The Tonal Structure of Organ Principal Stops
by Colin Pykett
Posted: October 2006
Last revised: 21 December 2011
Copyright © C E Pykett 2006-2011
Abstract An earlier article on this website looked at the tonal structure of organ flutes. It is quite a long article, mainly because of the diversity of flute stops and their different characters. This article takes the analysis further to examine in a similar manner the wide variety of sounds obtained when the dimensions of open metal flute pipes are varied by relatively small amounts so that they become Principals. In fact, it is remarkable in itself that our ears and brain assign a quite different perceptual character to the two classes of tone when the pipes which give rise to them are not so very different in construction. As with flutes, the range of different principal tones which exist is explored by relating it to the harmonic structures of the pipes. Other factors are also investigated, including some of the fads and fashions which have come and gone in principal stops.
(Click on the headings below to access the appropriate section)
An earlier article on this website  discussed the tonal structure of organ flute stops in detail, and this one now covers principals in a similar manner. Originally the article was to be entitled The Tonal Structure of Diapasons but this was deemed too restrictive - the term principal is better as it avoids unhelpful national connotations. Thus not only Anglo-American diapasons but the French montre and prestant are included, as well as the Italian principale and German prinzipal.
Probably the majority of discussions of principals by British authors are limited to the history and development of the Anglo-American diapason, emphasising again how important it is not to become trapped within national stereotypes, though an exception was the mid-twentieth century work of Bonavia-Hunt  which included all the classes of tone just mentioned. As an example of the narrower view, Sumner  defined a prinzipal merely as "the nearest equivalent to the English open diapason on German and Italian organs". That was it - not really a definition at all, and a semantically ambiguous sentence at that. On the face of it Williams and Owen did better with a more rigorous definition, ending with the statement that the diapason "helped to hinder the development of the English organ, its very name giving the stop a mystique absent from more traditional terms like Montre and Principal" . But a few pages further on they reveal a basically English take on the matter when they say that "Prinzipal has become useful as a term denoting the relatively colourless German basic 8' rank as opposed to the various English Diapason tones" . In passing, I wonder whether our Scots, Welsh or Ulster brethren have been annoyed by the rather thoughtless use of the word "English" rather than "British" by these and other authors? If this sounds irrelevant to non-Brit readers, think of it as a bit like an American writer who happens to live in glorious San Diego (ah, that climate and sea food!) referring to a Skinner diapason as "Californian".
So let's call them Principals rather than Diapasons or anything else when speaking of them generically, and put aside all this chauvinistic nonsense which does not seem to afflict discussions of flutes, strings or reeds nearly so much.
Principal tone is the hallmark of any organ worthy of the name (except the theatre organ, though even these often have a principal rank of sorts). It is a non-imitative tone unique to the instrument, and principal pipes have been used for centuries to create the major choruses of the organ from the lowest pitches to the highest, the latter in the form of mixture and mutation stops whose ranks sometimes go beyond the range of audibility of many people. The general appearance of a principal pipe is sketched in Figure 1, such pipes adorning the case of virtually every organ ever made. However if the pipe was to be turned round to conceal the mouth, even an expert could not always tell whether it was a principal or an open metal flute, because the scales of both types of pipe are comparable. Scale is a measure of width to speaking length, and for a 2 foot pipe (middle C of a unison 8 foot rank) the internal width of principals varies between 1 1/2 to 2 1/4 inches (38 to 57 mm). Open metal flutes are typically about 2 inches (51 mm) in diameter.
Figure 1. A Principal pipe
So if the pipes are similar, why do principals and flutes sound so obviously different? The clue lies above, in that the mouths of the two are different in detail. The main distinguishing feature is the cut-up, the height of the upper lip above the lower. Principals have a cut-up of about 3/8 inch (9.5 mm) at the middle C pipe whereas open flutes have have higher mouths, at least 1/2 inch (13 mm) and often more. In addition, the upper lip of flutes might be formed with an upwards arch. This is all very well, but how do such relatively small geometrical differences in the mouth parts lead to such pronounced differences in tone? To answer this question we need to move on to look at the harmonic frequencies emitted by the pipe.
