My subject is a stepchild -- live recording. Standing between the two officially canonized sources of musical experience, concert performance and studio recording, the recorded concert has had less than its due.

There has been a good deal of discussion about the differences and similarities between concerts and studio recordings. I should like to offer my own catalogue of distinctions (bearing in mind that a concert hall may turn into a studio if recording sessions take place in it).

In a concert one plays just once, in the studio several times if necessary. In a concert you must convince the audience at once; in the studio it is the accumulated result that counts.

In a concert the performance is only experienced once; in the studio it can be reproduced. In a concert the performer must get to the end of the piece without a chance to make corrections. In the studio he can make corrections, learn while he records and get rid of nerves.

The player before the public must do four things at the same time: he must imagine the performance, play it, project it and listen to it. In the studio he has the opportunity to hear it again after playing, and to react accordingly.

In a concert it is the broad sweep that counts. The studio demands control over a mosaic; while it offers the performer the possibility of gradually loosening up, there is also the danger of diminishing freshness. And there is the painful business of choosing between takes.

When playing before the public, details must be projected to the furthest ends of the auditorium, just as the whispers of an actor must be heard throughout the theatre. In front of the microphone one tries, on the contrary, to get away from exaggerations and aims for an interpretation that will bear frequent hearing.

In the concert hall the concentration of the audience brings about a mutual influence between the performer and his listeners. In the studio nobody has to be conquered -- but there is nobody to disturb you. The player sits as though in a tomb.

A fit of coughing or the chirping of the alarm on a watch may break the spell of the most delicate moment of the concert. The studio offers silence.

Weaknesses in a concert performance tend to result from spontaneity, from a break in concentration or from nervous pressure. In the studio they may have their roots in excessive critical awareness.

The ability to convince the public in the concert hall is quite independent of absolute perfection. The studio is ruled by the aesthetics of compulsive cleanliness.

All these are observations from the player's point of view. Concert-goers and listeners to records may like to add that a concert involves physical presence, while the 'pure music' of the record avoids it (a bonus for those who suffer from agoraphobia or feel uncomfortable in crowds); moreover, the sound reaches the listener unmanipulated and as directly as the acoustics of the concert hall permit. The sound of the recording, on the other hand, is decided by the technical staff, the musical effect depending on such factors as editing, balance, reverberation and the qualities of the reproducing equipment. Lastly, not only must the player perform an entire work in a concert, but the audience must sit still and listen until it is finished. (People rarely leave a concert during a performance. Such respect for the concentration of both musicians and audience is one of the tacit agreements of a cultured public.) But when you listen to a record you can turn the music off, savour it in instalments or try bits here and there; you can move, talk, eat and groan -- in a word, you feel at home.

'Accuracy and Soul'

Despite the funeral orations Glenn Gould delivered on concert halls, they continue to be the setting for the most vivid music-making. I do not wish to be dogmatic and will admit that there are concerts without a breath of life, and records of electrifying vigour. All the same, it follows from the way they come about that concerts are more likely to be characterized by spontaneity, tension and risk, studio recordings rather by reflection and superior method. To quote Robert Musil's The Man without Qualities, with its 'Generalsekretariat für Genauigkeit und Seele' (Administration of Accuracy and Soul), I may say that in the studio accuracy is more readily manageable than 'soul'.

Studio recordings have enormously increased the acuteness of detailed listening, including that of the musician listening to himself. In conjunction with the influence of modern Urtext editions and the demands of contemporary music, the gramophone record has profoundly upset listening habits. Its effect on the player, however, may not only be concentrating and distilling. The interpreter who aims at accuracy risks less panache, lesser tempi, less self-effacement. The gramophone record today sets standards of perfection, mechanical not musical, which the concert hall seldom confirms. It induces some artists to play in a concert as though for a record, in the fear that the audience is listening as though to a record.

But a concert has a different message and a different way of delivering it. Now that we listeners to records and studio troglodytes have learned so much from studio recordings, it seems time to turn back and learn from concerts once again.

Players and Public

For the sake of objectivity, let us consider the recordings of the 1930s, such as those of Cortot, Fischer or Schnabel. One may not have been aware then of certain imprecisions, in the way modern wrong-note fiends are. Where the leading of voices, the grading of dynamics, the control of character and atmosphere, timbre and rhythm are handled with the mastery of Cortot at his best, it appears to me that momentary lapses in concentration are not only irrelevant but almost add to the excitement of the impact.

In the 1930s people seemed to have played in the studio almost as in a concert. But was this really so? Even then, players must have been worried about providing lasting evidence on a record, unless they could summon up the unbelievable nonchalance of a Richard Strauss. Apart from that, the limited duration of the 78 rpm disc was basically at variance with the nature of playing longer pieces as a whole. But then, as Emil Gilels told me about his own early recordings, a side may have had to be repeated thirty times if the producer so commanded, and the players had not opportunity to hear the results themselves, since a wax matrix was destroyed by one playing.

Above all, there was no audience. Why, if I may believe my own experience as a listener, does an impressive concert tend to leave stronger traces than a record? Because the listener, no less than the player, has had a physical experience, not only hearing the performance but breathing in it, contributing to it by his presence and sharing his enthusiasm with many others. The listener encounters the composer together with the performer and the rest of the audience in one place and at one time.

In the studio the player is alone with his own self-criticism and the Argus ears of the producer. Even if he possesses the important gift of playing there with all the tension of the concert platform, and however vividly he might imagine the presence of the public, it is still imaginary. There is no direct exchange. He will, of course, try to remain as close as possible to his concert performances, using takes of a complete piece as a basis. But whoever subscribes to the belief that tape editing is a deception and that only complete takes should be used, deceives himself; he would renounce the advantages of the studio and still fall short of the enchantment of the concert, for it is not just the tension of the single uninterrupted performance that counts. (No one listening to my records could tell which movements remained unedited and which were put together from a number of takes.)

