The invitation to write this editorial came at the perfect time. I was embarking on a six-month sabbatical at Girton College, Cambridge and reflecting on some of the reasons for taking a career break. It would be a chance to explore new ideas and to escape from some of the ruts I had created for myself during a quarter century of performing (primarily leading and directing) and lecturing about performance. In particular, it would give me time to reflect on those aspects of eighteenth-century repertoire that have shaped my approach to preparation and rehearsal, as well as to performance itself. I have become increasingly interested in the experience of music as a social event – above all, how the social dynamics of ensembles can promote and restrict creative potential.

In the academy and among the listening public, there is still ambivalence towards and misunderstanding of performers interested in historically informed performance (HIP) using period instruments. Yet even though HIP has taken knocks, it appears to have survived, and it is certainly not the niche market once predicted. The debates of the 1980s may be behind us but, I would argue, there remains much to be explored of their legacy; the impact of HIP on the music industry, on concert halls and major opera houses, on CD and media production, on ‘modern’ orchestral practice, on conservatoire and school education – if not on the ‘new’ musicology – has been far-reaching. The most significant and frequently recurring theme during this time has been that performance be ‘informed’: by what and in what ways are questions that should create an interface between performance and musicology.

My own reasons for finding creative partnerships in the HIP movement are not unique. When I was an undergraduate at Cambridge University, some exceptional teachers and composers taught me never to accept too many assumptions about the musical object or stop questioning my own response to it. Perhaps this was why I felt temperamentally unsuited to a job in a symphony orchestra and instead joined contemporary music groups, enjoying the challenges of new pieces and of working with composers, discussing issues of performance and interpretation with them in depth. I wanted to take this approach to other repertoire, in particular to the music of the eighteenth century. I didn't, however, find this method of music-making in some of the well-known chamber orchestras I had idolized. There was no discussion except between the leader and the ever-present conductor, nor did rehearsal breaks seem to elicit much talk about the music. The social seemed to remain merely social in the colloquial sense. Indirectly, and subtly, this probably informed the musical experience. Nevertheless, some musicians performing new music had swapped their metal strings for gut and abandoned chin rests and were exploring the versatility of bows without ferrules. It was with these colleagues as well as early music enthusiasts that I found a common language and a spirit of enquiry. Maybe the approaches of contemporary music groups and period-instrument ensembles were similar – or perhaps it was simply a matter of finding people with whom one communicated well. We discussed how to make the most of the vertical dimension of the music alongside the horizontal, musical structure, phrasing, gesture and rhetoric. Many of these elements seemed to be improved by our using period instruments: greater instrumental clarity, for example, brought immediate textural benefits. In short, the music began to make sense in the light of my experience and education. New repertoires opened up; the revelation of non-equal tuning systems added piquancy to familiar works. I didn't find anybody who felt he or she was part of a musical Sealed Knot. There was no sense of re-enactment. Nor were we engaged in musical time-travel. The recording companies, who funded much of the exciting work in the 1980s, inevitably needed marketing tools, but as performers we were in no doubt that this new style was for our age. There were breaks with tradition; it was challenging to cut a link with the long performance history of such well-known repertoire. Some performances were experimental and at the limit of what players were discovering about the old instruments and their techniques – but the results were often thrilling.

A generation later, it is interesting to consider what remain key issues for performers curious to cultivate some historical performance techniques alongside modern aesthetic values. The methods and tools of preparation – texts and editions, historical evidence and rehearsal techniques – that inform an interpretation are constantly in flux as new evidence comes to light and new fashions creep in. In the early days of HIP the idea of finding the ‘truth’ of a work in an ultimate text, the Grail of any given composition, was a strongly held conviction. It is now clear that the whole notion of what constitutes the musical work can be questioned. Does one attempt to source a composer’s final thoughts, if they ever existed? How do these relate to early publications? Do new ‘Urtext’ editions provide a clearer guide to editorial practices? The worst material possible is a version that adds so many markings to the score that an arrangement of a composition results rather than an edition. (I might add that conductors/directors who heavily mark parts before even hearing how players will perform do not endear themselves to an ensemble of knowledgeable instrumentalists.) There can, however, be so many possible versions of a work that performers are learning to be aware of the choices made in preparing even the most scholarly-looking material.

Nor is examining contemporary evidence regarding performance practices unproblematic. There are dangers in looking for prescriptive evidence in treatises. Treating them as code-breakers is not always productive or creative. For whom were they written – for students studying to be professionals or for an amateur market? Were they a rant about bad habits, a deliberate attempt to establish a new style (or the result of a fear that new practices were creeping in) or a critical look at current practice? Is a highly personalized style detectable, a narrow local aesthetic, or are the directions based on comparisons with other writers and wide experience? In short, how far is the writer to be trusted?