The earlier article dealing with flutes  showed that their frequency spectra, representing the strengths of the various harmonics, were almost always characterised by even-numbered harmonics of lower amplitude than the odd ones. This is true whether the pipes are open or closed. However it is not the case for principals, whose spectra invariably have stronger even harmonics than flutes, particularly the second harmonic which has twice the frequency of the fundamental and therefore sounds the octave above it.
Another article elsewhere on this website  explains in more detail how flue pipes speak, and it contains an example of a principal spectrum. This diagram is reproduced here at Figure 2. The strengths of the harmonics fall off gradually, and although they jump up and down a bit, there is no evidence of the systematic separation between the odd and even harmonics which is so characteristic of flutes.
Figure 2. Typical acoustic spectrum of a Principal pipe
The total number of harmonics, 13 in this case, is generally greater than the number emitted by flute pipes although this cannot always be assumed to be a reliable distinguishing feature. I have analysed some of Gottfried Silbermann's flute pipes to find that they generate 15 harmonics or more in some cases, yet they still sound fluty compared to his principals! More than anything else, it is the strength of the second harmonic relative to the others which determines whether a pipe sounds like a principal or not, and moreover it is this feature which also largely characterises the tone of the various types of principal which have been made by different builders over the last few centuries.
The main reason why the even-numbered harmonics are stronger in principals than in flutes is because the upper lip is offset from the lower - because principal pipes are made of metal rather than wood, the upper lip can easily be pushed in or out relative to the lower. The offset means that the oscillating air sheet at the mouth generates both odd and even harmonics because it excites the air column within the pipe asymmetrically. As the sheet flips back and forth across the upper lip, it does not spend equal amounts of time inside and outside the pipe because of the offset lip - in other words its mark-space ratio is asymmetric, and a Fourier analysis of the resulting pulse-type waveform shows it contains both odd and even harmonics. In contrast, for a flute pipe rather than a principal, the voicer in effect ensures that mark-space ratio is more nearly symmetrical (1:1) by adjusting the lip offset to zero or thereabouts (this happens automatically for most wooden flute pipes because of the way they are made), and such a waveform contains mainly the odd harmonics.
Even if there is no lip offset, substantially the same effect can be produced by directing the air jet at an angle towards the upper lip through adjustments of the position of the languid relative to the lower lip.
The lower cut-up of a principal pipe is responsible for the generation of more harmonics than many flute pipes exhibit (stand fast the remarks about Silbermann's flutes made earlier). This is because the air sheet increases in width as it emerges from the flue at the lower lip, and therefore it will be narrower at the upper lip for a pipe with a low cut-up than it would be if the cut-up was higher. The narrower width results in more harmonics being generated because the generating impulses are sharper.
As with the earlier article on flute stops, it is not possible to include every conceivable type of principal in this one, so just six examples have been chosen to illustrate the main directions of development over the last 300 years or so. They have been selected, and will shortly be described, using an abridged version of Bonavia-Hunt's taxonomy of principals . It might be helpful to listen to them first before diving into a detailed discussion about each one, and the mp3 file below contains an identical clip played on each stop courtesy of Mendelssohn. All of the clips are digital reconstructions made from recorded samples of the actual pipes, because I flinched at the thought of having to travel thousands of miles just to record a few seconds of music on each of the organs. However I am confident that what you hear is a tolerably faithful reproduction of how the original stops would sound when rendering the same piece, subject to the characteristics of your particular listening set-up of course. I felt it was better to use digital approximations in this article rather than none at all, as it avoided having to rely entirely on the vague adjectival descriptions of how pipes sound which have been used in all other work to date. If you object to this then don't listen to them, just read on.
Various Principals - 3.28 MB/3m 35s
The stops, all at 8 foot pitch, were recorded in the order shown in Table 1 below, which shows the name of each stop endowed by its builder and its approximate date.
Table 1. Principal stops as used in the recorded examples (listed in order).
After listening to the clips for the first time it would not be surprising if your initial reaction was "so what?". Apart from one or two obvious differences, you might find the examples are characterised more by their similarities than by their contrasts at first. If this is your experience I recommend you try listening to the file several times, after which you will probably be able to draw out more from it. It is an unnatural listening experience to compare such short musical snippets from several stops which are all from the same family. The following discussions might assist also.
