- Proposal Format: PDF file listing the paper title, proposal (up to 500 words), and up to two pages of supplemental material (examples, bibliography, etc.). The author’s name should not be included in the body or metadata of the file.
- Submission Deadline: Sunday, May 28, 2017 at midnight (ET)
- Submission Email: Email your proposal as an attachment to Edward [dot] Klorman [at] McGill [dot] ca. Use subject line “PAIG Proposal.” In the body of the email, please list the author’s name, affiliation, email, and the title of the paper.
By Edward Klorman (McGill University)
Author’s note: This blog post is adapted from Chapter 6 of my book Mozart’s Music of Friends: Social Interplay in the Chamber Works (Cambridge University Press, 2016).
When examining a metrically ambiguous passage, many musical analysts will seek to untangle problems by explaining which among multiple metrical interpretations is “the correct one.” In this blog post, I explore metrical conflicts from a different point of view—namely, that of the performers.
Musicians often strive to find ways to play better together, to match their phrasing and shaping in order to “feel” the music in a unified way. But what happens when they encounter a passage in which their parts don’t match, but are in conflict? The subordinate theme from Brahms’s Sonata in E♭ Major for Piano and Clarinet, op. 120, no. 2 is a case in point.
You can get a sense of the issue—and imagine how you might approach it in performance—from this video. (Since my clarinet skills are a little rusty, I perform on viola instead!)
To provide some context for the subordinate theme, let’s first look briefly at the main theme and transition. The sonata opens with fairly regular four-bar hypermeter, as I’ve shown in my markings above the staves.
Some metrical dissonance arises in mm. 15–17, which tend to sound more like 3/2 meter than the notated 4/4. Within these four bars, the piano’s left hand emphasizes the off beats, playing on the weak quarters of each half-note pulse. The four-bar hypermeter is restored in mm. 18–21, but the piano’s syncopations continue to emphasize weak quarter notes. (These syncopations foreshadow an ambiguity I will discuss in the subordinate theme.)
In m. 21, the final bar of the transition, the diminuendo hairpin and the rest on beat 4 encourage the players to use some rubato, allowing the sound and momentum to dissipate as the passage comes to rest on a tentative German augmented-sixth chord in B♭ major (a harmony that yearns to resolve to V in the new key, the traditional cadential goal of an expositional transition).
And then what? After the general pause, the clarinet begins the subordinate theme on the notated downbeat, with the piano imitating one beat later.
The Subordinate Theme in Brahms’s Notation
This imitation constitutes a canon per arsin et thesin, whereby strong beats in one part coincide with weak beats in the other. Coming right after a general pause, this passage would challenge many listeners to determine whether it is the clarinet’s entrance or the piano’s that falls on the “true” downbeat (if the listener is not following the score).
Indeed, for the first several bars of the theme, it seems plausible that the bar lines might be located one quarter note later than how Brahms notated them, as shown below. Only by m. 28 (best seen above, in the print of Brahms’s score) is the notated bar line unequivocally confirmed: the clarinet’s eighth-note upbeat clearly establishes the downbeat of m. 28, and moreover it is at this juncture that the piano gives up the metrically ambiguating canon. Simply put, my alternative notation below seems plausible initially at m. 22 (making the placement of bar lines difficult to discern for several measures), but by m. 28 it is absurd.
Subordinate Theme in a Plausible Alternative Notation
Why do I hear this passage as so metrically ambiguous? And how should musicians approach their performance of such a metrical conflict? I will explain my own understanding and performance approach:
After the German augmented-sixth chord and general pause in m. 21, the clarinet’s entrance on concert F may initially seem to represent dominant harmony, bolstering the sense that it is an upbeat to the piano’s tonic entrance. But when I play the passage (on viola), I try to make the entrance in a metrically neutral way, stressing neither note of the (concert) F–F gesture. Observing Brahms’s sotto voce marking and emulating a clarinet’s capacity for niente entrances helps to achieve this. Once the piano enters in canon, the two players might emphasize their conflicting meters with a gentle persistence—that is, with the piano emphasizing the long slurs beginning on beat 2 (and thus suggesting my alternate notation). Brahms’s beams across bar lines encourage the pianist to count in terms of my alternate notation. This performance approach heightens the conflict and suspends the sense of metrical ambiguity as long as possible (ideally until around m. 28), thus taking advantage of a special opportunity afforded by Brahms’s passage.
