2. Theoretical Matters

Music performance is a rich, multi-dimensional phenomenon that has fascinated philosophers, historians, analysts, psychologists, cognitive and neuro-scientists as well as anthropologists and cultural theorists. People have studied it from various angles and disciplines arriving at important partial insights. In this chapter I review (necessarily very selectively) some of the key developments in this broad field leading to my proposition that music performance is too complex to be understood by any one approach. We need multi-modal and transdisciplinary, comprehensive accounts that are data-driven yet embrace the phenomenological and cultural if we wish to lessen the problem of verbalizing an embodied aural experience. Ultimately I argue that music performance is a complex dynamical system; as such it requires a robust and dynamic investigative approach. Gilles Deleuze’s theory of difference and consistency may just provide the necessary theoretical framework and toolbox of terms.

The last twenty years or so has seen an exponential growth in academic studies of musical performance. Previously this area of humanistic musicology was largely limited to investigating historical performing practices of the seventeenth and eighteenth centuries or earlier periods of western literate culture. At the same time, it was maintained that modern-day performances of later repertoires represented an unbroken tradition. Renowned composers and performers of the nineteenth century would hand down their understanding of stylistic requirements to their pupils in conservatoires or private studios, who in turn passed this tradition on to the next generation, in a continuous flow. Subsequent generations constantly interpreted the opinions and insights of past masters—composers, performers and teachers—while holding them to be gospel. The availability of increased quantities of written evidence from the 1800s onwards, including more detailed notated instructions in scores, as well as letters, memoirs, press reports and concert reviews, together with the relatively small, insular, hierarchical and authoritative nature of the world of classical music (“this is how Beethoven goes because the maestro says so, and he knows because he was the student of the student of the student of Czerny who was a student of Beethoven”) created the impression of an ongoing, living tradition. The master–apprentice model and prestige of educational institutions ensured that change was slow and seemingly imperceptible. Thus emerged what the musicological literature tends to call “mainstream performance” (MSP) style. This is the style that most experimental music psychologists have been studying since the early 1980s, though systematic investigations began in earnest with Seashore and his group at the University of Iowa in the 1930s.1 However, recent musicological interest in sound recordings has started to provide important counter-evidence that questions the existence of such an unbroken tradition and supports studies that map a different history.2

Meanwhile, generations of twentieth-century musicians and music historians dedicated to the rediscovery of historical performing practices of music composed prior to 1800 (roughly speaking) followed a very different path during the last 70-100 years. Instead of learning their craft from master teachers, they turned to written sources and surviving old instruments. From sporadic, individual endeavours at the end of the nineteenth century and during the first half of the twentieth century (e.g. Edward Dannreuter, Wanda Landowska, Carl Dolmetsch, Ralph Kirkpatrick) the early music movement, as it was first labelled, gradually grew in popularity during the 1950s and 1960s and, by the late 1970s and early 1980s, had developed into the immensely popular authenticity or period instrument movement. There are many readily available accounts of this development.3 What concerns me here is what has happened since; however, to explain my views properly, I have to take a detour.

As I showed in my earlier book on the performance history of Bach’s works in the twentieth century, this phase of the early music or authenticity movement had developed rather differently in Europe than in England. Contrasting customs, economic circumstances, cultural practices and musical personalities led to different musical results. During the 1970s and early 1980s the recorded performances of leading Continental musicians and ensembles (e.g. Leonhardt, Harnoncourt, Concentus Musicus Wien) had a very different sound and style of playing compared to their UK-based colleagues (e.g. Hogwood, Academy of Ancient Music). This difference was largely ignored by the most important English language criticisms and evaluations of the movement that appeared between 1982 and 1988 and thereafter.4 Ever since Taruskin’s immensely influential essay, The Pastness of the Present and the Presence of the Past (1988) that crowned this growing criticism, historical musicologists have looked at the twentieth-century trajectory of music performing styles in a different light and the movement itself gained a new name: historically informed performance (HIP). However, as I will outline below and throughout this book, the differences between performers and groups specializing in HIP have often continued to be ignored. This enabled generalizations that parroted Taruskin’s claim that HIP is a modern invention fitting modernist aesthetics to become entrenched in spite of several important refinements of his articulation being voiced in subsequent publications. As this issue is a crucial inspiration for my endeavour to show the importance of detailed and systematic data gathering, analysis and reporting, I will start my discussion of music performance literature with this key debate.

2.1. Cultural Theories

HIP and Modernism

Focusing primarily on British musicians, Taruskin pointed out essential similarities between mainstream and historically informed performing styles and linked both to modernist ideals. He showed that the tendency for rigid tempos, technical accuracy and literal reading of scores was common in both styles of playing and reflected general cultural trends that aimed to eschew the personal, the passionate, the subjective in favour of a distanced stance that paraded as objective, technocratic and scientifically based. An approach that subdues the performer’s role into a neutral, impersonal vehicle or narrowly functioning mediator-transmitter that simply allows the music to “speak for itself” is not, he argued, historically accurate at all, but reflective of our own age and preferences: music-making that aims for clarity, precision and economy of language is modernist. The sleek technical proficiency and clockwork-like style of playing may be compared to the smooth lines and functional design of modernist architecture à la Le Corbusier and show closer links to Stravinsky’s and Toscanini’s aesthetics than to the documented sensibilities of any composer-performer of the seventeenth or eighteenth centuries.

What Taruskin did not mention in his 1988 essay, apart from a short discussion towards the end, was the fact that not all musicians active in HIP at the time played like that, especially not on the Continent. There a different style was gradually emerging that allowed a greater role for the old instruments to guide technique and more time for experimentation, and to put into practice what historical treatises and instrumental tutors seemed to be recommending. As I have shown in my previous book, initially musicians of Leonhardt’s and Harnoncourt’s circles presented no exception to the modernist approach. They too went through the phase of the depersonalized, no-expression style of playing, but much earlier.5 Leonhardt reflected on this “new objectivity” period, when he said “Oh well, the neue Sachlickeit period […] We had to strip down our playing to bare essentials in order to find and bring up something different, something new that may be closer to how it was back then centuries ago.”6 However, by the later 1960s-early 1970s they had had their debates about the utopian nature of historical authenticity—a discussion that happened in the English-speaking world only well into the 1980s—and had created a radically different style of playing, rich in novel means of expression.7

This new style has since been described and codified by various musicologists and developed into what is currently regarded as the historically informed way of performing music from the seventeenth and eighteenth centuries. It is important to remember, however, that not before it was available in sound did performer-researchers manage to really grasp its essence and find a way to write about it meaningfully.8 It is also essential to stress that what we currently regard HIP is likely to be quite different from what performances actually sounded like during the eighteenth century. Historical accuracy is not really the issue here. We will never know for sure. What matters is the existence during the second half of the twentieth century of two distinct styles of playing baroque music; one identified as MSP, the other as HIP. I shall explain the differences, as I see them, throughout this book.

Contrary to the style critiqued by Taruskin, this version of HIP is orientated towards creating musical gestures; to make the music “speak” as a good orator would using the conventions and classical art of rhetoric. This approach and aesthetics foster rhythmic flexibility and localized rubato, dynamic inflections, free ornamentation and other expressive liberties. Although a radically different sound is thus created, this is of course no evidence for it being similar to how music-making originally sounded in the baroque period. Without surviving sound objects, aurally perceivable artefacts, it remains impossible to know. Taruskin’s argument that the HIP style is reflective of our own time still stands. But is it really just a twentieth-century invention, reflective of modernist aesthetic? Does it not have any historical grounding? Should we not need a fresh look at what is happening now instead of mindlessly continuing to repeat Taruskin’s finding about practices typical of the 1970s and 1980s in certain commercially successful circles?

Surely I am not the first to ask such questions. Music performance studies have grown enormously during the past two decades and several publications corroborated Taruskin’s basic claims while refining the argument. In relation to historically informed contemporary baroque performance, the most important contribution is John Butt’s monograph, Playing with History (2002) and the most often cited “defence” is the late Bruce Haynes’ The End of Early Music (2007). Butt’s discussion places a premium on linking the HIP movement to broader post World War II cultural developments; Haynes is more concerned with detailing stylistic differences and defending the ideology and historical validity of the HIP approach.

As I stated earlier, I am not at all interested in the debate about whether any performance of the Six Solos is more historically accurate than any other. The lack of sonic evidence from Bach’s time and circle makes this a moot point. Rather, I wish to provide an account of developments in baroque performance practice since the 1980s and explore how Taruskin’s and Butt’s positions may be applied or expanded. Therefore I will first investigate further the potential links between contemporary performance practice and broader cultural and aesthetic trends.

Modernism versus Postmodernism

In performance studies terms like modernism and postmodernism are often used rather loosely or only implied by way of reference to qualities and modes of thinking associated with broadly held modern or postmodern viewpoints and aesthetics. In a simple formulation the modernist stance pursues absolutes, believing in the possibility of an objective (scientific) knowledge of a dispassionately observed world. It tends to put a premium on abstraction; it tends to be logocentric, authoritarian, and hierarchical. Modernist art tends to be austere or ascetic to the degree of being unpleasant or incomprehensible. Its promotion of the cult of genius and originality goes hand in hand with denigrating conventional attitudes: the “philistines” and products that cater to such tastes.9 The postmodern stance rejects this and values relativism and pluralism instead. Postmodernism is linked to post-structuralism and deconstruction, which are “often presented in anti-scientific terminology that stresses the proliferation of meaning, the breaking down of existing hierarchies, the shortcomings of logic, and the failures of analytical approaches.”10 It emphasizes the importance of culture and its products, holding entertainment and the decorative in the same elevated esteem that the modernists would prefer to reserve for “High Art.”