In many ways Silbermann (Gottfried, not his brother Andreas) can be considered the inventor of the principals we have today, and to justify this assertion it is first necessary to look at what he inherited in craft and tradition when he started as an organ builder. This means we have to look at the work of Arp Schnitger first.
Until the early 1700's, principal tone in Holland and much of Germany was more or less defined and dominated by the principals in Arp Schnitger's organs and those of his disciples. These were of quite restrained power by modern standards, often comparable to the flutes. Even their tone quality inclined to flutiness rather than the brighter and more edgy tone which we associate with principals today. Some writers imply more certainty about the pipe scales in those far-off days than the evidence warrants, but insofar as they can be ascertained they were usually simple in that pipes of the same pitch in different stops (e.g. middle C on an 8 foot principal and tenor C on a 4 foot one) were often rolled on the same mandrel. Within a rank the scales sometimes appear chaotic to our eyes, though this might be due to the frequent incorporation and re-use of very old pipework by these builders. The desired differences in tone and power across a rank and between ranks were largely introduced by the voicer who varied the cut-up to alter the number of harmonics in the sound and hence its subjective brightness. Mixtures were not allowed to scream in the trebles by lavishing attention on each and every pipe, rather than relying more on pre-determined pipe scales and sloppier tonal finishing as we generally do today.
Pipes were regulated in power by adjusting the width of the flue slit rather than the size of the foot hole, which was large and unobstructed. This meant that the wind pressure appearing at the languid was more or less the same as that in the chest, and its rise time at the languid when the pallet opened was rapid. These factors endowed an overall promptness of speech and sometimes (though by no means always) a pronounced attack transient. At close quarters the pipes often sounded breathy because of the narrowness of the flues. Although the technique of nicking the languids to control the onset of speech (cutting closely spaced slots into them with a knife) was known, it did not seem to be used often. Wind pressures were low or very low, between 2 and 3 inches (50 to 75 mm) or less, thus the pipes were usually underblown in the sense that they were unlikely to fly to the octave when beginning to speak - quite different to many neo-Baroque organs today. The voicing and finishing processes of these 17th century organs are said to have occupied an enormous amount of time which would be inconceivable today on grounds of cost. Considering that the incomparably skilled journeymen who actually built the organs decamped with their families and lived on-site for years until the job was done, this can be understood.
All this helps to explain why Schnitger's principals had a beautiful transparent quality, and a principal chorus built up in this way was penetrating without being overwhelming, even though 11 ranks or so of mixture work could quite possibly be singing away simultaneously on a single department of a two manual organ. Senator Emerson Richards, the designer of the largest organ ever built at the Atlantic City Convention Hall (as it was then called), seemed humbled when he first encountered a Schnitger organ in the 1940's. He wrote of this epiphany:
"Full organ fills the rather large church with a flood of pure tone - no rumble or muddiness. Bach ... came out with an entirely new meaning. A precise, bell-like tone, rich in harmonics, but characterized by a lightness and transparency, gave an interest to the music never achieved by the romantic organ to which we are accustomed. There is plenty of power; the Mixtures are responsible for that; but it is a different kind of power. After becoming accustomed to it one never has the same interest in chorus reeds as instruments of power" 
To give some idea of what a Schnitger principal sounded like, the first example on the audio file above is of a digitally simulated Schnitger stop, based on the sounds of real pipes as are all the subsequent ones. It is of interest to bear in mind that principals like this might also have made an appearance in Britain in the 17th century with the work of "Father" Smith, who learnt his art on similar organs in Holland or north Germany before arriving here in the 1660's.
So if Schnitger's principals were so wonderful, why have they not survived? The answer to the question is similar to that which asks why J S Bach was virtually forgotten for so long after his death. He grew up to the sound of Schnitger's organs and similar ones and his music fits them sublimely, but both he and Schnitger were the end of a line. Once humanity reaches perfection it moves onto something else, and that is what happened when Silbermann came on the scene in the first half of the 18th century.