An interesting solution to these textual problems arose for the London Haydn Quartet (of which I was a member) when we came to choose an edition from which to record Haydn’s Op. 9 string quartets. We decided upon a single London one by Longman and Broderick from 1790. This was based on a desire not to recreate an eighteenth-century experience but to use all our twenty-first-century musical resources. Only very occasionally did we need to decide whether a note was an actual mistake. The closely engraved plates enable an entire quartet to appear in a single view – which, in addition to eliminating page turns, also reminds one of the content of the whole. Occasionally material has been omitted (for example, the Adagio from the E flat Quartet doesn’t have a decorated repeat) that changes the character of the movement. The freedom to experiment with the differences in articulation between parts opened up many alternative readings that modern editions iron out. But we couldn’t unlearn our knowledge of other editions, nor did we try to. That seemed an unnecessary conceit. Anomalies became the opportunity to define a common musical language between us based on experience and, ultimately, on musical taste. The experience was enlightening and the results well worth hearing.

In a string quartet there is clearly scope for the four players to negotiate among themselves. When we chose to perform from the eighteenth-century edition of the Op. 9 quartets, we also abandoned some conventional approaches to rehearsing. We relied on the parts rather than scores. The parts don’t come with bar numbers or rehearsal letters. Nor do they have pencils attached. It was illuminating, liberating and creatively inspiring to work from the material in front of us. Nor, interestingly, did it require more time. The lack of pencil markings was neither experimental nor ‘historical’. It was the consequence of a belief that rehearsals should explore the possible, maybe the probable, nature of a performance – but not necessarily its inevitable manifestation. Our common language was often created through playing alone, but when we did debate it was often useful to describe the music with analytical terminology. Among musicians who are comfortable with this approach as a complement to other forms of expression, the language of analysis can make a valuable contribution to the rehearsal process. Such ‘analysis’ highlights large-scale structures, of both a thematic and a tonal nature, and small-scale gestures, phrasing and harmonic events. It can be an aid to a closer understanding of the music and act as another creative contribution to an ‘informed’ performance.

Recent analyses ‘of’ rather than ‘for’ performance – focusing particularly on recordings – have begun to address the previous authoritarian position of formal analysis. However, as a performer, I am wary of too high a status being given to many recordings. The circumstances of production need to be placed alongside any conclusions that analysis reaches – and these circumstances have altered with technological advances. The recording sessions for the Op. 9s were a good example of the tensions that constantly arise between the spontaneity and uniqueness of a performance and the necessity when recording to reproduce a phrase exactly in order to cover traffic noise, a squeaky string or a technical lapse. Such accidents are the very substance of live performance, and are valued as such, but are unacceptable for a studio recording. There may exist a whole Pandora’s box of a history of recording. I realize that this is the subject of another study in itself, but I would urge its consideration. And I would warn that, in examining these data, one should not ignore the production process.

If negotiation between members of a string quartet during rehearsals is a given, in a large ensemble such as a chamber orchestra performing a Mozart or Haydn symphony the rehearsal process is more complex. Three questions stand out. Who might lead such a process? What decisions are taken on an individual level? Are there decisions taken subconsciously within a group of players performing the same material, as, for example, in a string section? While not discounting the usefulness of visual signals between players, relying on listening as a first resource to play eighteenth-century repertoire is a good measure of its complexity. This also reflects the salient lack of the silent conductor as a central figure in the organization of a performance.

Leadership in the eighteenth century was the preserve of a keyboard or violin director, or was shared between them. In a classical symphony with a dozen or more lines and players reading from single parts, controlling the flow and drama of the music also passes between instruments. A good chamber orchestra knows – sometimes instinctively, sometimes assisted by direction – how to apportion this responsibility. Prescriptive direction rarely engages players’ respect for long. The instructions loud/soft, short/long, fast/slow are not creatively stimulating to a player who has practised for the supposed ten thousand hours needed to be in a first-class chamber orchestra. Alternatively, creativity in performance can be fostered by and for the whole group – from within as well as from leadership that ‘facilitates’ (a term used recently by a colleague to describe what she most admired about John Butt’s directing). I do not want to negate the potential role of a more conventional conductor in performance, particularly of repertoire that has iconic status. In Bach’s day, his very presence, together with clarity of local style and the relative absence of a diverse repertoire, must have contributed to a unified approach in a performance of the St Matthew Passion, for example. Today, we enter upon such a project armed not only with our own personal experience of the work, views on possible performance practices and theological interpretations, but also the wealth of previous performances all stored in our memories. Channelling all these experiences into a coherent performance may benefit from a central figure. There are complexities of ensemble that are almost too demanding to be comfortably negotiated without a director, yet to coordinate the conceptual approach and pace the unfolding of a work is not the same thing as sending out a policeman onto the podium to patrol the players.