However, what modernism is or is not is a complicated matter and has been argued and reviewed by many, much better qualified to do so than I. It is common to posit that modernism stems from the emergence of self-consciousness made explicit by Descartes’s famous “cogito ergo sum” and the Enlightenment’s attempts at empiricist rationalism (or indeed from the radically changed sensibility that emerged with the Renaissance, Humanism and Copernicus’ discovery). Others link its beginning to Baudelaire’s mid-nineteenth century definition of modernité as the new aesthetic. Modernism’s various strands mean that it also entails aspects of romanticism and the cult of the individual.11 Complications regarding definitions and periodization show clearly in literary studies, for instance. They seem to form two camps: those who see modernism as late romanticism, and those who see it as anti-romanticism. Both are probably right; a situation that reflects modernism’s fraught relation with its past.

In music studies, modernism is more closely related to compositional style; the break-down of tonality, the rejection of melody, the cultivation of mathematically derived material, and the eschewing of the emotional, the expressive. In relation to music performance, and perhaps mirroring debates in literature, Butt distinguishes between romantic-modernism and classical-modernism, the latter aligning more closely with what Taruskin describes as typical of modernist performance style.12 The romantic-modernist (Taruskin calls them “vitalists”13) tends to read the musical score with the modernist aesthetic of seriousness and reverence granted to deserving artefacts representing “High Art”—for instance slow tempi, monumental sound, intensely felt and expressed melody lines in performances of Bach’s Passions as on recordings conducted by Karl Richter, or Wilhelm Furtwängler’s Fifth Brandenburg Concerto recording much discussed by Taruskin,14 or, to remain closer to repertoire under examination here, Itzhak Perlman or Oscar Shumsky’s recordings of the Violin Solos. The classical-modernist approach is dispassionate, matter-of fact, clear, precise, sleek and smooth; “modern” in the most general and commonly understood sense of the word—for instance Shlomo Mintz’s version of the Solos, or the “sportive” Brandenburg Concerto recordings of the Boyd Neal Orchestra, the Busch Chamber Orchestra or those conducted by Karl Münchinger as well as most of Taruskin’s examples.

Similarly, commentators’ discussion of what is or is not postmodern varies greatly. There are authors who regard postmodernism to be essentially a continuation of modernism, “except that confidence in […] reason has been abandoned.”15 Fredric Jameson identifies “a moment of late modernism” and posits that for “the late modernists themselves, what is often called postmodernism or postmodernity will simply document yet a further internal break and the production of yet another, even later, still essentially modernist moment.”16

To me this resonates well with the view that at times Taruskin labels as modernist certain attitudes manifest in some HIP recordings that may also be regarded as postmodernist. In his richly textured argument, Taruskin highlights the aiming for novelty and variety evident in many HIP projects and interprets this as a sign of modernist aesthetic. In discussing Christopher Hogwood and the Academy of Ancient Music’s 1985 recording of Bach’s Brandenburg Concertos, he criticises their decision to replace the famous sixty-five-bar harpsichord cadenza at the end of the first movement of the fifth concerto with a mere nineteen-bar solo found in secondary eighteenth-century sources. Taruskin recalls the arch-modernist Stravinsky’s “strictures [...] about the ‘seduction of variety’” and notes the commercial ploy of the recording being “billed […] as the Urfassung, the original version of the set, bringing with it a promise of hitherto unprecedented ‘authenticity.’” He argues that “the elevation of what amounts to a rejected draft to the status of a viable alternative—and even a preferable one—because it is earlier […] and less demanding on the listeners” amounts to “immoderate irony”; non-reverence for the canon and devaluing “both the work and the critical sensibility that impelled its revision” by Bach. Taruskin concludes, “By being rendered so much less impressive […] Hogwood’s Bach is rendered correspondingly more modern.”17 Maybe so, but perhaps the diversification, the commercial orientation, the de-canonization, the simplification, the “frivolization” if you will, could also be signs of postmodernism.

Taruskin’s explanation of this potential paradox is provided on the next page, where he writes:

The same critics who can be counted upon predictably to tout the latterday representatives of High Modernism of music—Carter, Xenakis, Boulez—and also who stand ready zealously to defend them against the vulgarian incursions of various so-called postmodernist trends, are the very ones most intransigently committed […] to the use of “original instruments” and the rest of “historical” paraphernalia. For we have become prevaricators and no longer call novelty by its right name.18

One reason for the paradoxical situation may be that postmodernism also thrives on novelty, increasingly so. In a fully commercialized and commoditized art world, novelty is essential although can be retro in style.19

Whether postmodernity is regarded a continuation of or a break with modernity, most authors define the terms in relation to each other. Jean-François Lyotard, in the classic text on the subject, explains that “scientific knowledge” and the modernist quest for it, “legitimates itself with reference to a metadiscourse […] making an explicit appeal to some grand narrative.” He posits that postmodernism can therefore be defined as “incredulity toward metanarratives” even though “[t]his incredulity is undoubtedly a product of progress in the sciences”20—another reason perhaps for seeing postmodernism as simply another stage in modernism. Jameson describes postmodernism by listing those features of modernism that it rejects, including authoritarianism and “the non-pleasurable demands made on the audience or public.” He also notes the “refusal of concepts of self-consciousness, reflexivity, irony or self-reference in the postmodern aesthetic and also in postmodern values.”21

How might such “incredulity toward [historical] metanarratives” manifest in music performance and its research? In terms of current performing styles of Bach’s music one might look for signs of an acceptance that it is impossible to accurately recreate bygone historical styles or to gauge long-dead composers’ intentions. A rejection of the authority of sources—manifest in increasing tempo and rhythmic flexibility, ornamentation, improvisation and transcriptions / arrangements—may also be indicative of “incredulity toward the metanarrative” of canonical composers, “fetishized” versions of scores and “etalon” performances. The postmodern tendency to discard attitudes that made “non-pleasurable demands” on audiences is also demonstrable in current performances of baroque music. It is evidenced in the revival of more sensuous sounds, less rigid tempi, less geometric meter, more modulated tone and varied use of vibrato, and more general freedom and flexibility to bring expressive qualities to the fore. In short, postmodern aesthetics—defined as “incredulity toward scientific knowledge” and “elevation of the decorative”—is seen in the increasing number of recordings that dare to “interpret” rather than just “dispassionately transmit” pieces; that show violinists being more personally invested in the performance, experimenting and playing with the music.

HIP as a Mirror of Cultural Change

Comparing performance aesthetics to modernist or postmodernist thinking seems fruitful up to a certain point. There are obvious similarities between the two: the modernist-scientific search for absolute truths and the concern for the composer’s intention; the establishing of correct scores (Urtext), execution, instruments, ensemble size and constitution; the searching for performance rules; the revering of great masters and their texts / scores; the pursuit of technical mastery, machine-like reliability and evenness, and so on. On the other hand, parallels can also be drawn between the postmodern focus on the “obsolescence of the metanarrative apparatus”22 or the “overwhelming failure of the rationalist project”23 and recent cross-over and arrangement projects such as Uri Caine’s jazz versions of Bach, Beethoven and Mahler, The Bad Plus arrangement of Stravinsky’s The Rite of Spring or the liberal approach to Bach’s or Vivaldi’s pieces (among others) by the British ensemble, Red Priest, to name but a few. Musicians’ publicly stated opinions convey a decidedly loosened verbal rhetoric regarding the possibility of ever knowing the composer’s precise intentions and the subjectivity of these intentions. Artists and researchers alike have started to emphasize the role of the perceiver (whether performer, listener, or reader), and the inadequacy of musical notation to convey how the music may sound.24

There are more covert analogies as well. Jameson notes that “classical modernism was an oppositional art; it emerged […] as scandalous and offensive […] subversive within the established order.”25 This could also be said of the early music movement of the 1950s to 1970s. Back then it represented an alternative performance tradition, often belittled and ridiculed by the academic and musical mainstream,26 which became more and more standardized and aggrandized by commodification through the recording industry and the promotion of stars.27 Jameson suggests dating the emergence of postmodernism to the early 1960s, when “the position of high modernism and its dominant aesthetics [became] established in the academy and [were] henceforth felt to be academic by a whole new generation of poets, painters and musicians.”28 In terms of music history this was the time when modernist composers, from the safety of university posts and state subsidies, shrugged at the lack of audience support, and when a positivistic outlook gave impetus to philological and analytical studies and the rise in prestige of musicology and music theory.29 With a delay of a decade or two (i.e. by the end of 1980s or later) the once radical art of HIP had also become established in pockets of Western Europe (e.g. The Netherlands) and in British and North American higher education institutions. So much so that it was common in the 1980s to jokingly refer to the “Early Music Police,” an umbrella term for purists who would insist on the use of period instruments and would sanction certain solutions and practices while vehemently condemning others as inauthentic.

