This seminar will explore some of the challenges to musical analysis raised by popular music in the twentieth century. Some of the topics we will explore include innovations in tonal syntax introduced in popular songs of the 1920s and 1930s, extensions to tonal and rhythmic syntax effected by jazz musicians, the rhythmic structure of early and "classic" blues and rock and roll, relationships between musical structure and dramatic structure in popular songs, the identity/ontology of popular songs, the relationship between words and music throughout popular song, and relationships between the value ascribed to popular music and the enterprise of musical analysis.
In addition to examining critically a portion of the existing analytical literature on popular music, the seminar will also undertake a close reading of Part II of Simon Frith’s 1996 Performing Rites with a view toward developing new approaches to the analysis of popular music. Some of these new approaches will be adaptations of existing work, but some may also draw upon the resources suggested by recent work in cognitive science.
Meeting 1 (1/3/02)
Assigned material
We began the seminar with a few orientations. First, personal history on my part. As a youngster who came to the guitar out of a fascination with the instrument rather than because of what it represented to popular music, and who learned "classical" guitar as simply another style (like "jazz" guitar, or "fingerstyle" guitar), the distinctions between "high" and "low" music have not had enduring value for me. I've also been aware, for some time, that there are all sorts of things that happen in popular music that I have a hard time understanding. Not necessarily a hard time playing, or a hard time listening to, but a hard time developing a comprehensive intellectual framework for. (Note that what is at issue is not whether a given piece of popular music is somehow "intellectual," but whether there exists a context for extended discourse about the piece.) This situation is all the more frustrating because I'm intimately aware of quite comprehensive intellectual frameworks—what I call, from the perspective I've developed elsewhere, theories of music—for music that is arguably more involved than a lot of popular music. Which leads me toward an interest in issues related to the analysis of popular music.
The situation with the analysis of popular music is all the more curious in that popular music has been a topic in the academy for close on to thirty years now. However, most of the discussion has been around the music. It is as if one talked about the social and historical circumstances of music in Vienna in the late 18th century without ever considering the music seriously, such that Mozart's Don Giovanni became the functional equivalent of, say, Giuseppi Gazzaniga's. This is not to say that, in some respects, Mozart's and Gazzaniga's operas might not be equally interesting. Nonetheless, there is something fundamentally dishonest or naive in this approach. If one really considers the resources of musical composition that the social and historical conditions of late 18th century Vienna made available to musicians living at the time, then it seems quite clear that Mozart was able to exploit these resources with a level of technical and artistic accomplishment that had no equal. To be unable to capture this fact would indeed be a grave deficiency in any truly comprehensive account of this period. It seems, however, that this is exactly where we've been with regard to popular music. We now have some pretty sophisticated theories about the social, cultural, and historical circumstances under which this repertoire was produced, but are still limited in our ability to characterize why some of these products are considered more valuable—both by learned critics and by consumers of popular culture—than others.
I do not believe this situation can be put to rights by a simple application of existing theoretical models to popular music. This is because theoretical models are constructed with reference to what is believed to be valuable about a particular repertoire: no theory is value-free. And to assume that the same things that are valuable in Schubert's "Nacht und Träume" are those that are valuable in Victor Young's "Stella by Starlight" should give us pause, for in the hundred-and-twenty odd years that separate the songs much transpired. It is thus necessary to explore how we might develop analytical models for popular music that will allow us to capture what we find valuable in the repertoire. (That there might be valued attributes in common between highly diverse repertoires seems completely possible to me, but it does seem less likely that, in their diversity, these repertoires would share all their essential attributes.) Which leads, finally, to this seminar.
The first step is to come to terms with how we might characterize the essential attributes of any particular example of popular music. This step we took after break (or at least started to take), when we turned to some examples of popular music brought in by members of the seminar and considered them in light of current theories about the process of categorization (drawing on a chapter of my Conceptualizing Music that discussed two popular songs from the early 20th century in just these terms). We will continue this highly selective and idiosyncratic survey of popular music at our next meeting, and take up the matter of musical value in light of readings from Simon Frith and Carl Dahlhaus.
