Let us start by discussing the approaches you take towards performance practice, especially of the nineteenth and twentieth centuries, where your experience in the field of early music allows you to work from a radically different perspective.

'Basically, the idea is that the principles of historical performance practice can be applied to other repertoire. For example, Stravinsky's neo-classical works, and especially pieces like Pulcinella, need to be approached with a sense of the tradition that existed between the eighteenth and early twentieth centuries. In the twentieth century, most particularly after the Second World War, everything appears to have fallen apart, not just materially, but culturally. I don't think this was immediate, because we can still listen to recordings made in the 70s that are very musical, even if at times they are somewhat generalised. However, in present day 'post-recording' age performances, everything is controlled rigidly and free expression has tended to become sidelined in favour of total accuracy. Oddly enough, this seems to be the way musicians want to hear it, in my view an example of our alienation from any sense of tradition in performance; in terms of singing, for example, one can cite the lack of bel canto, in both technical and musical senses, as a perfect example.'

Is it is also because risk-taking, perhaps alien to the practice of today, is a much harder process to make effective and so musicians prefer to take an easier road? Both record producers and performers want to play safe for obvious reasons: after all, something that the public perceives as odd is not likely to sell well, no matter how musical.

'The way that musicians listen in recording is, in my view, often quite false; they seem to listen only in order to get it in tune and make sure that it is together. After all, the tape recorder is very unforgiving! In a concert situation, the human ear doesn't pick up the same things: in a large space, one doesn't really notice discrepancies of intonation and ensemble, for example. Things like that don't worry me, I have a Schnabel-like attitude to the recording studio: if one is prepared to take chances in performance, then one should do the same in recordings. Many listeners, however, take their 'recording' ears into a concert hall and somehow manage to trick themselves into the belief that something which is in fact very small and dry, and frankly is very boring, is golden. In addition, I find this attitude prevalent amongst singers who are preparing Classical repertoire, such as Haydn and Mozart, and early 19th-century bel canto composers such as Rossini and Donizetti. They seem to have the idea that it should be performed in an absolutely pristine manner, in strict time with no sense of portamenti, rubato, or even dramatic pace - in other words, any of the devices that we know were in use at that time. Indeed, singers are often applauded for not doing these things, thus repressing not just bel canto style, but also their musical instincts in general!

'But it is important to have a sense of context in no matter what one does. My approach to early keyboard repertoire is a case in point. Each year I gave a harpsichord recital at the Wigmore Hall in London. Each centred on a particular pair of countries, such as Germany and Italy, Germany and France, England etc., in the course of which, I tried to explore the repertoire of each country from the sixteenth through to the eighteenth, a potted history of musical style. This overview is something I always achieve, whether as a keyboard player, or as a conductor or when working with voices. Also, I read around the repertoire, placing it into a cultural context by looking at other arts and literature. I was also looking into changing attitudes towards performance practice to gain a sideways sense of perspective in addition to basic historical knowledge. In a dislocated era like ours, it is important to have a sense of both perspective and tradition, which in my opinion we have lost altogether.'

In many respects, this seems to be a reaction to what is very much a communications-driven society where, ironically, we seem to have lost the power to communicate. I think this is where musicians often go wrong. There are those who purport to be early music specialists, who claim to be up on current scholarship, and there are those who might be up on contemporary performance practice, but haven't the performing flair or imagination to be able to bring it off. We are now a society that wants results immediately and so we sacrifice true communication. The listening public seem not to want anything that challenges their perceived ideas on music; they want it straight, it's less exciting, but is nevertheless safe.

'I think there is a lot of truth in that, but what always surprises me (and gives me a lot of hope as well) is that there are a few conductors around like Simon Rattle and Charles Mackerras, who are in different ways true to basic musical instincts and stylistic traditions. Although Rattle cut his teeth in new music, he tended to work back into the earlier classics through his involvement with the Orchestra of the Age of Enlightenment. Consequently, he developed an understanding of not only how period instruments sound and work, but also the consequent implications for how the music should go. He is one of a few who are both genuinely and deeply musical and who are developing a sense of historical perspective and tradition, and who have a need to develop their ideas through their performances. With Rattle, it is perhaps more striking, since he is a much younger man who has moved back towards Beethoven; Mackerras, who is much older, performs with a great sense of the tradition he grew from. Using a basic awareness of tradition, discovering it in the music, and then expressing it in performance should be an important aspect of anyone's training.'

What about the research of earlier performance techniques? There are many instances where it is effective, but often the results seem to be counterproductive: musicians seem not to have the time, or the inclination for that matter, to put them into practice successfully.

'One can do the research and put it into practice, but it can misfire if taken too literally or specifically. On the other hand, if you simply do what your teacher tells you - which is often what happens - you end up doing nothing with the music until you are told to, either by a teacher or a conductor. In my own teaching, above all else, I try to make pupils think for themselves and make decisions about how the music ought to go. To reinforce their basic instincts with specific knowledge about any given style period or style of music they might be working on should be an ultimate goal. Certainly, as far as early performance practice is concerned, there is still a lot to be discovered, but this is the case with any period. It is the duty of any professional musician to mug up on all aspects of the subject - otherwise it's like trying to learn a language without studying grammar - you are likely to end up with a complete nonsense! If musicians cannot understand the implications of any given piece and express these in a coherent way, the result will be musical illiteracy. Today, standards of musicality seem to be nowhere as high as they used to: there's a lot of technical competence about, but basic musicality, as well as obviously necessary skills such as sight-reading etc., seem to be much less prevalent.

'Technical standards, on the other hand, seem to becoming higher all the time, but therein lies a danger. Thinking in terms of the early music revival, the standards of period instrument playing 25 years ago in the UK were often atrocious; nowadays they are highly professional, but then they had only just learned how to play their instruments. It's taken that long to revive the tradition and although it's tremendous that the standards are continually on the increase, what is really worrying for me is that musicality is simultaneously being suppressed.'

But this seems to be almost a requirement-in recording sessions, for example, you have a producer who rules the roost, one who is often quite ignorant of either the meaning of the music, or the performers' intentions. Thus, we have a situation in which people who should know better, but who don't, are channelling good musicians into providing poor performances.

'It is also worrying that you play something that in the recording studio might not sound very interesting or is even flawed technically, but on playback sounds alarmingly good! This is a constant problem with recording and it brings about a false sense of perspective. 78rpm engineers didn't have the technology to make something sound better at the touch of a button, they simply recorded what they heard. In one sense, this might sound substandard, but there is also a naturalness about it that today we've lost touch with; you can hear not only good performances, but there is something behind the performance that is also very potent.

'Broadly speaking, I find that there can often be a lack of integrity in modern performances, whether live or recorded, which could easily be avoided were performers to maintain a clear sense of what they are trying to achieve by attempting to preserve a sense of tradition. Musicians cannot be automatons, since they have to respond not only to the music, but also to circumstances that govern the integrity of a performance. It requires not only an empathy with what they are doing, but also a deeper knowledge and understanding of culture out of which the music was born.'

Worrying words that will give performers and teachers alike much food for thought. The next issue contains the second part of the article, where David Roblou and John Woodford discuss problems surrounding contemporary opera performances.