He satisfied a need which was demanding more colour and power in the individual stops of an organ, as opposed to their choruses. Although he continued to provide magnificent chorus work, it was of a different kind to that of Schnitger. Moreover, because of the move away from Bach's style of restless contrapuntal polyphony towards compositions with an emphasis on melody, he ceased to provide complete Schnitger-style pedal organs in his smaller instruments, giving them only a small handful of stops with a manual coupler. Do not misunderstand my position - Silbermann's organs were just as breathtakingly beautiful as any Schnitger built, but they were built to serve different and evolving needs.
Silbermann used much higher wind pressures even in his smallest instruments, typically up to 3 3/4 inches (95 mm). Thus his principals were powerful, and a principal chorus could flood the largest spaces with sound. He made his ranks according to better-defined scales, halving their diameters every 17th note for some of the smaller pipes. This halving interval (used in so-called Normal Scaling) still forms a reference datum for the calculation of all the flue pipe scales used today. Below this his pipe widths varied more rapidly, and above it they varied more slowly. Such scales assisted the upwardly-voiced effect of his principal choruses, which were bright and colourful without screaming in the treble, yet not too thin and scratchy in the bass. These characteristics can be detected to some extent if you listen again to the clip of a Silbermann principal above, where they were carefully preserved in the digital simulation. The development of these scales was a quite extraordinary achievement for its time, and it must have played a significant part in what people punned as the silberklang - silvery sound - of his organs and to which Mozart himself alluded. Bach was apparently less complimentary, complaining about Silbermann's temperaments as well as the shortcomings of the pianoforte whose development the indefatigable Gottfried was also pursuing vigorously.
To enhance the final result, Silbermann was not averse to using foot hole regulation and nicking more frequently than did his predecessors when tonally finishing his organs. And to bring costs down, he also invented the concept of the factory organ in which the maximum possible degree of uniformity was included in his designs. The difference between him and others who later followed his lead was that Silbermann was a consummate genius who, on the basis of what his surviving organs tell us, would never sacrifice art for profit.
A major reason for the differences between the sounds of Schnitger's and Silbermann's principals can be discovered by examining their acoustic spectra. Those of Schnitger almost invariably have a second harmonic which is weaker than the fundamental, typically by between 9 and 15 dB (this means the second harmonic is only about 0.36 to 0.18 the strength of the fundamental). By contrast, Silbermann's second harmonics are frequently stronger than the fundamentals, typically by 6 dB (twice the strength of the fundamental). Otherwise they are of comparable strength and only seldom are they weaker, even then by only a few dB. These characteristics can be associated with little difficulty with the scaling and voicing practices used by the two builders.
I mentioned above that these types of principal sound "transparent", but what does this term actually mean? I cannot speak for others who use it, but when I use it here it means that every note being played simultaneously can be separately perceived. Thus in the audio clips above of homophonic (chordal) music played on these two principals, you can hear each note of every chord, including in the part of the compass below middle C. In polyphonic (contrapuntal) music you can likewise hear each of the parts in a fugue, say, moving independently of the others. This definition of the term is important because it will be examined again for each of the other types of principal to be discussed. But what factors in the tone result in transparency? It arises largely because the sound energy is spread across a number of harmonics whose strengths decrease relatively gently after the first (fundamental) and second. Another way of saying the same thing is that excessive power is not forced into these early harmonics, as it is in certain types of flute tone and later types of principal. If this did occur, the ear would be swamped by the amount of power in the fundamental and second harmonic and this would reduce its ability to perceive other notes sounding at the same time, especially lower in the compass. A full explanation of the matter is complex and related to issues such as masking which form part of the neural mechanisms and psychology of aural perception, and it cannot be taken further here.
After Silbermann's death the development of the German organ more or less ceased, never to regain its former pre-eminence. Both it and the works of Bach fell into comparative obscurity, although fortunately Silbermann's tonal achievements were quietly noted and used by various builders in other countries. This is why he could be said to have invented today's principals, which brings us to the next instalment of this story.