Meeting 2 (1/10/02)
Assigned material
We continued this meeting where we left off, with a consideration of ways to model the essential attributes of various examples of popular music: from Taiwan, from the Police and the movie Moulin Rouge, and from early '90s hip-hop. Our discussions were as wide-ranging as this music and our intellectual perspectives on it, but there emerged a somewhat more fleshed-out account of my notion of a conceptual model (understood as a correlated list of weighted attributes used to make judgments about a particular instance of popular music) and a distinction between conceptual models used for relatively straightforward characterizations of pieces (which I'll call "ontological models"), and conceptual models used for analysis (which, curiously, I'll call "analytical models"). I argued that the two are related, but that the latter is often synthetic: dispersed attributes within the ontological model—for instance, attributes that characterize rhythmic cycle, repeating patterns, and instrumentation—may be re-organized and recombined in an analytical model—perhaps under the attribute "groove." Re-organized in this fashion, they are then available for further analysis, and can be placed in correspondence with similar phenomena from other pieces. The analytical model, then, is an "act of the imagination" (to use a phrase developed in a slightly different connection by the sociologist Roger Brown) that, under the best conditions, has some fidelity to the ontological model but that is not simply a replication of that model.
In the latter portion of our meeting we turned to the first two chapters of Frith's Performing Rites. Our discussion was perhaps more abbreviated than was desirable, but what emerged was Frith's argument for the importance of taking popular discrimination seriously (Chapter 1) and for viewing the classical, folk, and pop music worlds in comparative terms, as contrasting solutions to shared problems (Chapter 2). For my part (and building on the perspective suggested by Carl Dahlhaus), I further suggested that matters of discrimination were key to the possibilities for musical analysis: a theory adequate to popular music must reflect what is valued within that music.
Meeting 3 (1/17/02)
Assigned material
Our points of departure were three more popular songs brought in by members of the seminar. Leonard Bernstein's "Somewhere," from West Side Story, raised a host of interesting questions, not the least of which was its liminal status as both "art song" and "pop song." Additional issues were those raised by the distinctive pitch structure of the tune, details of the orchestrated version we listened to in seminar, and the relationship between affect and pitch materials. A tune by the late Jim Croce led to a rather more wide-ranging discussion about the canonization consequent to the deaths of some popular artists, the appropriation of certain aspects of "folk" music for "pop" music of the early 1970s, and a few of the subtle details that can vivify a traditional—even conservative—AABA song form. But perhaps our lengthiest discussion was of Britney Spears's "Oops, I Did it Again." Indubitably pop music, the tune (or perhaps the persona known as "Britney") was also revealed to be very complex both in the semiotic field(s) it set up and in the way it was put together. Let it never be said that we are shy of a challenge.
After break we turned to Morgan Lewis's "How High the Moon." Discussion began with some trenchant observations about the hypermetric structure of the piece (which yields effects of both compression and elision in the middle of the tune), its melodic structure, and aspects of its harmonic structure. I then went out on a limb—perhaps very far out on a limb—and proposed that at least some conventional accounts of harmonic syntax (accounts developed primarily with reference to 18th and 19th century music) did not give a satisfactory explanation of how popular songs like "How High the Moon" realized (rather than simply assumed) the conceit of tonality. (I take this conceit to be a broadly syntactic one: that is, its central thrust is not that we hear a piece as being in a particular key, but that we can understand the pitch material of a sequence of musical phenomena as organized around a particular tonal center, even if this tonal center is different from the one the piece began in.) I then suggested that sequential structures—which are pervasive in this repertoire—might be a way to realize this broad notion of tonality. The music is still resolutely harmonic, but the harmonic structure represents not an independent syntax but a reinforcement of the melodic-sequential syntax, which is taken to be primary. (In its strong form this is, of course, a reversal of the perspective on the function of sequence in 18th and 19th century music.) Clearly this is a proposal that demands further consideration (or perhaps radical revision), something we shall undertake with our next meeting when we take up the deferred "All the Things You Are" and "Stella by Starlight."
Meeting 4 (1/24/02)
Assigned material
We began with a discussion of Allen Forte's analysis of Jerome Kern's "All the Things Your Are." In general, there was enthusiasm for many features of Forte's reading of the song, but also a fair amount of discussion about divergences between the way members of the seminar heard and/or construed the song and Forte's analysis. An argument was made for the importance of the lyrics, and also for alternate hearings of the last section of the chorus (after the bridge), hearings in part based on the way this section is commonly harmonized in performances by jazz musicians (who have indeed kept the tune in circulation). Building off some of my comments from last time, I suggested a somewhat altered analytical paradigm that would take into account the importance of the following features of popular songs of this type:
- the frame presented by the rhythmic structure, which shapes expectations about what will happen in the realm of pitch
- the opening interval or gesture, which often combines with a melodic phrase aimed toward a long note (something that can be seen in both "How High the Moon" and "All the Things You Are")
- the importance of melodic sequence to the basic syntax of the song as a whole
- the use of circle-of-fifth harmonizations to give a sense that harmonic syntax is orderly, even where connections between circle-of-fifth structures may not be orderly
We then turned to "Stella by Starlight," and some of the analyses presented in the 1997-98 volume of the Annual Review of Jazz Studies. Again, there was a broad appreciation of these analyses, but also some discussion of where they were wanting; together, this gave us a better sense of some of the challenges of accounting for pitch structure in songs such as these.