The long-standing debate regarding the performance of dotted rhythms is a good example to illustrate the point. Two of the key contributors to twentieth-century scholarship on seventeenth- and eighteenth-century performance practice, Robert Donington and Frederick Neumann, were at loggerheads for decades about the “correct” delivery of dotted rhythms. Many others, such as David Fuller, Graham Pont, Matthew Dirst and Stephen Hefling, have joined in, generating innumerable articles, books and responses to each other examining and interpreting the same sources.30 A more recent example is the debate regarding the size of Bach’s ensembles, especially the vocal forces in his Passions, Cantatas and the Mass in B minor.31 The modernist appeal to scientific truth and confidence in reason are clearly in evidence in these writings, making it is easy to agree with Taruskin and regard HIP as just another manifestation of musical modernism. However, noting the “immense weight of seventy or eighty years of classical modernism,” Jameson explains an aspect of the postmodern condition that also clearly resonates with the HIP project:

[I]n a world in which stylistic innovation is no longer possible, all that is left is to imitate dead styles, to speak through the masks and with the voices of the styles in the imaginary museum.32

So is HIP modern or postmodern? Does it matter? One way to reconcile the seeming contradictions is to examine the stylistic aspects of performance more closely. The differences may point to problems with dates: There is a discrepancy between the emergence of the broader manifestations of both modernism and postmodernism in general and the appearance of these attributes in performances of western art music. As far as twentieth-century music performance history is concerned, a change of style is more evident around the 1930s and then again since the late 1980s, rather than around the 1890s and the 1960s. The performances from the period between ca. 1930 and 1985 are what Taruskin originally labelled “modernist,” using the term “authentistic” for those that identified with the HIP movement. But is Taruskin’s verdict still valid? What has happened since 1988? What is the current style of playing? The next three chapters will explore these questions in great depth by examining the chosen violinists’ views and affiliations as well as analysing the performance features that characterise their respective interpretative styles.

For now we can note that already in the 1990s Michelle Dulak wrote about a mellowing of rhetoric regarding the possibility of recovering past practices or indeed the composer’s intentions.33 John Butt agreed while also aiming to explain the cultural roots of this new relativism:

As soon as it becomes acceptable to dislike what Bach might have done it is easier to allow [choices]. […] [H]istorical evidence can be treated critically, and one can acknowledge that there is no absolute distinction between the choice of personal insight—or opinion—and historical accuracy. […] If postmodernism means a more liberated attitude toward historical evidence, a less guilty (and more conscious) inclination to follow one’s own intuitions, then there are certainly more postmodern performers around than there were ten years ago.34

It is my intention to show how or to what extent pluralism and relativism conquered Bach performance practice at the turn of twentieth and twenty-first centuries; whether parallels can be found between performance styles and broader cultural trends. Butt has already paved the way. In this book I focus on subsequent decades to investigate further how performance practice issues have developed since Taruskin’s critique and Butt’s revisionist reading. It must be kept in mind, though, that this book only deals with recorded performances and only of Bach’s Six Solos for Violin. How valid my findings might be for other repertoires or what one may hear on concert platforms across the world is a task for another day!

Aesthetics and Value Judgment: Beauty and the Sublime

Trying to apply cultural theory to music performance, although potentially illuminating, also has limitations. It is worth looking for alternative explanations as well. Additionally, ideologically I tend to find myself agreeing with those who are frustrated by the relativism of postmodernism, “its overemphasis on reflexivity, its maddening efforts to write texts that do not carry any risk of presence.”35 Yet I much prefer performances that are unique, personally engaged, flexible, inventive and “interventionist” even to a degree when the interpretation considerably de-familiarizes or “recomposes” the piece. Music performance always carries risk and should not be impersonal. But is it not the modernist style that is criticised for lacking a presence, the performer’s personal conviction and “authenticity”?36 What about stylistic requirements? If prescriptive performance analysis necessarily leads to normative and thus authoritarian and absolutist-modernist expectations as Taruskin posits, then how are we to decide the quality of a performance?37

Jameson touches on this problem when he reviews the philosophy of aesthetics and its distinction between the sublime and the beautiful. He notes that “[m]odernism aspires to the Sublime as to its very essence, which we may call trans-aesthetic, insofar as it lays claims to the Absolute, that is, it believes that in order to be art at all, art must be something beyond art.”38 This is very much the standpoint of many mainstream musicians, who emphasize the timelessness of Bach’s music and reject the importance of historical practices when performing his works in the twenty-first century. Adorno’s oft quoted critique of the 1920s organ movement and bourgeoning music festivals that “reduce” Bach to Telemann is a famous quip stemming from such a modernist aesthetic.39 Itzhak Perlman’s contempt for HIP approaches to Bach’s violin works, or indeed for using period instruments, is another.40

Taruskin also discusses the difference between beauty and sublimity.41 In his view, the confusion started with Eduard Hanslick (1825-1904), who considered beauty “the legitimate domain of art,” the attribute that appeals to our intellect and invites contemplation.42 In contrast, as Taruskin shows, during the eighteenth-century, writers associated not the beautiful but the sublime with “boldness and grandeur,” with “the Pathetic, or the power of raising the passions to a violent and even enthusiastic degree.”43 Taruskin quotes Edmund Burke, who decreed: “Sublime objects are vast in their dimensions, beautiful ones comparatively small; beauty should be smooth and polished; the great is rugged and negligent […] beauty should be light and delicate; the great ought to be solid and even massive.”44 Taruskin then explains how, in the nineteenth century, the sublime encroached “upon the domain of the beautiful—of the ‘great’ upon the pleasant” to an extent that was “profoundly repugnant to the early generation of modernists” who wanted “art […] to be full clarity, high noon of the intellect.”45 Herein lies the source of Hanslick’s conflation of beauty and the sublime and the elevation of beauty to the centre of aesthetic contemplation. The smooth and polished, impeccably precise playing and the evenly vibrated, homogenous tone of contemporary mainstream musicians are the enactment of such a conception of beauty and art. In modernist performance there is no room for ruggedness or negligence. Rough sounds, risk taking, boldness are attributes one hears in recent recordings of select HIP and cross-over musicians; these, as I will show, blur the divide between the modern and the postmodern yet again.

This imperceptible yet confusing swap between the sublime and the beautiful underlies the difficulty in distinguishing between the modern and the postmodern in music performance. According to Jameson it was not until the 1960s that we came to the “end of the Sublime, the dissolution of art’s vocation to reach the absolute.”46 Thus began a celebration of the decorative and the return to seeking beauty; art as pleasure for the senses, which, incidentally, was very much the view of music’s function during Charles Burney’s (1726-1814) time.47 The new-found expressivity and playfulness of HIP that have been evolving on the Continent since the late 1960s and have become much more widespread since the late 1990s reflect this postmodern (as well as late eighteenth-century) sensibility: the geometrical, literal, smooth, modernist playing started to give way to a more engaged, flexible, interventionist, and reflexive interpretative style. What seems unique about this development is that while baroque music performance is apparently becoming more “decorative” (note the added embellishments, greater rhythmic freedom, etc.), it is also reclaiming boldness and depth through varied bowing and articulation, stronger pulse and a broader spectrum of dynamics, among others. While giving up on dogmas and rules, it is again becoming capable of “raising the passions,” to be daring and expressive; in letting go of absolutes, it is finding a way to resurrect the sublime.

It may be fruitful, then, to reverse the question of our investigation. That is, rather than trying to fit music performance into broad cultural trends, it could be instructive to examine instead if what is happening in music performance may provide cultural theorists with new ways of evaluating the status quo. Nowadays the performance of classical music is a small, peripheral phenomenon in global culture. Yet, through its technical discipline and concern with past repertoires and their aesthetic ideals, it upholds certain seemingly conservative values while managing to readjust and renew; to generate affect and move listeners. It is neither modern nor postmodern, neither absolutist nor relativist, neither scientific nor reflexive, but a never-ending search to get closer to the essence of pieces according to what this means to successive generations of performers and audiences.

2.2. Analytical Theories

The field of music performance studies (as opposed to historical performance practice studies) has generated its own theories. Here I am thinking of analytical and empirical models developed by music theorists, cognitive scientists and performers. The richest literature comes from music psychology and as such is of limited importance here because its questions relate more to psychological processes rather than stylistic considerations. Therefore my consideration of them will be necessarily selective. The main aim of this section is to provide an overview; to introduce different approaches and to discuss a select few recent theoretical propositions starting with musicological work and continuing with experimental and empirical contributions from the cognitive sciences.

Music Performance Studies

In the English-speaking world the study of music performance had originated in music analysis. As Nicholas Cook, among others, has pointed out the significant side-effect of this has been that the “flow of signification [in performance analyses] is from analysis to performance, from text to act” in both analytically and historically informed performance theory.48 Yet changes in performance style tend to precede their theoretical formulation. If Cook’s perception is correct, and reading the literature it seems apt, the situation betrays a lack of true dialogue between leading performers and academic music studies. Or rather, it indicates a one-way flow of innovation and an opposite flow of normative thinking that needs to be highlighted to correct the record.

My earlier comparative review of the development of HIP theory and HIP practice clearly showed performers leading the way and academically argued analysis and style description lagging behind. Key formulations of the essential characteristics of what is called the HIP style started to appear some two decades after the first recordings in which such characteristics can be heard (see fn. 8 above for references).49 While the performance style of Leonhardt and his circle or the Concentus Music Wien attained, by the late 1960s, a markedly different sound through its altered approach to articulation and musical meter, published performance practice research continued to be preoccupied with tabulating and debating the “correct” delivery of ornaments and dotted rhythms or, perhaps more usefully, with making sources available.50 Once the new style of playing had become wide-spread and its practice analysed and explained (primarily during the 1980s), the theory of performance practice (again) quickly became normative and prescriptive rather than descriptive, reinforcing the text to act flow of signification in academic writing while performers have again moved on, mostly unnoticed. This tendency of researchers limits insights into what actually happens in music performance at any particular time and is one of the reasons, I contend, why Taruskin’s verdict regarding HIP being modernist and “authentistic” prevails.

