Listeners will interpret and reinterpret rhythmic structure as the music unfolds, and these interpretations are continuously informed and influenced by a multitude of factors including linguistic and semantic content, the rap’s relationship with the musical background, and preexisting and emergent reference structures/schemata in the listener.

Using several excerpts of what could be described as “structurally malleable” rap, concepts like rhythmic tolerance (Johansson, 2010) and structural bothness (where a rhythmic event or group of events can be thought of as first being one structural “thing” before being reinterpreted as another) will be introduced and illuminated. What do emcees do when they stretch categorical boundaries in their microrhythmic phrasing? What does it do to the music? And how should we as analysts approach a musical structural parameter which is so fundamentally dependent on the listener and listening experience?