* * *
The approach I’ve suggested here is consistent with performance advice proffered in eighteenth-century performance manuals, such as those by Leopold Mozart, Johann Joachim Quantz, and Daniel Gottlob Türk. These authors encourage performers to emphasize beginnings of slurs, especially those that are syncopated against the meter (as in the piano part here). Brahms was certainly familiar with these ideas and harbored a highly traditional understanding of the meanings of slurs, one that is particularly relevant in a passage such as this. (On these eighteenth-century writers’ understanding of slurs and their execution, see my discussion in Mozart’s Music of Friends. On Brahms’s affiliation with older understandings about slurs, see Heinrich Schenker’s essay “Abolish the phrasing slur” in The Masterwork in Music, I ).
I’ll close with a perspective on performing metrical conflicts from an author more nearly contemporary to Brahms (and certainly one who felt a strong affinity to him), Heinrich Schenker:
It is the responsibility of the performer primarily to express the special characteristics of a composition, as they sometimes coincide with the meter, sometimes oppose it. Today, not only the failure to recognize rhythmic relationships but also sheer indolence creates a performance for the metric scheme alone—a dismaying evidence of decline (Free Composition, translated by Ernst Oster, emphasis added).
Leaving Schenker’s hortatory tone aside, he offers an idea well worth pondering. In everyday life, we strive to avoid conflicts or else to resolve them. But in playing music, our impulse to resolve conflicts should be questioned. A conflict-resolving performance strategy might conclude: “Brahms clearly intended the meter as he notated it, and this is confirmed by m. 28. Therefore, we should perform in accordance with the notated meter, minimizing signals that rub against it.” But a more satisfying approach, at least to me, is to allow the conflict to flourish and for the two musicians to each make an earnest bid for the meter as suggested by their own parts. The clarinetist’s bid for the notated meter ultimately prevails, of course. But our final-state understanding of the metrical structure is, like history, determined by the victor, and only after the conflict has ended.
by Richard Beaudoin (The Royal Academy of Music, London and Brandeis University)
Glenn Gould performed and recorded on an increasingly rickety, loose-jointed, swaying piano chair that his father, Bert Gould, had fashioned from a folding bridge chair in 1953. Its telltale clicking/cracking noises — made whenever Gould shifted his body — can be heard in this 1966 film of Gould discussing Schoenberg’s music with Yehudi Menuhin. Below are two photographs of the chair, in all its skeletal beauty:
I encountered Gould’s chair creaks up-close while microtiming his recording of Schoenberg’s Op. 19/1, to use as the basis for a series of new compositions called New York Mikrophon. I found the trajectory of Gould’s chair creaks intriguing, and I began to investigate.
My approach, which extends Paul Sanden’s writings on “corporeal liveness, considers Gould’s so-called “ambient” or “extraneous” chair noises as significant “sounded movements.” Quantifying sounds that are normally marginalized, my research connects sound studies, music theory, and performance analysis, and fuses published analyses about rhythm and meter in Schoenberg’s composition with the audio artifacts of Gould’s corporeality.
The elements of my analysis can be seen below:
Using the Lucerne Audio Recording Analyser [LARA], I made millisecond-level measurements of the number and location of all of the sound events — including each chair creak — in Gould’s recording of Schoenberg’s Op. 19/1, made in September 1965, at 30th Street Studios in New York.
This recording is very noisy, even by Gould’s standards; there are 85 creaks in this 86-second track. I noticed that the creaks were not spread uniformly across the performance, but, instead, followed a rough trajectory (see the image below). In addition, prominent gaps in the creaking were not easily explained.
Rather than propose my own analysis (which might be seen as self-serving to the microtiming data I had collected), I surveyed the published scholarship on rhythm and meter in Op. 19/1. This led me to publications by Jonathan Kramer, Charles Morrison, and John Roeder. Kramer devotes a chapter of The Time of Music to Schoenberg’s 17-measure work, and singled it out as exhibiting “the emergence of a foregrounded meter” (Kramer 1988). His tracing of the work’s metric evolution is outlined in the three score-based examples below:
I set about comparing the location the chair creaks in Gould’s recording with Kramer’s chart of the emerging metric hierarchy in Schoenberg’s work.