The reality is that in the wake of Taruskin’s criticism during the 1980s, HIP and performance practice research appear to be moving away from the “purist” standpoint. Nowadays many performers play on both period and modern instruments, and specialist groups and soloists emphasize the significance of their own sensibilities and readings of scores rather than the importance of knowing the “rules” and abiding by the fine print of historical sources. This more relaxed and relativist attitude is less typical of analytical and empirical music performance studies. Explicitly or implicitly these seem to hold on to the primacy of the notated text and to the view that the quality of a performance is dependent on how well it reflects or brings out compositional structure.51 Although such publications also seem to have had their apex around the turn of the 1980s and 1990s, the importance of structural signification has proven to be a notion even harder to let go of than the belief in the utopian ideal that the performer’s duty is to recreate exact historical practices and realise the dead composer’s ultimate intentions as represented by critical (Urtext) editions.

One example of these criteria continuing to make their mark is the work of pianist-researcher Julian Hellaby.52 He proposes a “framework” for analyzing performance: a “tower” based on nine “informants” grouped into four “levels”—all neatly fitting the broader principles of being historically and stylistically-analytically accurate, aware of the composition’s structural, genre-specific and affective content, and thus well disposed towards fulfilling the composer’s intentions (Figure 2.1). At the bottom of the tower Hellaby lists historical era and the score, followed by genre and topic (i.e. musical type associated with a particular function) one level up, then topical mode (i.e. performance qualifiers that appeal to imagination or emotion) and characterizers (i.e. distinctive musical features such as rhythmic, melodic or harmonic devices). Finally, on top of the tower, are tempo, duration and sonic manipulators.

Figure 2.1. Hellaby’s Interpretative tower.53

					executive
level four	Tempo	Duration manipulator		Sonic moderator
level three	Topical mode		Characterizer
level two	Genre		Topic
level one	Era (style)		Authorship (score)
					ideal

Although he states that the model represents an essentially interactive hierarchy, Hellaby implies that the top level is the most significant in terms of explaining variability between performances when he asserts: “level four [includes] those informants that engage most directly with a performer’s acoustic formations.”54 The sample analyses (Bach Toccata in D, BWV 912, among other works, each performed by four different pianists) are primarily descriptive, briefly commenting on the nine informants. The results are then summarized in a grid within the tower. Hellaby indicates the relative importance of both the nine informants and the four levels of an analyzed recording by varying the thickness of arrows that highlight the relationships. He thus provides a testable model that seems quite powerful in summarizing basic differences among performances. At the same time, its breadth and ability to show differences in detail (i.e. differences of degree rather than of kind) seems limited. Furthermore, while it aims to provide an objective account of a performance, it remains not just descriptive in its evaluation of the nine informants but also prescriptive regarding what a performer should consider when embarking on an interpretation of a composition. In other words, the model is an example of the “page to stage”55 approach to performance analysis.

This aesthetic seems a natural or logical standpoint in a literate culture where the functions of composing and performing have been separated. Yet the identity of a composed piece is not a simple matter, prompting several philosophers to argue about it.56 Adorno maintained, for instance, that “while neither the score nor the performance is in fact the actual ‘work,’ the score is closer than the performance, suggesting the need to examine the score when evaluating performances.”57 Exploring Adorno’s views, Paddison states:

While at the level of work as score multiple and contradictory readings may coexist as infinite potential performances, at the level of the work in performance, as ‘sounding object,’ no particular realization of the piece can fully meet the contradictory demands of the work as score.58

These contradictory demands of the score speak differently to diverse generations, leading to a variety of performance styles and interpretative approaches.59 When lamenting the “Urtext mentality” of modernist performers, we should keep in mind that earlier generations were often as well-versed in composition as in performance, and they often played more than one instrument (consider the violinist-pianists Fritz Kreisler, Jascha Heifetz and George Enesco, the last named being also noted for his original compositions, not just transcriptions or period pieces published under pseudonyms as in the case of Kreisler or Nathan Milstein and others).60 The renewed interest in improvisation since the turn of the twenty-first century may take performers back to composing as well. Some performers (e.g. Rachel Barton Pine among violinists) are already playing their own cadenzas and preparing transcriptions to expand their instrument’s repertoire.61 Once they feel confident in their command of speaking the musical language of particular eras or composers, performers can take scores as scripts rather than as literal texts. In any case, the fine line between fulfilling the composer’s assumed intentions and maintaining the performer’s prerogative of breathing life into the “work” by creating a “sound object” of it will always have to be negotiated anew, historically as well as culturally. Hellaby’s “tower” is a possible framework for analysing this negotiation.

Rink, writing earlier than Hellaby, proposed a framework for the study of music performance that illuminates the “process of refraction,” which may lead to performances “reflecting personal conviction and individual choice, at the same time demonstrating historical and analytical awareness and a given ‘programming’ (both physical and psychological).”62 The constituent parts of this “process of refraction” are similar to Hellaby’s “informants,” namely “genre, performing history, notational idiosyncrasies, compositional style, structure as ‘shape’ and physicality.” In Rink’s model these inform the “performer’s artistic prerogatives,” which then produce the performance conception.

The similarity of Rink’s and Hellaby’s lists of elements with which a performer should be concerned in developing an interpretation underline the currently accepted boundaries within which musicians of the western classical tradition work. In Rink’s conception, though, there is no preordained hierarchy among the constituents impacting on the performer’s decisions. Presumably each artist’s personal disposition and training dictates the process. As such, this framework seems more open-ended and less prescriptive than Hellaby’s; more accommodating of cultural and historical norms of any given era or performer. It is not a testable model, but it provides the analyst of performances with more room to actually study what performers do without channelling the investigation into an assessment of the “sound object” against predetermined hierarchical criteria. Hierarchical conceptions may easily lead to normative thinking and as such are not far removed from the modernist-absolutist standpoint. Framing performance and its analysis as a “process of refraction” allows for greater plurality, subjectivity and reflexivity. This can be useful if one seeks a diagnosis rather than an evaluation of the state of performance.

Empirical and Psychological Studies of Performance

Perhaps not surprisingly, investigations of music performance by the scientific community reflect even clearer preferences for the modernist approach. This is not necessarily for ideological-philosophical reasons but, as Richard Parncutt argues, because there is not enough time in a researcher’s life to keep up with all the musicological literature and knowledge to develop a deeper understanding of performance history and conventions.63 Also, music psychologists and cognitive scientists tend to look for “general mechanisms” rather than “individual manifestations.”64 For instance, when studying expressiveness in musical performance, they tend to focus on the relationship between the normative (notated) compositional structure and deviations from it in performance. Using scientific, empirical and experimental methods, they measure how performers mark musical units like cadence points or phrase boundaries and study listeners’ perceptual thresholds and aesthetic judgments.65 Once scientists have observed and mapped regularities they develop models for musical performance. In these rule-based systems preferences masquerading as normative “consensus” are programmed in, and even though the authors of algorithms keep fine-tuning the rule parameters, the systems remain reflective of stylistic characteristics common in a particular era (among currently available programs these largely mirror the mainstream aesthetic ideals of the 1950s to 1980s period).66 Such systems remain paradigmatic of the modernist approach, working with abstract forms of meaning (representation) and central control.

But music performance changes constantly, even if this is difficult to detect within shorter periods of time. Musicians, especially competition judges and teachers, often assert their conviction regarding what they believe to be artistic norms—for instance how Beethoven or Mozart “should” sound. Yet music performance changes because taste changes. Instead of taste, Leonhardt speaks of imagination. He explains:

I think that the changes made in the last fifty years are based on the fact that the imagination has changed. […] the crucial element behind the wish of some players not to play Bach on the piano was that they wanted to get rid of that dripping Romanticism they did not like. […] In the last fifty years, we have gradually begun to see that Baroque music is, if anything, more expressive than Romantic music, but in details rather than in large lines. With that, a technique developed, but not by itself; it’s only that the wish has changed, that our imagination of the music has changed completely.67

This ever renewing thinking and musical imagination is one reason why I propose—more fully in the next section—that music performance should be considered a “self-organising process in which meaning is generated through a dynamic process.” In other words, it is not the result of “the passive reflection of an autonomous agent” who authoritatively decides how the performance should go.68 Expressiveness is a feature of performance that can establish an artist’s personal authenticity; it is a performer’s way of interacting with and responding interpretatively to musical stimuli according to her or his cultural-educational background and musical persona.69 What may be a convincing gesture in one performance may sound alien in another; what one generation finds appealing and “true,” may be rejected by another as uninformed or false. This can be demonstrated by studying listeners’ reactions to old recordings or subsequent published reviews of earlier performances re-issued first as long playing record and then as compact disk. It is also evidenced in diverse generations’ emphatic assertion that they are “serving the composer’s intention” while playing completely differently from each-other.

The constraints a performer brings to what may be heard as appropriate expression has been confirmed scientifically as well. Through a set of experiments Renee Timmers showed that “choices at the beginning of a phrase provide constraints and expectations for the rest of the phrase […] if ornaments are chosen to be performed long, this may set the trend for future ornaments.”70 The musical context, the quality of the measured gesture and the historically-culturally defined aesthetic expectations are crucial and provide the basis of the infinite variety and creativity manifest in music performance. The various mathematical models developed so far tend to neglect these more intuitive interactions between a performer and a piece of music.71 Although some computer controlled piano performances sound very successful (because they come close to current aesthetic criteria), they lack dynamism: the potential for creativity and renewal.

