Analysis and Performance Studies

As a sub-discipline of modern music theory and analysis, analysis and performance studies, or simply “performance studies” as it has occasionally been called by some authors (Dunsby, 1997, 12), is only about forty years old. Although the pioneering contributions of Roger Sessions (Sessions, 1950) have been noted particularly by Edward T. Cone, Erwin Stein’s seminal study of form and phrase structure as they relate to performance (Stein, 1962) is more frequently cited as the foundational study of the discipline in its modern form. Stein’s declaration that “discrepancies between performances are often a matter of different emphases, but often they are caused by misconceptions” (Stein, 1962, 12) ushered in an era of prescriptive writing on the part of music theorists eager to diagnose and cure the performer’s ‘malady’ – making “sound without sense” (13).

Stein identified issues of tempo, dynamics and articulation as the most pressing interpretive concerns of the performer, and declared that, while “structural considerations cannot guarantee a good performance,” they can “help to avoid a faulty one” (21). For Stein, these structural considerations primarily involved form and phrase structure – how music unfolds at different temporal levels in balanced, proportional segments. Though not explicitly stated as central to his book, his implicit emphasis on temporality and structural levels in music set the stage for the next two decades of research in analysis and performance. In his 1968 book Musical Form and Musical Performance (a near duplicate of the title of Stein’s book), Edward T. Cone combined the study of temporality and structural levels, introducing the concept of “hypermeter,” a deeper level of meter that groups measures together into “hypermeasures” with a single “structural downbeat.” While many readers may be more familiar with the concept of hypermeter from the later writings of Schenkerian theorists such as Carl Schachter and William Rothstein, for Cone hypermeter was integrally linked to performance, since “valid performance depends primarily on the perception and communication of the rhythmic life of a composition” (Cone, 1968, 38). Continuing the prescriptive trend begun by Stein, he advocated “accentual adjustment” to bring out the groupings revealed by hypermetric analysis.

Unlike Stein, who, as reported by Benjamin Britten in his foreword, enjoyed a career as a German opera conductor (Stein, 1962, 7) and therefore intended his book to cover “all types of music, vocal as well as instrumental” (19), Cone separated the study of instrumental and vocal music, writing a second book focused more on the latter that addressed “persona” and dramatic narrative in place of hypermeter (Cone, 1974). In addition, in both books he selected the majority of his musical examples from the common-practice period, with a particular emphasis on solo keyboard works and orchestral literature in the earlier book.[1] This narrow focus, reinforced by a vast majority of the subsequent contributors to the field, prompted a group of subsequent authors to offer studies that dealt with alternative repertoires, including twentieth-century chamber music (Hermann, 1981; Folio, 1993) and orchestral repertoire (Wintle, 1982), vocal music (Rolf and Marvin, 1990; Rosenwald, 1993; Stein and Spillman, 1996), solo instrumental works (Lester, 1999) and popular music (Cook, 1996).

Cone’s work proved influential in that it served as both touchstone and foil for Wallace Berry’s 1989 book, Musical Structure and Performance, one of the most widely known and controversial works in the field. In his review of Cone’s 1968 book (Berry, 1971), Berry criticized the lack of specific recommendations to the performer regarding dynamics, tempo and articulation (271) and worried that Cone’s higher rhythms, particularly the “terminal downbeat” may lead to “insensitive mechanical articulation” (282). In a subsequent article (Berry, 1988) and in his book, he addressed many of the issues that had preoccupied Stein – an emphasis on tempo and articulation as interpretive tools, a plea for restraint with regard to artistic license, the privileging of form – but went a great deal further in detailing what “interpretive interventions” the performer(s) should undertake to highlight the musical structures revealed by analysis.

Musical Structure and Performance remains one of the most ambitious analysis and performance studies to date, yet its impact will likely be measured by the negative reactions it provoked, some of which accelerated the offshoot of entirely new branches within the discipline. The midpoint of the 1980s had brought the simultaneous publication of three important articles: Cone’s affirmation and further development of hypermeter (Cone, 1985), Janet Schmalfeldt’s dialogue between a performer and an analyst regarding issues of interpretation (Schmalfeldt, 1985), and David Epstein’s application of cognitive science to the study of tempo (Epstein, 1985). While Cone’s study demonstrated the continuing relevance of hypermeter for performance (cf. Burkhart, 1994), Schmalfeldt and Epstein’s articles encouraged the growth of performance analysis (i.e., analysis by / for performers) and analysis of performance itself, respectively.

The fortuitous dialogical format chosen by Schmalfeldt for her article was a thinly veiled reference to her ongoing personal struggle to reconcile the two sides of her own musical persona, and the honesty and integrity with which she pursued that reconciliation – not to mention the keen observations rendered by the pursuit – gradually inspired subsequent authors to abandon the stern, imperative prescriptions of Stein and Berry in favor of an equal consideration of the insights performers and analysts produce through their musical studies. New authors began to advocate a ‘separate but equal’ philosophy, whereby “performer’s analysis” (Rink 1990, 323; Cone, 1994, 734) was associated with specific, local problem solving (Dunsby, 1989, 8) and the means for conveying a given interpretation were left to the performer (Lester, 1992, 79). An effort was made to problematize the dichotomy between intuition and reason (Rink, 1990, 320; Kerman, 1996, 95), and compromise approaches such as “informed intuition” (Rink, 1990, 323) were proposed to replace imposed prescriptions by the analyst. Although some reviewers felt that Jonathan Dunsby’s 1995 book Performing Music: Shared Concerns, a primary example of the field’s desire to be more inclusive of performers, wound up being a jack-of-all-trades and master of none (Bradshaw, 1995; Rink, 1996), the recent publication of a literal collaborative dialogue between analysts and performers (Clarke, 2005) testifies to the enduring power of Schmalfeldt’s attempted leveling of the playing field.

Given the discipline’s newfound respect for performers’ insights, some authors began to develop the idea of performance itself as analysis in a different medium (Rink, 1990, 328; Lester, 1992, 78), leading to the notion of analyzing performances for the light they might be able to shed on the printed score, and not vice versa (Clarke, 1991; Rink, 1995; Dunsby, 1997). The publication of an entire volume dedicated to the analysis of performance (Campbell, 1996), as well as two important studies by a leading music theorist (Cook, 1999 and Cook, 2001) confirmed the arrival of this new branch of analysis and performance studies, which continues to be one of the fastest-growing areas of the discipline.

When David Epstein, one of the first to advocate the analysis of recorded performance, published Shaping Time: Music, the Brain, and Performance (Epstein, 1995), the temporary equality between performers and analysts was placed in jeopardy. Although he did analyze recordings, he was criticized for returning to the prescriptive, uni-directional (analysis – performance) model (Rothstein, 1998, 126) and for circular reasoning and an inaccessible prose style (Adlington, 1997, 159–60).

And where were the Schenkerians in all of this? As noted by two of his reviewers (Folio and Larson, 1991), Berry’s approach is quasi-Schenkerian in its reliance upon structural levels, yet most Schenkerian analysts (Burkhart, 1983; Larson, 1983; Beach, 1987) preferred returning to Schenker’s own, rather limited pronouncements on the topic of performance (Rothstein, 1994; Schenker, 2000), thereby temporarily ceding the development of new analytical methods to music psychology and cognition. Analysis and performance studies continues to grow: the Society for Music Theory, a flagship professional society, sanctioned the creation of a special interest group devoted to the topic in 2004, and the number of conference proposals and presentations related to performance studies is steadily increasing every year. Who knows what the next forty years might hold?