These are, of course, synonymous with the name of Aristide Cavaillé-Coll. His Montres (the 8 foot variety) and Prestants (4 foot) were key ingredients of the jeux de fonds. Respectively, they derived their names from the front show pipes and those which stand in front, i.e. also on show in the organ case. They are often said to have had a "hard" tone which might imply a lack of blend with other stops, though quite the opposite was the case. The way they blended with other stops of the same pitch, particularly the Flûtes Harmoniques, resulted in some of the most characteristic and beautiful sounds of the romantic French organ.
In effect, Cavaillé-Coll took Silbermann's principal and made it speak slower and with less harmonic development. The relative slowness of attack arose because of the time taken for the pipes to come fully onto speech, typically 35 cycles of the fundamental frequency rather than the 15 cycles or so which characterised Silbermann's quick-speaking stops. In turn, the slowness was due to the emphasis he placed on foot hole regulation rather than regulation at the flue. Constricting the foot hole meant that the wind pressure appearing at the languid was less than in the chest, and it also had a slower rise time when the pallet opened because of the constriction. By this means Cavaillé-Coll separated the function of regulation, to adjust loudness, from that of adjustment of tone quality by manipulation of the mouth parts. And because of the slow rise time a Montre will seldom speak with a chiff or any other sort of transient - its sound just seems to appear from within the depths of the organ when you key it. In fact, subjectively, the attack is not unlike that of the contemporary and popular French harmonium (not the same as the American organ) which seemed to be an acceptable substitute for the organ for many French composers of the day. Because Cavaillé-Coll used even higher wind pressures than Silbermann his stops were also louder, though seldom overwhelming.
The harmonic structure of a Cavaillé-Coll Montre pipe is unique and interesting. Seldom does the all-important second harmonic take a greater strength than the fundamental, and more often it is comparable or slightly weaker. In broad terms therefore, the frequency spectra were something between Schnitger's principals (which were fluty) and Silbermann's (which were bright and zingy). You can hear this on the recorded samples. But there is more. The spectra of many of his pipes seem to linger with significant amplitudes until the 7th-9th harmonics are reached before decaying more rapidly, and more than anything else this gives them their characteristic flavour - hardness, or whatever you want to call it. Using a digital synthesiser to reconstitute a Montre tone shows clearly how critical are the levels of these harmonics - too much and the tone is ruined; too little and it loses character. Cavaillé-Coll discovered and refined this property and bent it to his will, though quite what led him to it I do not know. He had no digital synthesiser. The skill and knowledge died with him, and as far as I am aware his techniques have not yet been properly analysed in the light of modern acoustical physics.
The method he used was to cut slots in the backs of the pipes, the dimensions of the slots and their positions being critical. He slotted virtually all his principals, sometimes even the tiny pipes in the top octave of a mixture. He was undoubtedly well versed in the acoustics of his day, having been something of a minor prodigy in mathematics and science in his youth, and because of the interest he took in the mathematics of pipe scales and in the end corrections of pipes, it is clear that he continued to hone his knowledge in these matters as his career progressed. He seems to have grasped that slotting a pipe was analogous in some ways to what happens when an orchestral player covers or uncovers the various finger holes in a woodwind instrument - the pitch changes, certainly, but so also does the tone quality because of changes in the harmonic spectrum. Although it takes a keen ear to detect this, it requires an even keener mind to understand what is happening and how to exploit it for the creation of deliberate tonal effects. I believe Cavaillé-Coll went a long way towards achieving this understanding.
Slotting the pipes was nothing new in the 19th century because it had long been used to reduce the effective speaking length of the principal pipes used in organ casework. Thus their actual lengths would match the visual aspect which the case designer wanted and it is still used widely today for the same reason, often unthinkingly in view of the effect on their tone. Slotting is also sometimes used, again unthinkingly, merely to assist tuning the interior pipes in an organ. However, because he slotted all of his principal pipes whether forming part of an external display or not, I think Cavaillé-Coll had discovered something which was not capable of being fully understood until well into the 20th century, long after his death. Unlike the tone holes in a woodwind instrument, the case of the slotted tube does not appear to have been researched in detail, but I have done an initial analysis which suggests that the slot itself does not radiate the fundamental frequency very much whereas the higher harmonics are radiated from it more efficiently. Those harmonics in the spectrum which lie above the fundamental frequency in effect seem to take progressively less notice of the slot as their frequency increases, therefore the sounds of the higher harmonics are radiated into the air with different efficiencies to those of the lower ones. This behaviour depends on the length of the slot and its position relative to the top of the pipe, and therefore adjustments to these dimensions will determine how the upper harmonics are emphasised.