In the last portion of seminar we brought our attention to bear on chapter 6 of Frith's Performing Rites, where he engages the topic of rhythm. There was much lively discussion—occasioned in part by Frith's fronting the issues of racial and sexual essentialism in his consideration of rhythm in popular music—as well as some thoughts from me about how the physical aspect of rhythm highlighted by Frith could be construed in terms of processes of cross-domain mapping. Frith, for his part, writes
It is, in fact, the rhythm-focused experience of music-in-the-process-of-production that explains the appeal of African-American music and not its supposed "direct" sensuality. The body, that is to say, is engaged with this music in a way that it is not engaged with European musics, but in musical rather than sexual terms. (Performing Rites, p. 140)
I expanded this by suggesting that it is the possibilities for mapping between various sorts of physical action and the musical materials typical of popular music (especially that derived from African-American music) that draw our attention to the "processive" side of music, and that allow us to read this music as sexually charged.
Meeting 5 (1/31/02)
Assigned material
Our discussion of rhythm continued during this meeting, with our focus on Mark Butler's recent article in Music Theory Online on electronic dance music. A topic that came up early and persisted, in some form or other, throughout our consideration of the issues Butler raised was how one determined what the frame of reference should be in a rhythmic context (a frame that might provide the basis for descriptions of a particular rhythm as "syncopated"). I likened the situation to the figure-ground paradox in the visual domain: once you determine what the ground is, you can then determine the figure; but with many patterns you can also flip things around, such that the figure becomes the ground, the ground the figure. Another touchstone was the relationship between physical interactions with rhythm (through dance or other forms of movement) and our conceptualizations of rhythm. As always, I was loath to draw a sharp distinction between the two, and yet this very figure-ground issue suggested that how we think of rhythm can inform how we interact with it.
During the latter portion of our meeting we discussed some of Pete Martens's analyses of Led Zeppelin tunes, with "Rock and Roll," "Kashmir," and "The Ocean" coming in for particular scrutiny. Pete's analyses gave us a very interesting view of these pieces (and a glimpse at issues in this repertoire as a whole), and also led me to suggest the importance of four-measure units in framing the dynamic patterns that rhythmic figures at the level of the measure combine to create. And Led Zeppelin's music pointed quite naturally toward our next topic: the blues, and the transformations this genre underwent over the course of the twentieth century.
Some mention should also be made of Frith's chapter 7, which continues his treatment of rhythm, working largely within the framework Jonathan Kramer set up in his The Time of Music. Frith comments
To grasp the "rhythm" of a piece of music (which is, in the end, to listen to it) means both participating actively in its unfolding and trusting that this unfolding has been (or is being) shaped—that it will lead us somewhere. It is at once a physical and mental process; it involves aesthetic and ethical judgments. By entering this world of "virtual" or "inner" time we effectively (willingly, trustfully) leave the world of "real" time: hence the common experience of music as timeless (the common use of music to achieve the state of timelessness). (Performing Rites, p. 153)
Although there is much worthy of merit in this, I also pointed out that it wasn't necessary to assume that particular kinds of temporal experiences are coded into specific pieces of music (which seems to follow from Kramer's approach). Indeed, certain kinds of music may lend themselves to one or another kind temporal experience, but it is also the case that humans can—and indeed regularly do—impose interpretations of temporal experience on all types of phenomena, as meditation exercises suggest. Rather than look for temporal experience in music, we should look for it where it ultimately obtains: in the ways humans structure their experience of the world.
Meeting 7 (2/14/02)
Assigned material
We began this meeting with a return to the blues, and the approach to the genre that I've developed in my recent work. That approach is motivated by the question of the relationship between syntax and semantics in music, a question analogous to (but different from in some substantive ways) that explored recently by cognitive linguists. My supposition is that the manipulation of syntax is a primary way to construct meaning in music. From this viewpoint the blues (especially in its instrumental guises) offers a relatively straightforward object for study. The genre makes use of comparatively restricted musical materials, and constructing meaning—that is, saying something significant through blues performance—is understood as basic to blues practice. In my own work I've found that the most effective way to describe the manipulation of syntax in the blues for expressive ends is in terms of multiple layers of syntax: that is, syntaxes of harmony, melody, and rhythm (as well as the words in sung blues). Musical events proper to these various syntactic layers interact, and performers can use these interactions to shape the process of constructing meaning specific to their particular goals.