Nicholas Cook also advocates a more complex approach that embraces ethnographic-cultural as well as empirical-experimental methods. He takes his cue from Philip Auslander, starting with the premise that “to be a musician […] is to perform an identity in a social realm.” This approach, Cook claims, “involves not a predefined meaning reproduced in performance but rather a meaning that emerges from performance.”72 It assures the foregrounding of the performer and his or her “relationships to audiences.”73 Referring to Robert Philip’s work, Cook too notes the abyss between the quantifiable and the qualitative, how “empirical analysis can leave behind essential aspects of the phenomenon” it studies.74 For instance, Philip claims that although an accelerando can be measured, to a listener it “can seem impulsive, or uncontrolled, it can seem to be aiming precisely at a target, or to be dangerously wild. It can seem spontaneous or calculated.”75 In other words, the meaning of the accelerando is not deciphered by acoustic-physical measurements; the same measured value may be perceived to create varied and even contrasting affects, it may move the listener in diverse ways while different listeners may also hear the same performance differently.76 Similarly, performers regularly note how they adjust their playing according to the venue or the instrument at hand (and various other “externals”).77 The solution Cook proposes is to adopt a broadly ethnographic approach, with the performer and analyst taking part in a participant-observational dialogue or the analyst ensuring that empirical data is contextualized adequately through detailed engagement with cultural, biographical, historical and other available sources.78 Cook concludes by asserting

Performance […] is an indefinitely multi-layered and complex phenomenon, the multiple aspects of which demand multiple analytical perspectives. […] [T]he approach of interdisciplinary performance studies helps to clarify what performances mean, while more empirical approaches help to clarify how performances mean what they mean.79

2.3. Music Performance and Complex Systems

Given the considerable achievements of the various approaches reviewed so far, it seems useful to try to develop an understanding that may assist in moving away from the “either / or” divide both in terms of cultural theory (modernist versus postmodernist) and also in terms of approach (empirical or analytical, ethnographic or experimental). Both the study of music performance and its object (i.e. performance) are multi-faceted. Acknowledging that music performance is a complex phenomenon, indeed a complex system, fosters the “and / as well as” outlook; the possibility of multiple characteristics and groupings. An understanding of complex systems can be developed by finding the simplest aspects and building on those; or by working with the complexity. Cook has argued we need both and I agree. However, the former approach has been privileged long enough and therefore I would like to shift the attention, without becoming dogmatic, to the latter. As Cilliers states:

Instead of looking for a simple discourse that can unify all forms of knowledge, we have to cope with a multiplicity of discourses, many different language games—all of which are determined locally, not legitimated externally.80

Such a standpoint may prove productive for a study of the current landscape of Bach performance practice: the differences between MSP and HIP, but especially the varied strands within these broad categories readily lend themselves to explication once it is accepted that “[d]ifferent institutions and different contexts produce different narratives which are not reducible to each other.”81 But before I discuss similarities between music performance and complex dynamical systems I want to turn to philosophy, the more traditional seat of theoretical thinking about music. My contention is that Gilles Deleuze’s theory of difference and consistency can be usefully harnessed as it fosters non-hierarchical thinking and the non-linear inter-relationship of individual constructs and elements or “intensities.”82

Gilles Deleuze and Difference in Music Performance

In performing arts, thinking categorically and scientifically can easily lead to stifling rules and the formation of one dominant type of thought: “This is how Beethoven goes because Czerny says so.” Aiming to pin down what the composer might have wanted leads to “territorialization”; normative, recurring patterns of thinking and a limitation of possibilities. The emphasis on Urtext scores leads to “Urtext”—and “Maestro”— mentality, an authoritative hierarchy that discourages experimentation and to be different. Instead it promotes “obedience” and repetition, resulting in homogeneity of performing style. Hierarchical conception of difference reduces performance styles to static, binary opposites and (negative) categories. For instance, a performance is considered to be either HIP or not; it is evaluated against a specific category such as HIP or MSP, a context within which one or the other tends to be regarded inferior to the alternative.

As Sally Macarthur notes while discussing the binary between male and female composers, “In its pluralism and chaos, Deleuzian philosophy enables the construction of a web of interrelations for [performance styles] which opens [them] out to multiple possibilities.”83 Instead of “subordinating difference to instances of the Same, the Similar, the Analogous and the Opposed,” Deleuze argues for “the state of free, oceanic differences, of nomadic distributions and crowned anarchy.”84 He claims that “to ground is to determine. […] Grounding is the operation of the logos, or of sufficient reason.”85 And later reiterates “that to ground is to determine the indeterminate.”86 On the other hand

Systems in which different relates to different through difference itself are systems of simulacra. Such systems are intensive; they rest ultimately upon the nature of intensive quantities, which precisely communicate through their differences. […] Systems of simulacra affirm divergence and decentring: the only unity, the only convergence of all the series, is an informal chaos in which they are all included. No series enjoys privilege over others, none possesses the identity of a model, none the resemblance of a copy. None is either opposed or analogous to another. Each is constituted by differences, and communicates with the others through differences of differences. Crowned anarchies are substituted for the hierarchies of representation; nomadic distributions for the sedentary distributions of representation.87

As Deleuze considers such systems “sites for the actualisation of Ideas,” in other words active, individuating “multiplicities constituted of differential elements, differential relations between those elements, and singularities corresponding to those relations,”88 they are open and dynamic systems with ever expanding potential for something new and different. In these systems

Repetition is no longer a repetition of successive elements or external parts, but of totalities which coexist on different levels or degrees. Difference is no longer drawn from an elementary repetition but is between the levels or degrees of a repetition which is total and totalizing every time; it is displaced and disguised from one level to another, each level including its own singularities or privileged points.89

Individual performances of western art music, with its history, traditions, conventions, and ever renewed imagination, easily fit this description: when we compare recordings (or live performances) of a particular composition we are not dealing with “repetition of successive elements” but with “totalities which coexist on different levels or degrees.” And the differences between them are as Deleuze describes in the above quote.

Because Deleuze considers the “highest object of art […] to bring into play simultaneously all these repetitions, with their differences in kind and rhythm, their respective displacements and disguises, their divergences and decentrings,” he critiques the pursuit of universally agreed-upon suppositions. In his view these create an impasse; they deny the potential to renew, to create difference:

The more our daily life appears standardized, stereotyped and subject to an accelerated reproduction of object consumption, the more art must be injected into it in order to extract from it that little difference which plays simultaneously between other levels of repetition […]”90

If this is (one) role of art, including western classical music performance, then it is questionable what we gain by categorizing its various specimens. Description and interrogation of music performance as “systems of simulacra,” as examples of “multiplicities” and “difference” might be more meaningful. The analytical chapters demonstrate the rich diversity and interaction of performance features that come to light when we are willing to forget about neat categories or wanting to explain, to “ground […] the indeterminate.” Chapter three shows that hardly any of the selected recordings fits perfectly the opposing “One” of MSP or HIP. It is much more appropriate to think of a continuum along a vector or even a three-dimensional space of performance style where each performance hovers over several instances of “multiplicities.” These findings are further explored and illuminated in chapters four and five until enough evidence is amassed to demonstrate the analogies between complex dynamical systems and music performance. The pointing out of these parallels is assisted by occasionally referring to certain terms borrowed from Deleuze and Guattari.

For Deleuze’s philosophy of difference provides not just a broad aesthetic and analytical framework but also a toolbox of concepts. I hasten to say that my analysis is not going to use Deleuzian terminology directly. I am not conducting a Deleuzian analysis. But throughout chapters four to six I am signposting the parallels between Deleuze and Guattari’s concepts and my more mundanely formulated analyses of the recorded performances to highlight their persuasive support for the argument of this book. Therefore it is important to introduce them here while noting that since these terms are used only as potential analogies they are always placed in double quotation marks.

In his opus magnum co-authored with Felix Guattari, Deleuze distinguishes three “lines” in relation to strata (i.e. “phenomena of thickening; accumulations, coagulations, sedimentations, foldings”), assemblages (“produced in the strata by extracting a territory from the milieus”) and rhizome (a heterogeneous, non-linear chaotic system / multiplicity): The “molar lines” segment and categorize, they are subordinated to the “One”; they form a circular and binary system. The “molecular lines” are rhyzomic, they break and twist and pass “between things, between points.” They organize in a non-hierarchical fashion and are no longer subordinated to the “One” but are “multiplicities of becoming, or transformational multiplicities.” Finally there are the “lines of flight or rapture.”91 This is where transformation or metamorphosis occurs. They do not segment but completely break out of one form of construction (the “territorialized” and “repetitive” normative) and move towards an emerging other. Deleuze and Guattari emphasize the lack of symmetry which enables “the masses and flows” to constantly escape and invent new connections.92

I would argue that we can think of performance features as functioning like these three types of lines. “Molar lines” would constitute elements that define and categorize performance style. They “territorialize” and make a performance belong to a type (the “One”); for instance a clear example of seamless portato bowing denoting the MSP style. “Molecular lines” would be features that interact with other features and blur categories. They “deterritorialize,” in Deleuzian language, and create difference. They contribute to the transformation of style such as when articulation and bowing emulate period performance practice without using a period bow. The “lines of flight” might be those elements of a performance that create radical departures from the usual; for instance through extreme tempo or novel ornamentation and other variations. To analyse the complex interplay between these “lines” is a major aim of this book and Deleuze and Guattari’s conceptualization of difference and how it may be constituted in non-linear dynamical systems (“assemblages” and “multiplicities”) provides fitting theoretical underpinnings.