Not only did Cavaillé-Coll apparently discover this effect long before we developed today's understanding of musical acoustics, but it is likely he understood it in theoretical terms too, at least sufficiently to be able to make entire ranks of slotted pipes with the necessary uniform harmonic enhancement . It was a remarkable achievement when seen in this light, comparable to that of Silbermann in coming up with his understanding of pipe scales, and it makes him perhaps one of the most "scientific" organ builders to date. His published work on end corrections confirms this .
Whether Cavaillé-Coll's principals could be called transparent is a matter for debate. Although they have a more ponderous sound than the early German principals discussed above, by and large they are certainly not as opaque as some of those now to be discussed. The maintenance of their upper harmonic structure thanks to the slot technique gives them a somewhat penetrating edge which is of value in much organ music.
An example of how principal tone had evolved in Britain and America by the 20th century is included with the sound clip above, in the form of a simulated 1930's Rushworth and Dreaper open diapason of moderately "fat" tone. One of the most obvious things about its sound is its opaqueness, in the sense it is more difficult for the ear to resolve the separate notes than in the case of Schnitger's, Silbermann's and (to some extent) Cavaillé-Coll's principals. In other words, the 20th century foundational Anglo-American diapason was no longer transparent. A reason for this is that the fundamental is almost invariably stronger than the second harmonic, usually by at least 15 dB (a factor of 5.5 in sound pressure level) and often more. Also the remaining harmonics tend to fall off more rapidly in strength than do those of the principals discussed above, and there is nothing comparable to Cavaillé-Coll's slots to maintain their strength. The speech of such pipes is often ponderous and slow, especially below the middle of the keyboard. All these characteristics are more pronounced for "large" diapasons than "medium" or "small" ones. They explain why the foundational diapasons of the 20th century sound rather muddy and opaque, because such a relatively large proportion of the acoustic power is concentrated in the fundamental and this acts to mask any other tones which are present at the same time. Moreover, the lack of precise attack fails to signal to the ear that a new pipe has just come onto speech. Opaqueness, by this definition, results in aural confusion between the various parts in the music being played and such diapasons are anathema to the contrapuntal clarity required for rendering Bach's works.
The route taken in Britain which ultimately led to this type of principal is interesting, and it began innocuously enough in the mid-19th century with the work of Edmund Schulze from Germany. Prior to that were the post-Restoration Schnitger-type principals of Father Smith in the 17th century as mentioned already, together with the classical (pre-Revolutionary) French inspired ones of Renatus Harris of the same period. Later these were followed by those of Snetzler, yet another builder who had learnt his trade on the Continent, and who developed a more essentially British style of instrument in the 18th century including such novel stops of the day as the Dulciana. Hill was building large instruments in Britain by the mid-19th century incorporating some German tonal concepts, though he was sometimes overwhelmed by the mechanical and tonal difficulties of filling large spaces with sound which were presenting novel problems to our somewhat inexperienced native builders.
But back to Schulze. He was familiar with Silbermann's work of the previous century but wanted to get more power from his principals without spoiling the tone through the use of excessive wind pressures. As an aside on the matter of tone quality, it is worth mentioning that he loathed Cavaillé-Coll's slotted Montres. Thus he encouraged his pipes to consume as much wind as they could by using large foot holes and large mouths, thereby generating more power. His methods were introduced into Britain in places such as Leeds parish church where they attracted much attention, in particular from T C Lewis who more or less copied them.
Simultaneously another line of attack was opened by builders such as Willis and, later, Harrison who believed that still higher powers could only be achieved through the use of higher pressures. They seemed either unaware of the progress already made by their contemporaries both at home and abroad or they ignored it, leading to the appearance of the foundational type of principal discussed above which had an entirely different type of tone. Not only loud, it also became unmusical and opaque for the reasons mentioned . Lewis abhorred this trend, declaring that "the evil has proceeded so far that I cannot refrain any longer from raising my protest against it" . But the trend continued unabated, culminating in the tonal innovations of Robert Hope-Jones to which we shall now turn.