Here the live recording serves as a bridge. What is it able to convey? For me, there is above all the attractive feature that the uniqueness of a concert has been thwarted. The concert took place on a certain day: the public was present, as we can hear in the background, and we can imagine being present ourselves -- a fancy much less ridiculous than wishing to imagine ourselves in the bare studio.

What you hear and enjoy is an indiscretion, something that was only intended for those present and cannot be exactly reproduced. It is not the technical level of reproduction I am referring to: the fact that live recordings cannot always achieve the quality of the best studio products hardly worries me. It is the participation of the public, the aura of physical presence, the contribution of which cannot be altogether assessed on a live record; and yet, in some happy instances, these leave their mark in the heightened intensity of a performance, in the increase in the player's vision, courage and absorption.

'Trouvaille' or Production

Why have live records been so rare until now, except for those of famous artists of the past enjoyed by connoisseurs and collectors? First, because a concert becomes more difficult when it is recorded. The sight of microphones on the podium does not fill the artist with glee. Incidentally, one must make a clear distinction between radio recordings and live productions for commercial gramophone records. The former are more easily bearable since they will only be broadcast once or twice, while the latter are bound to terrify the player, being aimed at an international body of critical contemporaries, and future generations. Live productions are therefore only worthwhile in special cases, one of which I shall mention later. On the whole, live records should come about by chance; they should use radio or private recordings that give the artist pleasure. (Of course it is outrageous that, in some countries, they are still sold without the artist's agreement).

Here we come to the second difficulty about live records: the prejudice against their alleged technical, and even musical, inferiority. Losses in digital quality and realistic balance, accidental noises, inaccuracies in the playing or fatigue in the instruments are mentioned as deficiencies that cannot be tolerated. True enough, there is no call to make a commercial record of a performance that has caught the interpreter off colour, the public during a flu epidemic, or a fleet of fire-engines passing by. Apart from that, the latest developments in recording technique will sometimes make an expert in electroacoustics happier than a musician. There are chance recordings that bring a piece of music to life and studio performances that destroy it. Those who consider spotless perfection and undisturbed technical neatness the prerequisite of a moving musical experience no longer know how to listen to music.

Speaking for Myself

In pleading for live recordings here, I do not by any means wish to turn my back on the studio. I have spent innumerable interesting and some happy hours in it, owe it much essential experience, and shall continue to acknowledge my records, though with certain reservations. But in future I should like to place more frequent live recordings next to them. My first live record was devoted to the longest masterpiece of the older piano literature, Beethoven's Diabelli Variations. Since then I have been waiting for a suitable performance of the 'Hammerklavier' Sonata Op. 106, and recently found one in a London concert given in April 1983. Why am I drawn, of all things, to the biggest and most dangerous works? Because they best provide evidence of a mastery that is not available to some 'studio artists' and because it is works of that scope which stand to gain most in boldness, absorption and vision. The objection that no player can function uniformly well for an hour or so, however justified in itself, misses the point of a concert performance. Compared with the evened-out results of the studio, it may show greater dedication and that unexpected success that differs from a premeditated result as a poem differs from a timetable.

Beethoven in Chicago

Although I would not normally wish to make a planned live recording, an exception came about in Chicago in June 1983. On this occasion the effort of performing the cycle of the Beethoven piano concertos was added to the risks of a live recording.

Performances of cycles make the stature of a composer more clearly recognizable. They are especially appropriate to great composers like Beethoven, who constantly have something new to convey. The unmistakable character of each movement, when played in close succession with the rest of the cycle, shows its profile even more distinctly to performers and listeners alike.

The Chicago Symphony Orchestra, James Levine and I are old acquaintances. My contact with that splendid orchestra goes back to 1970 and has continued with welcome regularity. The Ravinia Festival of 1977 gave us the opportunity to play all the Beethoven concertos under James Levine. On the first evening the temperature was around 100°F and humidity 95 per cent, which disturbed the orchestra not a whit: concentration and control of the playing remained virtually unaffected. The cycle was repeated two years later. Finally, in June 1983, two series of Beethoven's concertos were recorded digitally in Chicago's Orchestra Hall.

The undertaking had a double goal: we intended to examine and to realize a concept of these works in several stages. The rehearsals were used to go, among other things, into the difference between sforzando and forte piano. Levine and I found ourselves in friendly agreement about what Beethoven's scores communicate, while the musicians of the orchestra never tired of re-examining pieces they had long ago mastered.

At the same time we hoped to document that tension and directness which manifest themselves more readily before the public, a kind of spontaneity within pre-set boundaries which would rather discover than reproduce.

The confidence of all concerned in one another was the safety factor that made the risk of this live recording for once a calculable one, but it would have been extremely unwise, not to say foolhardy, if only for technical reasons, to have relied on the tightrope-walk of a single series of performances. The availability of two cycles gave us the possibility to combine some benefits of concert and studio: the freshness of the moment with the advantages of having a choice. I am not giving away any secrets if I say that live productions nearly always work in this way, seeking a synthesis or a compromise between both worlds.

All the same, despite every precaution, the audience could still have spoilt the lot. Over the years I can remember concerts with screaming babies (Japan), a barking dog (New York), a mewing cat (Istanbul), somebody falling down in a faint, a maniac clapping in the most impossible places, and a power cut plunging us all into darkness. (In Chicago itself I once had to stop during a recital, a few bars into the hushed beginning of Liszt's 'Sposalizio', and tell the audience: 'I can hear you, but you can't hear me.') None of these things occurred; on this occasion the exceptional stillness and concentration of the Chicago public filled me with gratitude. It was almost possible to forget how dangerously one lives when one records live.