Within this specific recording of this specific work, the level of Gould’s body motion that is transferred to the chair correlates to the gradual emergence of a metrical hierarchy: as Schoenberg’s written meter becomes the sounding meter, Gould’s physical shifting largely abates. The image below presents the first stage of my findings; I presented more detailed analyses in a paper given at SMT/AMS Vancouver 2016, as part of the session called Performing Meter.
You can listen to how Gould’s creaking relates to Kramer’s metrical observations here.
The microtiming also turned up some wonderful ‘hidden’ details in the recording, including:
- Gould re-attacking a tied note on the downbeat of measure 8 (the tie exists in Schoenberg’s manuscript and all printed scores), which affects the perceived metric clarity,
- a peculiar gap in the creaking that corresponds precisely to measure 7, which Kramer highlights as the clearest meter thus far in the piece,
- an unusual bit of vocalizing in measure 2, in which Gould sings a motive which is not simultaneously occurring in the piano (as was his common practice), but which instead occurs a few moments later — a kind of subtle, improvised vorimitation.
I’ll write more about these in the future, as I prepare a paper on this research.
I don’t, however, take these findings to be general proof of what pianists do when playing metrically irregular music. Nor do I use them to hypothesize about body movements by pianists, twentieth-century pianists, mid-twentieth-century Canadian pianists, or Glenn Gould in his mid-30s, etc.
Rather, what interests me is how all of the sounds captured by the microphone in New York in September 1965 work together to create a distinct impression of this unique piece. And in this single case, the proliferation and exact placement of the creaks made by Gould’s chair does, subtly, guide the mind across a trajectory of movement that is sympathetic with a recognized structural aspect of Schoenberg’s piece. In this way, they are perhaps analogous to microexpressions.
My work — in some ways the opposite of corpus analysis —involves detailed cataloguing of all of the sounds within a single recording. Doing so, I unearth little signals provided by overlooked, over-heard, so-called extraneous noises. Removing or suppressing such “insignificant” sounds — a common practice in the recording industry — deprives the listener of unique types of intimacy and musical understanding.
by Victoria Tzotzkova (Harvard University)
“The dualism between subject and object disappears…. The world as this ‘cybernetics’ constructs it is a monism.… Because the worlds are coupled, they must in the last analysis be regarded as a single system.”
—N. Katherine Hayles, How We Became Posthuman
The cybernetic principle—the idea of a self-changing feedback loop, where each part is continually reinvented through its interaction with all others—has been at the core of different technological advances, and is also at the core of theoretical work in different fields. In some of its unfoldings (like the theory of enaction developed by Francisco Varela and his collaborators), the basic idea of a dynamic, generative, recursive interaction, linking the constituent parts of a process has served as a powerful framework for theorizing the fundamental relationship between an organism and its environment. Stemming from concerns in regulating automatic processes, research in cybernetics has been developed in much larger contexts, posing some profound philosophical challenges, but also engaging diverse communities, and opening avenues for research, design, and experimentation.
For all its promise and power as a model for conceptualizing different aspects of human existence, as well as for creating tools and technologies, the cybernetic principle can bear deeply unsettling implications: once in motion, the process continues on a path that cannot be reliably predicted at the outset; a self-changing process shapes itself as it goes along.
In regards to relating—or even integrating—performance and analysis, can we usefully consider this idea of a dynamic interaction of mutually engendering parts of a holistic process? What would it mean for performance and analysis to interact, such that each continually shapes the other? What would exploring the relationship between performance and analysis look like if we tried to imagine it with the idea of cybernetics in mind?
I. A personal turning point relating analysis and performance
Growing up in then still communist Bulgaria, I started formal music training at a very young age, a study that included solfège and basic theory. In my hours of practice, still as a child, I developed the habit of doodling around my pieces, going, for instance, between themes from different Haydn sonatas I was playing. I got chastised for resorting to a bit of that same practice during a concert, and, being a dutiful student, I stopped it.
