I'm a Rondo Today

Extracts from a conversation with Gillian Weir

The great thing about the organ is that everything in it affects everything else; everything has to interface and to be directed to a specific end. I'm a rondo today; I will keep coming back to this theme.

The organ grew up, out of its own principles of design, differently in each era and in each country. In each instance the organ came first and the music exploiting its special features came later. Our difficulty, in the 20th century, is that we now have the music and we have to work in the reverse direction if we want to plan organs to play the body of music that is the organ's repertoire. Unfortunately, the issue has not been thought about with very much objectivity. Various influences have shaped the organ with nobody stopping to say “Hey, wait a minute, do we really want that particular influence?”

This is not an attitude that has to be taken with any other instrument, for several obvious reasons, but the organ has the problem of being half a sophisticated engineering project and half an artistic project. There are people who build it, play it or listen to it who love it for a number of extra-musical reasons; the most outspoken enthusiasts seem to be those seduced by its engineering marvels, by the rumble of the 32ft or by historical idiosyncrasies.

As a keyboard player who came to the organ from the general musical world, my first concern is that it should function well in its diverse and fascinating repertoire. I feel the only way to make sense of the design problems is to have a very objective idea of its purpose. For me there is only one reason that makes sense of everything: that it should be built to play its body of music. Now some people think that is what they are doing when they include and design a Cromorne just like one they've heard in a Clicquot, then claim the organ can cope with Couperin et al. But this approach – the eclectic organ – leads to an instrument on which one can't really play anything successfully because the varied styles so badly represented will not integrate. The vital factor in an organ is that it should be integrated; it has to be a total concept.

I very much regret the idea of fashion in organ design. Builders have said to me “Oh but they did that in the fifties” in rejection of some idea, not making any explicit criticism but implying that it was old hat. Or, by contrast, “we want to try such and such ...”. I certainly think that the organ can and should develop, but that the development should be in a clearly delineated and positive direction, not with the negative aim of seeking excitement by rejecting an idea simply because it's considered passé. Of course the reverse is also true. With the music itself as rudder (not merely the subjectivity of “a musical sound”) you can still have organs that are diverse, choosing in each instrument a different area of emphasis.

This is especially important in the concert-hall organ. After playing so many, in often superb new halls around the world, I am astonished that so few are successful. Usually the reason is that they were built to a prevailing fashion, without consideration of the concert-hall organ's special needs. The appropriate repertoire needs, for example, full scaling with rich, robust principals. The foundation stops, the ‘fonds’, must cohere with an intensity that functions like a large orchestra's string section, as called for by Franck and the later French symphonists and by Liszt, Reubke or Reger. A Cavaillé-Coll is not always ‘beautiful’ in itself; its merit comes from the effectiveness with which it projects the musical ideas of its time. Then there are other considerations, such as acoustics, which are not nearly so sympathetic to an organ as those in comparably-sized churches. Air-conditioning systems are another, not only their noise but also the sound absorption character of the duct openings.

An organ with one or two 8fts on the Great and a plethora of high mixtures is not going to suit the concert-hall repertoire, partly because of the need for a much more telling foundation in the appropriate repertoire (orchestral as well as solo) and for the acoustical conditions, and partly because those high pitches are often permanently out of tune due to erratic air conditioning. Also, mixtures that are too high and thin do not blend with orchestral tone. A scaling and pressure that favours the treble (as with Cavaillé-Coll) is especially important when one will need to dominate in a concerto (e.g. the Jongen Symphonie Concertante) sometimes with only a few flutes, and to project the soaring melodies of the music of this period. So few people play the concerto repertoire regularly that it is particularly misunderstood simply commissioning the currently fashionable organ-builder and following current fashion has led to many dismal failures.