Music Performance as Complex Dynamical System

As I have indicated already several times, I propose that it is time for researchers to investigate music performance in its complexity because it is a phenomenon that displays the characteristics of complex dynamical systems. These need to be approached with an appropriately complex method. The aim, therefore, is to provide a theoretical framework and a model for such a comprehensive and multimodal approach. The ensuing chapters demonstrate the necessity of dealing with music performance in its complexity if the goal is to understand the phenomenon and its experience. Studying merely its constituent parts is not just inadequate for a proper insight but often leads to misconstrued information. To introduce my position, a review of what I consider important about complex dynamical systems is in order. How these characteristics manifest in music performance is argued more fully in chapter six, once sufficient evidence has been established through chapters three, four and five.

As is well known, music and its performance is a cultural product with a rich history. Paul Cilliers asserts that complex cultural phenomena are “never symmetrical” because the “history of the system is vitally important for the way in which meaning is generated.”93 Self-organizing complex systems make self-adjustments “to improve [their] own performance.”94 Music performance is no exception. As I have shown and will detail further in subsequent chapters, post-1980s HIP had embraced valid criticisms and altered its verbal rhetoric and, more gradually, its interpretative practices to “improve” and renew. The history of performing styles is never circular or symmetrical. The study of sound recordings shows, when older styles of expressive mannerisms seem to resurface at a later time, these tend to be ever so slightly or fundamentally different from their previous incarnations. It is quite unlikely that we would ever recreate the sound as it was originally produced, whether we have aural evidence or only verbal descriptions of a particular performing convention.

The current scene of HIP in nineteenth- and early twentieth-century repertoire is a case in point. The majority of such performances adapt precious little of the original conventions evidenced on early recordings and piano rolls. One fledgling exception, the work of Neal Peres Da Costa and the Ironwood Ensemble, shows preliminary attempts at re-instating nineteenth-century piano, string and ensemble practices.95 I say preliminary, because the extent to which they use particular characteristics (such as dislocation of hands, arpeggiating chords, portamento, asynchrony of parts, non-unified bowing, tempo irregularities, etc.) is far more modest than what one hears on early sound recordings. Eventually they may go further with some or all of these performance techniques but whether they will ever sound like a carbon copy of artists recording at the beginning of the twentieth century is highly doubtful and not particularly desirable.

Rachel Barton Pine, among others, has noted that the clear and rich sound achieved through current recording technology makes her look for a style of portamento (sliding between notes) and bowing that she feels is more suitable than a direct imitation of what one hears on old recordings.96 Barton Pine seems to refer here to a pre-conscious affect, a sensation of intensity. Listening to her playing and comparing it to the recordings of Maud Powell—in homage of whom Barton Pine’s recording was issued—I think I know what she means, but I am unable to objectively show the difference. My measuring tools (software that can be used to visualize and analyze audio signals) are not sophisticated enough to help quantify the differences in kind and degree. Lacking empirical evidence, we are left with our aural impression. This confirms a difference in the frequency of introducing portamento and a tendency to mostly use it in upward motion while Powell and others used it in both directions, often more frequently downward. The fascinating revelation, however, is not this. Rather, the techniques by which expressive gestures like portamento are delivered, but at least the resultant sound, are utterly different. I would venture to propose that the modern expressive portamento is fingered and bowed much more lightly while those heard in old recordings are more intense. When late nineteenth-century violinists introduce portamento for expressive purposes they play them with more weighted bow-strokes and greater definition. Performance does not circle back, it is “not symmetrical”; it moves more like a spiral because each generation reinterprets and further develops what it learns from the past. We simply may not have the right vocabulary to adequately describe what we hear nor the technology to study it empirically.

Music performance is a complex system not just because it is a culturally constructed yet individual artistic practice that is temporal and ephemeral, resisting verbal discourse, but also because the act of performance involves a multitude of elements from the technical to the aesthetic, and from the physical and measurable to the embodied, implied and subjectively perceived. All of these are inter-dependent, forming a complex web of non-linear interactions. The non-linear and asymmetrical nature of complex systems means that “the same piece of information has different effects on different individuals, and small causes can have large effects.”97 The infinite variability of music performance and its diverse effect on listeners are well established facts, lending further support to viewing it as a complex system. As such it is best grasped, at least metaphorically, by models that “can dynamically adjust themselves in order to select that which is to be inhibited and which is to be enhanced.”98 Music performance is robust and flexible, constantly changing and adjusting to cultural norms and expectations as well as to individual artistic personalities and technical abilities.

Interdisciplinary approaches and collaborations are therefore important because complex systems are more than the sum of their parts; the “intricate relationship” among these parts is equally, if not more, significant. By cutting up the system, by focusing on one or the other of its aspects, the “analytical method destroys what it seeks to understand.”99 Complexity is misrepresented by simplistic explanation but “an analysis of [its] characteristics […] can be attempted in order to develop a general description that is not constrained by a specific, a priori definition.”100 Distortions are inevitable since only certain aspects of a performance can be analysed at a time.

Frustratingly, even attempts at describing the holistic effect leads to dissecting because of the linearity of words and sentences: we can only speak of one thing at a time while our mind can decode-perceive the full aural image at once (surely not in all its detail at a conscious level but in its essence, something that can prove very elusive otherwise). I will try to overcome this linearity by frequent cross-referencing to earlier and later discussions of the same excerpts. Ultimately, we simply need to accept that our inquiries lead along a road of discoveries and increasing knowledge rather than towards absolute truth (or how the performance “ought to go”). If investigators of music performance adopt such an attitude they will make less lofty claims and act as researchers of the “nonmodern” era as opposed to assertive modernist scientists or postmodernist debunkers.101 An approach that focuses on the complex of what we hear and experience when we listen to a particular recording; an approach that acknowledges and interrogates the dynamical, non-linear interactions at play promises to enrich our knowledge and understanding of music performance in a way that science and its reductionist, positivist paradigm cannot.

But what are these potential non-linear, dynamical interactions? According to Cilliers, the characteristics of complex systems include:

1. A large number of elements are necessary but not sufficient […] the elements have to interact and their interaction must be dynamic. A complex system changes with time. The interactions do not have to be physical; they can also be thought of as the transference of information.
2. The interaction is fairly rich, i.e. any element in the system influences, and is influenced by, quite a few other ones.
3. The interactions themselves have a number of important characteristics. Firstly, the interactions are non-linear […] [which means] that small causes can have large results and vice versa. This is a precondition for complexity.
4. There are loops in the interactions. The effect of any activity can feed back onto itself, sometimes directly, sometimes after a number of intervening stages.
5. Complex systems are usually open systems, i.e. they interact with their environment.
6. Complex systems operate under conditions far from equilibrium [or stability]. There has to be a constant flow of energy […] to ensure survival.
7. Complex systems have history. Not only do they evolve through time, but their past is co-responsible for their present behaviour.
8. The interactions often take the form of clusters of elements which co-operate with each-other and also compete with other clusters. An element in the system may belong to more than one clustering. Clusters should not be interpreted in a special sense, or seen as fixed, hermetically sealed entities. They can grow or shrink, be subdivided or absorbed, flourish or decay.102

Looking at this list, I see many parallels with music performance. As we all know, music performance has many elements and they interact with each other in many ways. We do not just hear pitch but also its timbre and therefore the same pitch or note can have different character and affect; phrasing and articulation interact with tempo and intensity to create diverse meanings while articulation and tempo impact on the perception of rhythm and pulse, just to name the most obvious interactions. These may also form “loops,” and differences in nuances can have significant impact (“small causes having large results”). My analyses in the subsequent chapters show the history of performing Bach’s music in recent times; how the past is “co-responsible for present behaviour”; how this evolving style interacts with the players’ personality and musical biography as well as their cultural environment, and how all this is fed by the dynamism of ever new generations of musicians wanting to make their mark, providing a “constant flow of energy” and change. The discussion of similarities and differences between current MSP and HIP highlights the “clusters of elements which co-operate with each other and also compete with other clusters.” They also show how certain features may belong to several clusters. This in turn explains why their relative importance (as well as interaction with other elements) in a particular context determines their function; how strongly, at any given moment, the feature contributes to the style of the performance and its overall effect. There are no absolute indicators, only “multiplicities” with different proportions of “molar” and “molecular” lines “territorializing” and “deterritorializing” various “assemblages.”

Whether we see the relevance of Deleuze’s thinking or take heed from Cilliers’ assertion—that clusters are not “hermetically sealed entities” but “can grow or shrink, be subdivided or absorbed, flourish or decay”—we are surely facing a complex dynamical system in the phenomenon of music performance. The exploration of these parallels between music performance and complex systems infuses all of the subsequent chapters of the book. I return to engage with them specifically in the last one.

2.4. Performance Studies, Oral Cultures and Academia

There are two more issues I wish to touch upon before turning to the chosen repertoire. These are, firstly, the respective roles of “practice-based musician-researchers” and musicologists, and secondly, the differences between aural / oral and literate cultures.103 Although the notions of modernism and postmodernism could be adequate to explain certain observations, looking at them from these additional angles adds further depth and nuance to the study of music performance.

Research Roles: Performing Music or Analysing Performance?