Years later I began to think my doodling may have been a good idea, but I was so completely out of practice, that it was pretty much impossible to reclaim. I had become a typical, non-improvising, classical performer. Yet some years later (and after a few encounters with French traditions in keyboard harmony), I gradually reclaimed my doodling abilities, and the path was through theory: the countless harmonic dictations and simple tonal progressions that had been drilled into me as a child turned out to be a perfect beginning for a course in re-learning how to improvise. I now practice my doodles, and from a performer’s standpoint, find it to be the best anti-anxiety routine I know, as well as a great way to build up my classical performer’s sense of agency. In this sort of practice routine, studying harmony or motivic gestures, devising practice exercises, or building up my performer’s familiarity with a piece are aspects of the same process: a sort of continual feedback, merging performance and analysis.
II. A brief overview of a movement
The Performance and Analysis movement (and the Performance and Analysis Interest Group [PAIG] of the Society for Music Theory) arose at least partly in response to a perceived divide between performance and analysis, which many felt did not capture their musical experience. In recent decades, the movement has expanded the focus of music analysis, from the score (predominant in traditional analysis) to also include recordings, embodied aspects of performance, and performers’ voices.
In British musicology—as well as elsewhere in Europe—performance studies and artistic-/ practice-based research are becoming a relatively significant part of the study of music, which has given impetus for emergent institutions specifically devoted to such activity (e.g., The Orpheus Institute, Belgium), as well as projects specifically conceived to cross between university and conservatory settings (e.g., Centre for Music Performance as Creative Practice, UK). This trend has resulted in events like masterclasses in artistic research (e.g., by Paulo de Assis), the Performance Studies Network conferences, as well as research positions for practicing musicians and research studies involving conservatory and professional musicians. The Centre for Performance Science (Royal College of Music, London) is also a catalyst for a related move to develop the more explicitly scientific study of performers and their performances.
Yet, most research in this vein—which explicitly involves or addresses practicing musicians—comes typically from musicologists, psychologists, empiricists, etc., and rarely from music theorists. Even those of us in music theory who specifically work on performance would generally agree that, by and large, our discipline has not yet found ways to embrace performance as an integral concern.
III. Carving out spaces for dialogue: The PAIG special session
What stops music theory from fully embracing performance? Are we looking to address performance and performers through beliefs and practices that do not facilitate a fluid exchange? If so, do we do this consciously, or out of habits we have not yet clearly examined? What might these habits be and can they be revised in fruitful ways, to build on cherished skills and approaches while inviting performers’ insight and participation?
One of the aims of the PAIG-sponsored special session at last year’s SMT conference in Vancouver was precisely to address questions like these from an unusual angle, namely, by giving the floor to performers. The three presentations of the session were given by performers—Patrick Boyle (jazz trumpet, joined by his trio), John Lutterman (baroque cello), and Charles Neidich (clarinet)—with a response from British empirical musicologist and performance scholar Eric Clarke. Each of the presenters addressed concerns that emerge in practice and through the experience of performance, and spent considerable portion of their time playing their instruments. One might say that these particular performers were all of a special kind: those who have already developed their performance concerns into research programs. But they still spoke as performers, and developed their presentations from that standpoint. They each brought to the session their love for the music they play, as well as their enthusiasm for engaging with their scholarly audience.
Whatever might be keeping us—as scholars—from fully embracing music performance as an area of study, hearing out performers’ voices might point some ways out. In listening to performers talk, we undoubtedly hear some ideas and expressions that are not entirely native to music theory. But we also recognize many of the insights as deeply musical and also deeply familiar. So we might try listening specifically with an ear to finding the common vocabulary and common commitments, and use these points of contact as points of departure, as a springboard from which we can reflect back on what we do as music theorists, and on the sorts of things music theory can be.
If we let them, performers might take us into directions we have not already taken. These directions would be as of yet unknown, and following them would be to embark on a journey with no preexisting roadmap. By the same token, in talking to us, performers themselves—like the presenters at the PAIG special session—might approach, experience, and perhaps even do performance a little differently.
IV. “How should music theory change to accommodate this kind of work?”
I was asked the question above at my dissertation defense and am forever grateful for it. It opened the possibility that music theory is a fluid and evolving endeavor, and that the concerns and commitments I have by virtue of being a pianist may not only one day be included; they may even shape the concerns and commitments of the discipline.