In designing for the performance of music, you may find that its composer is unlikely ever to have heard an organ with all the characteristics his music needs, which is why simply copying an old organ is not necessarily any solution. I am puzzled as to why people in the 20th century are so timid about making artistic decisions. There seem to be two polarized positions: one in which ideas are rejected simply because they are thought to be out-of-date, and the other where nothing new is admitted, in favour of an unfocussed copying of historical examples. There is an analogous situation with performance. One of the most important things I've learned is that it is possible to have all the right facts and to draw all the wrong conclusions from them. In one's studies one must look for a unifying principle, which is not the same thing as a rule. For example, in performance, articulation and fingering must serve the unifying principle of rhythm. Once the principle is discerned you can “play it with your nose” (Praetorius) because you now know why you are doing what you are doing. But blindly following someone else's rule won't necessarily produce the desired result. It's the same with organ-building; we need to think “What do I want this instrument to achieve, to be capable of ?” – apart from simply sounding ‘exciting’ or ‘thrilling’.

The polarization is expressed in organ-building as, on the one hand, “Let's do it our way; let's have an English organ”. The result is often an instrument that isn't right for any music even our own, but only for improvisation or accompaniment, both of which could have been served by an organ designed for the repertoire. On the other hand are those who say “Old organs are best”. The organ chosen for imitation is frequently not very effective in any specific repertoire but will have appealingly idiosyncratic mannerisms. (It is hard to discover how old “old” is – 1390, as at Sion? – pre-1800-? No clear answer comes.) It seems to me perfectly reasonable to say “I am of the 20th century; I want the organ's music to be played with fidelity to its meaning, it may even be that an organ can be built now that will serve that music more successfully than those of its own time”. Bach's son spoke of the poor organs his father had at his disposal for most of his life and indeed for the last 27 years of his life he taught the violin six days a week and conducted from the harpsichord in the organ loft on Sundays. His organ music reflects this: it is string writing (think of the great G minor fugue for example) and his counterpoint is best served not by the richly coloured but relatively unclear Silbermann but by an instrument with polyphonic scaling and voicing that will display the true glory of the music. Similarly, the logic of the werk-prinzip perfectly suits the inner logic of the German baroque school.

On the other hand, French Classical music grew entirely out of the organ of its time. It celebrates the vibrant colours and the way they are balanced, and the music is wedded to these characteristic sounds. If a good copy of a superb example could be made one would have an ideal instrument for that repertoire. But it would not suit counterpoint and of course would lack an independent pedal division. And that is to consider only two of the national schools. Clearly there is an enormous problem in attempting to build an organ that will play all the repertoire and I would agree that it is extremely difficult to do. Indeed it is clearly not possible to have a single instrument that plays Bach and Widor, Cabanilles and Howells, absolutely ideally. But it is possible to devise an organ that serves the most important elements of the music of each. However, to do this the idea of copying a historical example must be abandoned.

Today, one would like each builder to develop a style that is not a copy but which embraces those principles that are fundamental to the success of an organ physically and acoustically and which makes use of those factors in the tonal design that relate to the needs of the repertoire. I do not see the point of building an organ which, for example, lacks the obvious benefits of encasement, has a rudimentary and insensitive action or poor quality pipe-work, or in which one division is masked by another, simply because some organ of the past engendered these shortcomings. This is to put either the visual aspect of the organ or some other consideration before what should surely be the function of a musical instrument: the best possible performance of its literature. Yet often we see such an organ not merely tolerated but proudly displayed, even presented as “a challenge” should one timidly attempt to point out that it cannot do justice to the music. I do not see the proper performance of music as an obstacle course to be run for the fun of it; such an attitude just seems to me irresponsible, as well as disappointing and wasteful. A museum piece is of course another matter; but a vital, living instrument should be purposeful and functional.

What are some of the ways in which the apparently irreconcilable can be reconciled? Polyphonic scaling will not make an organ unsuitable for French Classical music but it is essential for counterpoint, so perhaps we should start with that. What to do about the fact that French mutations were much fatter than German ones? Considering the way mutations are used will help to resolve that, a scale a little wider in the tenor than is customary with a German organ will make a tierce en taille possible without jeopardising their use as a sesquialtera elsewhere. Will it spoil a chorale prelude if the solo mutations are a little wider than their historical counterparts? No, but jeux de combinaisons won't work if they are too thin, as the pitches won't cohere. Positioning needs careful thought, to determine on which division a colour will appear (apart from the choruses etc.), depending always on how it will be used in the repertoire. And so on; an intimate knowledge of the music and the way it is used will be necessary, but that should not be too much to ask.