During the last few years there has been an increased emphasis, especially in the UK, but also in Australia, on “practice-based” or “practice-led” (practice-informed) research and a questioning of the adequacy of performance research conducted by musicologists “to get to the heart of what underlies performed music.”104 The proliferation of self-reflections and musical self-analyses (at times referred to as auto-ethnography) is clearly noticeable, causing some concern in science-based circles of academia. I sense a growing tension between two camps, a tension that, to me, seems to be perpetuated and exacerbated primarily by simplistic, “one-size-fits all” funding models. The situation is not unique to music performance research. Speaking of academia more broadly Latour states:

One camp deems the sciences accurate only when they have been purged of any contamination by subjectivity, politics, or passion; the other camp, spread out much more widely, deems humanity, morality, subjectivity, or rights worthwhile only when they have been protected from any contact with science, technology, and objectivity.105

He argues for synthesis and mutual respect, adding:

We are […] so accustomed to taking for granted the abyss between the wisdom of the practice and the lessons of theory, that we seem to have entirely forgotten that this most cherished analytical clarity was reached at the price of an incredibly costly invention: one physical world “out there” versus many mental worlds “in there.”106

I could not agree more, as everything else so far in this book hopefully demonstrates. Surely, the study of phenomenology, dialogue, collaboration, and genuine communication are in the interest of all who wish to gain a better understanding of artistic processes. But none of this should mean changing or abandoning roles and giving up on rigorous standards. Musicological performance studies as I perceive them are not primarily for the benefit of performers but for the listening public and those interested in culture and history. Practicing musicians’ main contribution to society is to create high-quality performances and my task as a music performance researcher is to account for how and what they achieve and offer to listeners. These are two entirely separate activities, with different goals and means, in the contrasting domains of aurality and textuality. Both are valuable in their own right and dependent on the other to gain greater insight. If musicologists are not asked to perform why are performers required to write?

I am not denying that performers should be well-informed about and reflect on what they do, or that they may even wish to communicate their thoughts in the form of published research. These can offer different and important perspectives and I am always curious to read them. I simply would like to reaffirm the right of performers to “just” perform, to communicate their knowledge aurally, through performance. By the same token, I wish to affirm the validity of “traditional” musicological investigations of performers and performance that deal with the product of performance, the “sound object.” Or, whether in collaboration or not, with the process that helped it come to life.

I find it contentious to do away with the subject-object dichotomy and agree with Latour that we need to overcome it instead because “the object that sits before the subject and the subject that faces the object are polemical entities, not innocent metaphysical inhabitants of the world.”107 The self-reflective practitioner may not be the most reliable source for exploring certain processes and mechanisms. Psychologists speak of demand characteristics and attribution as essential problems that may arise in such contexts. Patrik Juslin asserts that “because many of the processes and mechanisms are ‘implicit’ in nature and could occur in parallel, researchers cannot rely merely on phenomenological report or introspection to explain musical emotion. (The music experience is the thing that needs explaining, rather than being that explanation). Most of what goes on in the causal process might, in fact, not be consciously available.”108 Juslin refers to Paul Silvia who also “notes that ‘perceived causality and true causality diverge,’ since ‘processes irrelevant to causality influence the attribution that people make’: ‘people tend to attribute causality to salient stimuli, even when salient is unrelated to the effect.’”109 From a more philosophical, humanistic and subjectively felt viewpoint we may ask: “Who has ever mastered an action? Show me a novelist, a painter, an architect, a cook, who has not, like God, been surprised, overcome, ravished by what she was—what they were—no longer doing.”110

The problem as I see it is that the performer is not necessarily better placed than the musicologist to verbalize the characteristics of a sound object or the processes that generated it. As introduced in chapter one, the typical inadequacy of words (including metaphors) for the particulars of bodily, somatic, kinaesthetic, aural and psychological experience (or action) is the biggest obstacle in researching music, especially performance. A good example comes from a recent DVD where the eminently articulate and inquisitive Pieter Wispelwey is struggling to explain the difference between the sounds of his two different baroque cellos. First he states that “at 392 the sound is even more relaxed” but also “rustic and raw” adding that “it’s all a matter of colour.”111 Later he comes back to this issue and again notes the difference in tuning but then simply keeps repeating “it’s very different” while shaking his head and eventually just plays a passage on both instruments ending with raised open arms and shoulders and a huge smiling question mark on his face, as if saying with delight “don’t you hear? It’s all very different—wonderful / amazing, no?”112 I am not sure the audible difference is so noticeable, or entirely stemming from the instrument (or the tuning) rather than the way he plays the passage on each cello. It is possible though, that he plays differently because the different instruments “prompt” him or react differently. What this episode makes blatantly clear, however, is the phenomenon that what is in the mind (arms, fingers, ears, whole body) of a performer may not always be audible for the outside listener. Still, I wonder with Latour “Why is it that we cannot readily recover for our ordinary speech what is so tantalizingly offered by practice?”113 and suggest we look for the answer in contrasting oral and literate cultures.

Oral Cultures and the Aurality of Music Performance

Others have also drawn attention to the importance of aurality in music making.114 But in relation to the study of western art music performance this issue has largely been neglected. Instead, there has been an increase in “how to” books, tutors, and manuals (just like in the empiricist eighteenth century!) that provide written accounts of how to play the piano, the violin, the flute, the guitar; how to have a healthy singing technique; how to interpret particular repertoire, how to compose hit songs; how to be creative; how to practice, how to memorize, and so on. Almost every “famous” teacher since Carl Czerny (1791-1858) and Rodolphe Kreutzer (1766-1831) has put pen to paper to disseminate their knowledge—accrued from learning mostly by ear from other famous musicians—to eager students and interested amateurs around the globe.

This emphasis on the written word is at odds with our multi-modal but primarily aurally perceived appreciation of music. As neuroscience has shown, “Musical activity involves nearly every region of the brain that we know about, and nearly every neural subsystem.”115 If so, we are limiting our potential to understand our interaction with music if we over-emphasize the analytical, the abstract, the written or notated; what we can verbalize. It leads us to look for what we can measure and make us think we explained what is not (yet) measurable. Such an approach fosters categorization, ossification, homogenization and normative thinking. Oral cultures thrive on variation, on “thinking forward,”116 on experiential and communal learning, communicating and being in the world.

Music performance parallels oral cultures in that it is based primarily on experience—kinaesthetic, aural and emotional memory—that is honed and kept alive through repetition and daily practice. Although music notation has become increasingly sophisticated over time and western classical musicians rely on score reading, much of the musician’s training and active professional life takes place in the aural realm. Surely teachers often use metaphors and imagery, occasionally specific musical terms, but most often they demonstrate, playing their instruments or singing / humming.117 Anyone who has observed rehearsals of experienced musicians knows how little they speak and how well they seem to communicate through sound and gestures in shaping their interpretation. This is also confirmed in experimental music psychologists’ research.118 If there is any discussion at a rehearsal, it is often combined with brief demonstrations on the instrument, as if to confirm proper understanding. Knowledge is “conceptualized” in sound and movement rather than words. As Mine Doğantan-Dack states, “music making requires mentally hearing and imagining the notation as music.”119 Because of this and because western music notation is limited in capturing phrasing, articulation, rhythm, intensity / volume (dynamics) and tempo while being silent on timbre, the cognitive processes of the performer’s musical world resemble those typically found in oral cultures and this is what we need to tap into and investigate.

The perception of sound is limited to a few seconds—similar to the time-limitations of short-term memory—and experienced as “the now.” According to cognitive science,

Echoic memory is an auditory sensory memory that persists for several seconds, after which it is lost unless attended to. […] Echoic memory enables us to relate what we are hearing at this very moment to what we have just heard. It permits us to maintain a temporal window wide enough to recognize a dynamic sound or parse a phrase.120

Importantly, “this sensory memory is image-like […], in that it exists independent of and prior to language and it is often difficult to capture in words.”121

In contrast, reading texts engages sight and sight isolates, dissects. As Walter Ong has argued in his foundational text on orality, “A typical visual ideal is clarity and distinctness, a taking apart.” Listening to sound is the opposite. “The auditory ideal […] is harmony, a putting together.”122

Whereas sight situates the observer outside what he views, at a distance, sound pours into the hearer. […] Vision comes to a human being from one direction at a time: to look at a room or a landscape, I must move my eyes around from one part to another. When I hear, however, I gather sound simultaneously from every direction at once: I am at the centre of my auditory world, which envelops me, establishing me at a kind of core of sensation and existence.123

Noting that the concepts of interior and exterior are “existentially grounded concepts, based on experience of one’s own body,” Ong also explains that

Interiority and harmony are characteristics of human consciousness. The consciousness of each human person is totally interiorized, known to the person from the inside and inaccessible to any other person directly from the inside. [...] What is “I” to me is only “you” to you. And this “I” incorporates experience into itself by “getting it all together.” Knowledge is ultimately not a fractioning but a unifying phenomenon, a striving for harmony. […] In a primary oral culture […] the phenomenology of sound enters deeply into human beings’ feel for existence.124

As musicians tend to spend long hours playing music from an early age, we can think of them as living in a “primary oral culture.” In such worlds words or musical notes are not thought of as something “laid out before their eyes […] ready to be explored,” dissected, analysed, understood; they are experienced and acted, although they can also be mentally represented.125 Pianist-musicologist Mine Doğantan-Dack has recently formulated something similar in relation to classical music performance:

The experiential reality for the performer is such that the visual symbols in a score are always already perceived as “music,” together with various expressive details that are understood immediately as constitutive of the music and are not inferred from the score. […] Consequently, the visual, objectively identifiable and fixed entity that researchers regard as the musical score, and the audible, subjectively construed phenomenon that defines the score for performers are not ontologically the same phenomenon.126

At the end of the book I shall return to the problem that an over-emphasis on seeing, on analytical and abstract thinking may pause. There I shall propose reasons why we should try harder to find ways of interrogating how and why we hear and react to music performance the way we do in a more holistic manner. Here I would just like to note the accumulating literature promoting the “revaluation” and “rehabilitation” of our aural sense.127

Exploring Jean-Luc Nancy’s phenomenology of listening, Anthony Gritten refers to research on music’s role in human evolution and concludes that “there is a strong case for claiming that the ear is the primary sense organ of the human body.”128 He comes to this conclusion by arguing that

the extraordinarily deep-seated and often unthinking bias towards visual modes of cognition, action and judgement is a necessary detour in human development that finds its real significance within the narrow context of the rise of Enlightenment Modernity in the techno-scientific developed world. […] we should acknowledge that we have been listening for longer, and that we have been listening quicker for that matter (the human brain processes audio data faster than visual data). Given that listening is central to many ways of being in the world otherwise than, and often older than, the dominant Western model of communicative consensus […] it behoves us to rethink the function of listening, and by extension musicking.129

How can we do that, in our current academic climate where even visual artists, let alone musicians employed by tertiary institutions are required to write as much as to produce creative work? When we are fixated on the written, we miss an opportunity to study our ability of sensing and communicating aurally. Yet this is what musicians do best.

Keeping Music Performance in the Aural Domain

Although my focus is on performing western art music, there is no need to underline that most other musical traditions do not use notation but pass on everything purely aurally. There is considerable emphasis on learning by ear even among pedagogues of western classical music as well, especially during the foundational years. According to Ong, people who have no recourse to writing and have to rely on their memory for everything tend to think much more functionally and eschew abstraction. For instance, if they are asked to define an object, they do not provide “a sharp-focused description of visual appearance […] but a definition in terms of its operations”—in our case, a demonstration of a musical phrase or gesture, rather than a scientific explication.130 Oral cultures do not

deal in such items as […] abstract categorization, formally logical reasoning processes, definitions, or even comprehensive descriptions, or articulate self-analysis, all of which derive not simply from thought itself but from text-formed thought.131

Although modern musicians of the western art tradition have recourse to sound recordings as well as music notation and thus are able to take a snapshot of sound or the score for close inspection, their learning or polishing performance still largely parallels oral cultures in its ways of enabling recall of solutions. Just as in oral cultures, musicians learn to express the full existential context of performing particular pieces through repetitions and the adoption of conventions.

Back in the seventeenth and eighteenth centuries there was no sound recording technology and notation was much more skeletal. Oral traditions were even more alive partly because fewer people were (fully) literate. Also, musicians came from families of musicians stretching several generations. They learnt composition and ornamentation primarily by imitation and daily practice as well as absorption of a formulaic “vocabulary.” Unlike their modern-day colleagues, they only played in their local, contemporary style and were not required to “speak in dialect,”132 let alone the different “languages” of music composed over a period of more than 400 years. For them, just like for the bards of ancient times and oral cultures, thinking of a single text, of one fixed version was still quite an anathema. Notation was severely limited in conveying information essential for an ideal performance so they interpreted them at will, adding and changing according to their fancy even if relying on local customs. They performed a living, vernacular tradition that was malleable and fluid typical of the artistic output of oral cultures. They also tended to play their own music and so had control over the contributing elements. They worked with patterns and “templates,” and the learning of articulation and figuration were embedded in their instrumental training. What we find in modern institutionalized musical learning is a process, at least partly, of “training [the professional musicians] out of that flexible mode of thinking.”133

An overemphasis on text, on being “true to the score,” on playing exactly as the score prescribes, goes against the long tradition of living in the aural realm, especially since scores are so hopelessly inadequate in conveying what a musician is required to do when performing a piece (and equally inadequate to transcribe what performers have actually done).134 Similarly, in my view, an overemphasis on self-reflective research, on insisting that musicians verbally articulate and publish academic papers about what they are doing simply reinforces the primacy of literacy and delegates an unwarranted superiority to verbal discourse. It is a sign of a dominant left-hemisphere, according to Ian McGilchrist.135 The question to which I will return in the Epilogue of this book is whether an over-reliance on what the left hemisphere does best (such as analysis, categorization and abstraction) is ever going to be adequate to investigate processes (like music performance and listening) that are deeply multi-modal and heavily implicating the involvement of the right-hemisphere.

Academia Once More

As I indicated at the start of this section, people lament the complacency of analysts taking the upper hand in performance analysis (the text to act flow of signification), or the unequal balance of power between academics and performers or between musicology and performance.136 Yet funding principles of tertiary institutions perpetuate the situation by insisting that performers should write and publish “original research”; verbal discourses rather than (or as well as, in somewhat better scenarios) create exciting and illuminating performances (whether informed by research or not). I find it a demeaning situation that stifles creativity when musicians are required to demonstrate “original research” in their performance or recording of a repertoire piece. It is not just difficult but near impossible to do so when performing a Beethoven or Brahms sonata or Bach’s Solos for Violin, for that matter, or any other oft performed music. Yet the performance may be stunning and worth a hundred “original” research papers about them!

In oral cultures the “integrity of the past [is] subordinate to the integrity of the present. […] Oral traditions reflect a society’s present cultural values rather than idle curiosity about the past.”137 The musical mainstream’s claim to a living tradition reaching back to Beethoven reflects a mind-frame focusing on the integrity of the present. Even the development of HIP is a testimony to the residual power of this aural mode of existence: When the written sources were first recovered and studied musicians could only interpret them according to their modern experiential boundaries of aural and operational memory. That is why the use of old instruments proved so crucial yet it took so long to let go of playing them with modern technique.138 The older generation of HIP practitioners sometimes note that the younger players are not familiar enough with the sources but try to “cut corners” by quickly learning the essential “tricks” from their teachers and then use them liberally and routinely.139 This might be a lamentable attitude, but it also shows that music performance is primarily an aural practice that thrives on imitation and variation rather than abstraction and analysis. Musicians are like “skilled oral narrators [who] deliberately vary their traditional narratives because part of their skill is their ability to adjust to new audiences and new situations or simply to be coquettish,” to entertain.140 We should let them play and rejoice in the sensual pleasures they offer. We should let “the phenomenology of sound enter deeply into [our] feel for existence” because listening to music brings us closer to our own inner world, to our “interiorized consciousness” more than a thousand academic words.141 When we listen to music we are not “out there” contemplating an object but “in the music,” living it in the now.142

As a final point in my argument regarding practice-led / practice-informed research, I cite Latour again to reiterate “that action is slightly overtaken by what it acts upon, that it drifts through translation; that an experiment is an event which offers slightly more than its inputs; that chains of mediations are not the same thing as an effortless passage from cause to effect.”143 The “experiment” Latour refers to in this quote can easily be exchanged for “music performance” or its analysis. There are innumerable examples when performers’ verbal discourse does not match their performing practice. It may be that the intention was not executed (or not perceptibly executed) for a variety of reasons, or the verbal pronouncement might have been just empty rhetoric. The fact of the matter is that performers practice, research and reflect but then the performance takes over, and the musician’s preparation “drifts through translation.”144 The musicologist in turn tries to make sense of this translated act but while looking for adequate words and methods a new “chain of mediations” arises and the perceived or analytically derived cause and effect may not match entirely the performer’s understanding and mental construct of her own act. And this is all well and good, for the musicologist is primarily a listener and as such writes for other listeners. The performer on the other hand may best communicate to other performers and listeners her knowledge about performance and the work she is performing through the act itself, within the experiential aural domain.

2.5. Conclusion

At the end of this meandering tour of theoretical and methodological approaches to music performance we seem to be well positioned to launch into our designated material, the forty-odd selected recordings of Bach’s Violin Solos made since Sergiu Luca’s ground-breaking first with period apparatus from 1977. The development of and current views on HIP and MSP have been mapped from multiple angles—cultural, historical, analytical—and the proposition to consider the recordings as manifestations of multiple, non-linear, looping interactions in a complex dynamical system that is music performance has been put forth. Through the review of analytical, empirical and experimental research into music performance we realized the need for a more comprehensive approach that would offer an adequately complex and more nuanced account of this multifaceted human activity. My position that both objective and subjective, aural and written (as well as visual) representations and discussions are absolutely essential for a balanced and fair evaluation of what is happening in music performance has also been made clear.

In the following chapters I endeavour to showcase a method that engages with music performance in its complexity. I examine the selected recordings of Bach’s solo violin music against this theoretical backdrop and explore the potentials and limits of an academic approach to the study of music performance. Some might readily object that recordings are not performances but I believe they are.145 Those disputing it generally cite the unnatural environment of the recording studio, the inhibiting effect of the microphone and, above all, the editing processes involved. But in my experience the differences between Arturo Toscanini and Herbert von Karajan or Sviatoslav Richter and Alfred Brendel are fairly clear. According to empirical data I collected in 2006, most recordings are released when all parties are satisfied with it and the majority of soloists would agree that their recordings reflect their ideal performance of the work at the time of recording.146 Most importantly, recordings are experienced by listeners as performances, in my experience they may even sound different each time one listens to them, and so I make no apologies for studying them as such.