Subsequent to this introduction we turned to Bessie Smith's "Thinking Blues" (discussed by Susan McClary in the second chapter of her Conventional Wisdom, published in 2000). After a brief review of the overall structure of the tune (as revealed by the lyrics) we turned to ways Smith manipulates melodic and (perhaps more importantly) rhythmic syntax for expressive ends in the first stanza. This detailed look at the actual pitch and rhythmic material Smith used then fed back in to a reading of the larger process through which she created meaning by the way she sang this blues.
In the latter portion of our meeting we turned to the Beatles's "Hey Jude," and tried to get a sense of what (if anything) in the musical materials contributed to the status of this tune as a No. 1 hit. In this we were ably assisted by D. Davies and J. Rockwell, who gave us an in-depth look at some of the features of the song and the issues that were relevant in elevating it to the top of the charts. I endeavored to contribute by noting small-scale, but nonetheless significant rhythmic devices in the melody: for instance, in the first part ("Hey Jude. . .") there is a gradual shift in the first four measures away from the downbeat, which is recovered in measures five and six only to slip away again in the final two measures. In the second part of the tune ("And any time you feel the pain. . .") there are almost no melodic notes placed on the beat, which (together with the change in harmonic strategy) tends to set this section in relief and draw attention back to the first part (as the site of a stable beat). These strategies suggest a way to account for the sense of forward motion attendant on the conclusion of the first part, which then leads in to the famous extended coda, and distinctiveness worthy of a top-ten hit.
Meeting 8 (2/21/02)
Assigned material
Our topic for this meeting was the relationship between folk music and popular music, a topic specific to the projects of quite a few of our number. I began by noting connections between folk and popular music (realizing that both terms are deeply problematical), most particularly in their musical materials: both tend toward simplicity in pitch materials, with complex contrapuntal or harmonic structures being relatively rare; both tend toward immediacy rather than abstraction, and so on. As a mark of these connections, there are frequently confusions between the two: Stephen Foster's songs (such as "Oh! Susannah," and "Hard Times Come Again No More") have been adopted by many "folk" musicians, and are often taken as "folk" music. There are, however, differences between these two kinds of music: the folk repertoire is typically part of an integrated set of social practices, in which different kinds of music are associated with different social and cultural circumstances; the popular repertoire is invariably associated with the profession of music, and is thus implicitly part of a market economy. Finally, technology (as two of our readings for this week suggested, and as summoned by everything from movable type to digital recording) also factors into the question of relationships between folk and popular musics, although it is unclear whether this is because it represents a new tradition in opposition to established traditions or because it is associated with the threat of reification concomitant with professionalism.
With these thoughts in mind we turned to Charles Hamm's "The Acculturation of Musical Styles," with particular attention to Hamm's division of musical practice in the first half of the twentieth century (and to some extent, before as well) into popular, country, folk, and African-American music. Although we did not find the division, nor the means by which it was effected, to be satisfactory, it did provide quite a bit of fodder for discussion, engaging as it does issues of class, race, and to some extent even geography. Distinctions between "popular" music and "folk" music seemed more reflective of marketing positions than musicological fact, or were (as D. Davies put it) self-fashionings intended to establish a pragmatic identity (with the pragmatic aspect returning once again to the matter of music as commodity). What seemed more promising was construing these distinctions as relating to ways of making music, or "styles" as we called them. Thus "folk" music is, perhaps, more properly thought of as a style of performance in which non-electric (i.e., acoustic) instruments are prominent, in which the lyrics tell a story and are often predominant, where diatonic or modal pitch structures predominate, and in which rhythmic structure is generally simple. Whatever the final merits of this view, it is quite useful for developing an analytical approach to the music, since the variables it engages—pitch, instrumentation, rhythm, and the like—are common currency in musical analyses.
After a short break our own Aibhlín Dillane shared some of her thoughts on how the essential features of Irish music have been characterized, and on how these were realized in performances with her group Anish. We listened to and discussed Anish's recording of the reel "Star of Munster," and were able to get a much better sense of how traditional materials could be framed within non-traditional contexts. Reversing this, we then revisted "Hey Jude" through De Danann's transformation of the tune into something that conformed significantly with a "folk style" of performance (taking the tune away from its "popular style" origins). We finally considered Solas's recording of "The Grey Selchie" (a tune with Shetland origins), with attention to the possibilities for musical expression that the band exploited through the kind of tight musical arrangement typical of popular music. Again, there remains much to be said about relationships between folk and popular musics; what is perhaps most interesting is that, for the sake of clarity and concision in argument, much of what needs to be said has to be grounded in specific features of specific musical performances.