Encasement is so vital to the projection and colouring of the sound that it seems to me to be a quintessential requirement. The idea it embodies of an individual character for each division is deeply imbedded in more schools of writing that the German baroque. I can show this principle at work in Liszt and Reger, for example, together with the allied ideas we refer to in shorthand as the werk-prinzip; the unity in diversity thus produced and the consequent illumination of the inner logic of the music helps to make the art of registration so much more than a game of playing with colours, and extends it to become the artist's way of conveying the structure and thus the message of the whole work.

I am very much opposed to placing the second manual division in a Swell box because I believe the basic principle of the main corpus of early music, including Bach's, is that it is a dialogue. This idea parallels the Baroque orchestra's ripieno and solisti; the music is a discussion, with a Voltaire like poise and elegance: “you put your view, I'll discuss it and put mine, and then we'll come to an agreement”. The Great or Hauptwerk puts his view first; it's then discussed by the Positiv, usually based on a pitch one octave higher and placed in a more intimate relationship with the hearers, being closer to them. It also has a different sparkle to its sound, being in a smaller case. As agreement is reached, the manuals are coupled the protagonists unite. Then you have not simply a louder or fuller sound but the completion of the chorus to its full pitch, the Positive chorus now on top of the Great; and the drama of the work's conclusion is reinforced as the two distinct divisions become one. The way in which the sound forces are used reflects the structure and emotion in the music. There's no need for frequent colour changes, legitimate and profitable in Karg-Elert or Messiaen but only confusing here. But if this equality of the divisions is threatened by enclosing the second division in a Swellbox with its consequent weakening of projection, something quite fundamental is destroyed. For me the organ is based firmly on the extended mixture, derived from the harmonic series, that began it all and that we now have in the shape of choruses; its design starts by recognising the idea of two equal but subtly defined divisions as its essential basis.

Much more diversity will be in order when one comes to the third manual division. Until then, unless one is working towards a specific organ for a repertoire intentionally limited, it virtually designs itself in its essentials; the stop-list could be written down very quickly. Much diversion from the basics will rob the organ of its ability to play the basic repertoire while adding little of any use musically. (I am often surprised by the stop-lists of new organs that manage to combine redundancy with inadequacy in astounding measure, the result of a kind of eclecticism that treats each division as the representative of a different school, thus crippling it for any music except that for single manual. The frustrated player attempts one stylistically appropriate registration after another without success, while at the same time wondering what to do with three 15ths all exactly the same.) If a Swell is to be the 3rd manual division, you need one that performs as is expected in Vierne symphonies or modern and Romantic concertos. For this it must be virtually a second Great in power, but will also need the lower-pitched multi-rank ‘Plein Jeu’, another reason a Swell as 2nd division is less than satisfactory, as it will not answer the Great, as described. It seems to be thought that Romantic music needs primarily certain colours, such as Harmonic Flutes or Voix Celestes, but these will not of themselves provide an organ that will successfully play symphonic music. They arc not in any case born of that period but derive from much earlier antecedents. What Romantic music does need is power; the point is to overwhelm. It argues in a way different from the classical and civilised dialogue of earlier times. In a Vivaldi concerto the outer movements present a cheerful face, the inner movement a more thoughtful idea, but these differing emotions are not on the whole presented in conflict (Bach's G minor Fantasia being an obvious exception in this genre). In Romantic music, on the other hand, opposites (God and Man, Life and Death, Good and Evil – call them what you will) enter into conflict and one of them emerges triumphant, perhaps incorporating the other into itself as in, say, Franck's Pièce Héroique. And to overwhelm in this way a large space is needed; it is as unproductive to play a symphonic work on a small organ in intimate surroundings simply because it contains so-called Romantic colours as it is to try to project a miniature in a vast building.