If we do engage performers and find ways to build and maintain live ties with the people and activities within music performance, there may be changes to what we know as music theory and what we do as music theorists. Similarly, a tight interaction with performers would also potentially bring changes to what we know as performance and what we (or they, performers) do as performers. An ongoing (cybernetic) interaction implies both.
For me—as both performer and theorist, or better yet, as a musician—these changes are exciting and welcome.
Su Yin Mak (The Chinese University of Hong Kong) has graciously allowed us to post the video of the paper she delivered this past Fall at SMT in Vancouver. The paper was presented in a session chaired by PAIG member Alan Dodson called, appropriately enough, “Performance and Analysis,” alongside PAIG co-chair Andrew Friedman’s paper “Reimagining (Motivic) Analysis in Light of Performance.”
You can find the video here, and the abstract reproduced below. Su Yin writes: “I am currently planning a follow-up project with collaborators based in Japan and the United States that focuses on the uses of metaphorical description in rehearsal communication by professional string quartets, and would greatly appreciate comments and suggestions from PAIG members.”
Abstract: Structural models for Western art music are primarily score-based and rarely incorporate the views of performers. I have attempted to redress the omission through a multi-phase study of rehearsal discourse by professional string quartets based in Hong Kong, China, Japan and the United States. This paper presents the findings from the Hong Kong phase of the project. Over a six-month period, I attended and recorded the Romer String Quartet’s rehearsals and public performances as a participant-observer. Quantitative and qualitative analysis of the rehearsal footage, along with interviews with the players, offer insights on how a professional string quartet perceive, conceptualize and communicate about musical structure. My research reveals that although parameters such as formal articulations, harmonic changes and motivic continuity were rarely singled out for discussion in rehearsals, the players did pay close attention to structure within the context of feeling and character or in relation to considerations of sound and ensemble co-ordination. While the quartet’s communication relied extensively on metaphorical descriptions rather than music-theoretical terminology, when asked to explain the meaning of their metaphors the players referred to very specific aspects of compositional syntax. Thus, in its combination of overt expressive considerations and latent structural understanding, the rehearsal discourse suggests that the relationship between the two is more complex and less exclusive than some have assumed. These observations prompt critical reflection on ways of mediating between theoretical and practical perspectives of musical structure, and demonstrate how methodological interactions between theory and ethnomusicology might contribute to such mediation.
by Daniel Barolsky (Beloit College)
At this year’s business meeting for PAIG at SMT/AMS in Vancouver, we ended with the relatively mundane task of discussing and voting on a new set of bylaws. One of the sub-topics that I had expected to be routine was the issue of interest-group membership. Much to my surprise and dismay, there were voices that expressed concern about our openness and proposed hypotheticals that some collection of individuals (especially those from organizations other than SMT) might decide to colonize the interest group. Happily the final decision was to define membership by the listserv, a collection of email addresses to which anyone, regardless of institutional or disciplinary affiliation, can join.
However jocular the suggestion about “colonization” or more serious the proposal to police membership, I couldn’t help reflect, as I left Vancouver, on the larger intellectual, disciplinary, and political context of this conversation. Not five hours after the business meeting, the AMS convened a session on “Race, Ethnicity, and the Profession” that made clear to everyone how the restriction of methodologies did more than merely help define a discipline but, rather, served to shut out individuals and identities. Less than a week before the US election, I couldn’t help but see parallels between the desire to “other” members of different musical societies (including their subjects and approaches) and the rhetoric of political candidates who espoused building walls to keep out people of certain faiths and ethnicities. And, finally, I reflected on the history of the Performance and Analysis Interest Group, an inclusive and welcoming organization that over a decade ago gave me, a graduate student in musicology who was interested in performers and recordings, a scholarly, intellectual, and professional home at a time when the majority within the AMS and SMT excluded methodologies that addressed the voices of performers and their performances. The question we need to grapple with, as we continue to revise and develop our bylaws and, in particular, our mission statement, is what would restricting our membership protect? What does it really mean to limit ourselves to theorists only?