One aspect of interpretation that is often settled at an individual level is what for many instrumentalists remains one of the most preoccupying and crucial dimensions of eighteenth-century performance practice: phrasing. Here the influence of Harnoncourt’s 1982 book Musik als Klangrede (translated by Mary O’Neill as Baroque Music Today: Music as Speech (Portland: Amadeus, 1988)) and its consequences for performance were far-reaching. Most importantly, this book encourages an unequal, inflected mode of sound production, with performance choices operating on the micro level. It also demands attention to the concept of rhetorical delivery; in my experience, audiences are attentive and responsive to this approach. Given the notational under-specification of eighteenth-century music, questions about what the notation presumes, or even means, have to be asked. In a Mozart or Haydn symphony, for example, with the possible exception of notes with individual articulation marks, the question ‘how shall I play this?’, however instinctively asked or answered, can arise several times per bar. Performers need to make choices that feel spontaneous and inventive yet sit comfortably within an orchestral setting. This can happen when a closer attention to local shaping is allowed to interact with longer lines. Slurs delineate phrasing but one must avoid the danger of a chopped and abrupt style that breaks up the music. Suggested with subtlety, phrasing, slurs and gesture present a musical argument with speech-like clarity through contrasts of light and shade. Rejecting an emphasis on phrase beginnings in favour of direction (a major feature of nineteenth-century performance style) brings out the long line at the expense of local shape, and may also result in an unwanted crescendo. This method of playing can also negate any sense of bar hierarchy pulse and, for string players, the beauty of unequal bowing. Such phrasing, shaping and linear control are often the realm of the individual musician, and offer great potential for creativity and originality.

String players in small sections collectively negotiate a playing style involving many complex technical and musical judgments. This is achieved with very little verbal communication and at a very high level of detail. Players must continuously pay attention to articulation, bow speed, shape, phrasing, timbre, ornamentation (including vibrato), fingerings, dynamic nuance, rubato and ensemble. This is not done to clone one another – computer-generated sections do not sound like good human versions, after all – but to create a vibrant, flexible section. In addition to their own group, players are listening to other lines, all the time making choices about which are most important.

The negotiation of leadership, phrasing and the infinitely subtle moulding of a string section take place, for the most part, in a rehearsal room. The greatest single influence on my musical development has been in the preparation for a performance – in rehearsal. Performing can be exciting, stimulating, rewarding, humbling and occasionally scary. In practice, though, rehearsals are themselves performances, since both acts require the revelation of passionately held, intensely personal views. It can be a tough experience to have these challenged. The social dynamic within an ensemble is a critical dimension of the rehearsal process. This is established in and outside the rehearsal space, on and off the concert platform, and has taken many years to coalesce into its current form in many well-known groups. In order to function well, rehearsals depend on cooperation and agreement about the way in which language is used. Describing the notation as a map or a blueprint or as a language in itself can assist in the communication of ideas and interpretations, and each contains its own agenda. The metaphor of spoken language is a useful way of describing articulation and small-scale phrase units. The vocabulary used can describe formal structures or offer a straightforward reminder of how individual parts relate to each other, or it can be more general, even philosophical in nature. People react very differently to the use of poetic images or humour, and each player has a preference for how they are directed through speech or gesture. Musicians often describe music as language beyond words. How frustrating, when one needs to search for an explanation to convince others of how something ‘goes’. Underlying a very strong musical conviction may be well-founded, informed intuition, but without adequate communication the idea is unlikely to be taken up by the group. Unless a wordless rehearsal technique is used, players who do not relish verbalizing their ideas often lose out, but demonstrating by example alone is not always sufficient. Rehearsals reveal the difficulties in a complex social environment of finding a language of and for music. Their potential to harness the wealth of experience and the expectations of individual players to be creative, socially cohesive, even inspirational, or at least satisfying and productive, is vast. The resulting performances can be transforming.

And what next? This article has reminded me that in describing the sophisticated qualities required of musicians I am surrounded by them in the form of colleagues. It would be a shame to stop there. New and continuing dialogue between performers, musicologists and other disciplines offers a real opportunity to ask what an ‘informed’ twenty-first-century performance of eighteenth-century music might be, and how it might be achieved. Ethnographic work could further uncover the skills of ensemble playing – since this has been underestimated and largely undocumented in Western art music, with the exception of some studies of string quartets. The social dynamics of a musical rehearsal may be shared by other performing arts, including the theatre. So it is to that discipline, to ethnomusicology and to the social sciences that I shall turn now, alongside research into the historical evidence for accounts of rehearsals and ensembles in the eighteenth century.