Specific colours are not the issue; the stop-list must be one of the means to an end, not the end itself. I am in love with the organ; I can be enthralled when the personality of a superb instrument speaks. But first and foremost, it is vital to me that it purveys, presents and projects the music itself. If it happens to be built by someone who has a spark of genius then it will be subjectively beautiful as well, but that is a bonus. Certainly it is always a disadvantage when the organ-builder/designer is not himself a voicer. The success or otherwise of an instrument is determined when it is on the drawing-board (although an otherwise completely successful organ can, sadly, still be ruined at the finishing stage) and it is not going to be successful if every part of the organ-building process is not already conceived and foreseen in the mind of the designer right at the start of that process. Some builders have organised their businesses splendidly with efficient computerization and splendid workmanship but, of itself, this efficiency will not produce a successful organ. Without some musically aware people who can both think in strictly functional terms and also say “This is my vision; I can hear it already”, an artist who can direct the entire process, you are never going to have an organ with any real personality; it lacks direction. The voicing process is not subjective: it is the final stage of the design process, the final step in giving the organ its voice, the last element in the integration of the myriad facets that make up an organ. To do this well it is not enough simply to cause the pipes to make a sound that a voicer considers pretty; they must make the sound that they were designed and created to make when the colour of that sound was being planned in conjunction with all the related factors. This is understood fully by very few people – which is why we prize so highly those famous organs that have succeeded and that touch us so deeply.

Great artists have a goal, they have a vision, they have a clear idea of where they are going. What they do comes therefore from a positive standpoint. A common mistake is to react so strongly against the immediate past that one goes too far the other way; this is seen in architecture, and in music it happens often with composition. In fear of being called derivative, composers will try to eliminate from their work any echoes of their predecessors. But a worthy work of art cannot come from a negative standpoint, merely the rejection of a set of ideas, any more than it is likely to result from a mere copy; a work of art is something that its creator has to express, an aspect of truth. The organ has to be functional and, to progress needs to combine originality with purposefulness. Others have accomplished this in the past, for example Cavaillé-Coll, a great innovator. I believe that we might not have organs at all now, except more or less as museum pieces, were it not for Cavaillé-Coll. He made many technical innovations that proved invaluable in developing the organ for the music of his time. We must maintain continuity with the past but neither endlessly repeat it nor thoughtlessly reject it. I see no reason that imagination should not flower, as long as it is directed always by an overriding principle that unifies the design and gives it functional purpose; just as a table can be made in G-Plan, Chippendale or Victorian style as long as it is flat so that things placed on it don't fall off.

Another example of riding off in the wrong direction as a reaction against the immediate past is seen in the aggressive so-called ‘neo-Baroque’ organs, which weren't baroque at all; a visit to North Germany would show that the originals are singing, not shrill. The distortions in the modern instruments resulted from a reaction against the muddiness of early 20th-century organs; the ‘American Classic’ organ was an extreme example of such a reaction. What was needed was not brightness but clarity, two completely different things. Brightness can be a gentle glitter but it can be harsh and unclear. Clarity comes from the design of the organ; the balance of the scaling, giving the pipes space to speak, thinking polyphonically. These are all fundamental physical principles, transgressed at our peril because we end up with a compromise. Compromise may be a diplomatic blessing, but it is an artistic tragedy, and in artistic projects in Britain it sometimes seems not merely tolerated but revered, almost a credo. That is not to say one should be rigid in one's view, but rather that it is important to be quite clear about what is essential to success and what merely peripheral, which means working with a clear purpose and working with conviction.

My whole argument, really, is that we should have a reason for building an organ; that reason should be the best possible performance of its music. To this end our builders need to understand the whole of the repertoire, and we must remember that everything in an organ affects everything else; it all interfaces and is a total concept. That's where I came in; I'm a rondo today.

Gillian Weir's career as one of the world's foremost concert organists has included regular performances with major symphony orchestras and leading conductors. Her vast repertoire ranges from the pre-baroque to first performances of works by Peter Racine Fricker, William Mathias and Olivier Messiaen. Her achievements were recognised in 1989 by the award of the CBE. She is the first woman organist to have been so recognised.

Gillian is married to Lawrence Phelps, the well-known American organ-builder. Her comments on such matters as layout, casework, tuning methods and tracker action will be the subject of a further article in the next issue.