As I flew home to a polarized and fractious political climate, I also took comfort in the excellent and discipline-bending session organized by PAIG that took place the evening before the business meeting (as well as in the other papers/sessions during the weekend that explored a range of issues pertaining to performance and analysis). In particular I took hope from the comments by the session moderator, Victoria Tzotzkova, who urged the audience, while listening to the presenters (none of whom were exclusively music theorists, some of them not even card-carrying members), to expect new voices, sounds, and perspectives, and to consider in what new and inclusive ways we could imagine music theory to be.
by Nathan Pell (The Graduate Center, CUNY)
The starting point for my post is a lyric from the theme song from the ‘70s sitcom Happy Days, which I’m sure was formative for many of us: “it feels so right, it can’t be wrong.” I’d like to apply this to matters of performance and analysis, in a way I hope The Fonz would approve of.
As performers, we spend much of our time trying to do things cleanly. So we practice tricky passages to get the notes fluent and even, and we strive constantly towards good phrasing, intonation, ensemble, diction, etc. These things are generally construed as “right,” their opposites “wrong.” Performances that adhere more or less to these standards are generally considered ideal. Those that epitomize them are virtuosic, especially if carried off with ease.
But I’d like to focus on performances that, by such standards, are generally deemed wrong: the ham-fisted, out-of-tune, misplaced, and erratic (either accidental or intentional). Funnily enough, these bloopers are frequently my favorite part of a performance, and I will sometimes miss the messy version when I hear a “clean” rendition of the same passage. At the heart of my “desert island” collection of recordings are many that feature faux pas in some way or another.
In other words, a performance can have moments that are “so wrong, they’re right.” I consider this a sort of extravagance. Of course it’s not how the passage “should go”; but the mistakes humanize the performance, and there’s a certain brilliance to the unorthodoxy that I savor. I would not be surprised if many of us have our own personal lists of similarly cherished, extravagant moments. I’ll share some of mine, and will then relate these comments to analysis.
The great pianist Edwin Fischer was famous for his wrong notes. His landmark recording with Wilhelm Furtwängler of the Beethoven “Emperor” Concerto is littered with them. They handle the transition from the Adagio to the Rondo with particularly exquisite subtlety. Less subtle, however, is the beginning of the Allegro, where Fischer’s left hand misses the mark a couple of times. I do not for a second wish that he played the right notes: the mistake (along with some heavy pedaling!) accords perfectly with the finale’s romping, jubilant character.
Furtwängler has long been my favorite conductor, despite his immensely complicated status as a political figure. His performances revel in spontaneity of all sorts, including some jarring mistakes. But I consider many of these endearing: for example, the fourth horn’s woeful intonation in the Ninth Symphony’s notorious solo. I find the effect rather nice, actually. The reduced forces here remind me of the village wind band in the Scherzo of the Sixth Symphony: inept, but very earnest. There is something touching about the halting performance, especially in context of the pious simplicity of this movement.
Speaking of the Sixth Symphony, there is a wrong note in Furtwängler’s recording with the Vienna Philiharmonic: in the finale, a first violinist loudly plays an E♭ where F is written. I have always enjoyed this mistake because I like to imagine that the player is hearing the E♭ as leading towards the subdominant (of course, not what Beethoven wrote!).
Outright mistakes are not the only way a performance can be “so wrong, it’s right.” Indeed, performance allows all sorts of heterodoxies. My favorite performers are often those who are not afraid to do something extravagant and “wrong” in the service of the music. The effect is usually not subtle, for this kind of transgression results from the performers intentionally flouting performance traditions.
In a performance of Don Giovanni’s Commendatore scene from the 1960 Aix-en-Provence Festival, Rolando Panerai sings Leporello’s triplets late, surely on purpose. The contrast between the resolute Don and Commendatore on the one hand and the hysterical Leporello on the other is immensely effective, and, having heard these late triplets, I cannot be completely satisfied with any on-time rendition. Later in the scene, Leporello begs Don Giovanni to heed the Commendatore’s calls of repentance. The Don refuses and is soon engulfed in flames for his sins—making Leporello’s “sì, sì” the very last interaction between servant and master. Panerai’s performance hardly resembles what Mozart wrote, but remains tremendously evocative.
As far as I’m concerned, there is only one recording that makes sense out of the finale of Bruckner’s Sixth Symphony: Furtwängler’s with the Berlin Philharmonic. The moment I like best is an unconventional, gleefully irreverent trombone slide in a fortissimo passage (it exists in no score). To me, it is completely in character: “so wrong, it’s right.”
The famous Boccherini Minuet is certainly a light piece, and as such admits a great many liberties in performance. (It is fascinating to consider the question of what makes an extravagant performance “work” in one case but not in another—this would require far more space than this post allows me. Suffice it to say that it must be very subjective, but that there are probably some moments that call for extravagance more than others.) Enter the idiosyncratic Russian cellist Daniil Shafran, known for his wide and fast vibrato (which my cello teacher once mimed by pretending to stick his finger into an electric wall socket). The vibrato is on full display in his performance of the Boccherini, but it’s hardly the most bizarre thing going on here! Of particular note are Shafran’s pecking bow articulations and very distinctive rubato (especially in pesante passages), for which his harpsichordist appears ill-prepared. Now consider the performance by nineteenth-century piano virtuoso Francis Planté, widely considered one of the finest players of his day, who was captured on record at the age of 89. His début, 80 years earlier, had been attended by Chopin. It is astounding, therefore, to hear his recording and to think: yes, people really did play like this! The left-hand accompaniment is quite loud, and the melody whirs along until the grinding halt of each cadence, whose “plunk, plunk, plunks” sound comical to modern ears. But a more “correct,” more virtuous performance sounds so dry in comparison!—even if it is not lacking in character, and its execution is blameless by any of the usual musical criteria.
* * *
All those who practice the art of analysis are well aware of its greatest danger: over-analyzing. We live in constant fear of explaining the special-ness out of music, of making it conform too snugly to a theory. In reality, we know that music is always messy, irreducible, and illogical; but we must make sense out of it, reduce it, make it look cleaner. In this way, the impulse towards “cleanliness” and “right execution” is as strong in analysis as it is in performance.
But analysts can counter this impulse in the same ways as performers. Sometimes the best way of serving a piece is to produce a “wrong” analysis, one where the extravagant idiosyncrasies of the music outweigh any obligations towards the decorum of good praxis.
What does an “extravagant analysis” look like? A famous example is Schenker’s graph of the C major Prelude from WTC I. Despite its canonic status, it is not a “clean” graph. In fact, it contains a sizeable contradiction!
Note how D4 is transferred up one octave to D5 (in red), creating a prolongational span. The saw-toothed beam that connects E4 to C5 (blue) indicates another prolongational span. But these two spans clash, representing two different harmonies! The contradiction is irreconcilable, and any introductory analysis student would be marked off for it: even more shocking when one considers that the Urlinie-Tafeln were designed to be teaching resources! His objectively “wrong” analysis is nevertheless of true value, though, because it captures something important about the work that a “correct” reading would surely miss.
Direct contradictions are only one sort of way analysis can be “wrong.” Just as performance does, analysis admits a veritable menagerie of extravagant possibilities. One can see a sort of extravagance in James Hepokoski and Warren Darcy’s view of Mozart’s G minor String Quintet, K. 516: the second theme begins in the tonic (!) after an extremely unusual type of “medial caesura” (“I:PAC MC”). Rather than forcing the piece into their theory, they chose instead to accommodate its remarkable divergences. This might make for difficult theory, but yields a sensitive analysis. Extravagance can take the form of an adventurous analysis; in Beethoven’s ‘Tempest’ Sonata, William Caplin locates the final cadence of the exposition at a point where “a literal tonic bass is not present.” This is not a shy decision, but it is clearly made with a deep concern for expressing the work’s idiosyncrasies. There is a stylistic “wrong-ness” in set-class analyses of ostensibly “tonal” music: David Lewin analyzes WTC I’s F♯ minor Fugue as centered around the set class (013), and Milton Babbitt prefers to describe the “Tristan chord” as (0268). Their extravagant analytic approach allows them to reveal latent musical relationships.
No matter whether one agrees or disagrees with the above decisions, I hope the expressive power of their extravagance is readily apparent.
As many of us attend SMT, a conference largely devoted to analysis, let’s remember the power of extravagance in music, and the dangers of over-correctness. Our discipline places so much emphasis on the perfectly tidy theory, the airtight argument, the flawlessly consistent analysis—and rightly so! But even as we cultivate these virtues, let’s not forget to keep a place for the slightly off-kilter, the imperfect and “wrong,” the undotted i’s and uncrossed t’s—in brief, the wonderful messiness of art.