In the closing decade of the 19th century and well into the 20th there existed an insatiable demand for nothing but power, more power and a dark profundity in British diapason tone. It ended with British diapasons whose harmonic structures were often little different to flutes but which were vastly louder, thus most of the acoustic power was emitted at the fundamental frequency alone. A vestigial retinue of a few harmonics trailed away, contributing little to the power nor brightening the tone quality. For the reasons discussed already, such diapasons became the ultimate in opaqueness as far as the music was concerned. As long as they produced sound which rolled impressively around a cathedral it did not seem to matter to many people, and for the simple homophony of hymns and Victorian musical pap maybe it didn't.
Robert Hope-Jones saw a way to make diapasons of this type by thickening their upper lips, simply by gluing leather over them rather than by using thicker metal which would have made the pipes difficult and expensive to manufacture. The leather also probably affected the acoustic impedance of the upper lip owing to its softness compared to the metal, as well as modifying the dynamics of the decaying air stream owing to its rougher surface. The leathered lips resulted in vaguely defined broad impulses being delivered to the air column in the pipe, such impulses containing fewer harmonics. Hope-Jones called this type of diapason a Diapason Phonon, a name probably invented by his brother Kenyon who had had the benefit of a classical education. It was copied by some others including that grossly over-rated builder, John Compton, whose regrettable legacy to the 20th century British organ is that his firm continued to propagate such tonal ideas until its merciful and inevitable collapse in the 1960's.
The diapason phonon was as far away as it was possible to get from a principal derived from Silbermann roots, and Lewis was quite right to draw attention to its shortcomings and those of its imitators. Because builders such as Willis, Norman & Beard and Harrison, as well as Hope-Jones and Compton, were all apparently vying with each other to subdue the populus with thick diapason tone, his was but a small voice crying in a wilderness of Victorian vulgarity. It took us the best part of a century to recover a sense of values, delayed by dilettantes such as George Dixon (Figure 3) who for some unfathomable reason had such a major yet malign influence on the tonal design of British organs. He dug an everlasting hole for himself by his contributions to the book referred to earlier . Anyone who regards the sound of a Schnitger or Silbermann organ as nothing more than "sausage frying", or who thinks it worthwhile to develop detailed registrational rules for use with Compton's extension organs, can be safely ignored. The pity is that he wasn't.
Figure 3. Hubris on horseback - Lt Col George Dixon
Maybe the one man who had the vision, tenacity and sheer guts to break through this barrier as far as British organs were concerned was the late Ralph Downes, and the tonal design of his landmark Royal Festival Hall instrument of c.1950 continues to evoke controversy and admiration to this day. Whatever its detractors might say, it proved beyond doubt that principal choruses designed along Schnitger and Silbermann lines have no difficulty at all in filling the largest spaces with transparent sound.
The word geigen is said by some British writers on the organ to mean a violin (for example, see good old Sumner ). In fact it does not mean any such thing; it is a German verb, not a noun, and it actually means to fiddle, as in playing a violin. We can exploit the richness of English to pun that geigen principals or diapasons were arrived at by fiddling around with the more conventional principals of the 19th century until they became thinner and stringier in tone, and that is more or less what happened. The pipes are certainly narrower, and in some ways the tone reverted to that of Silbermann's feistier principals a century earlier. The geigen therefore became less opaque than the general run of principals that were being made at that time, and so it has remained. Probably the only significant difference between today's geigen principals and those of Silbermann is that the pipes generally speak slower because of the use of constricted foot holes, higher wind pressures and modern voicing techniques.
Geigens are extremely valuable stops from a musical point of view, and even from the digital re-creation of a mid-20th century American example provided in this article, you can immediately hear how refreshingly transparent is its tone compared to the thick muddiness of the ordinary diapason and the phonon which precede it on the clip. Most modern geigen principals are probably over-regulated and rather too anodyne in attack; they do not quite have the character of a Silbermann principal for example. Nevertheless, it is invariably worth trying to build up your principal choruses on them if you encounter them in a modern British or north American organ, rather than on the ordinary diapasons. They are also excellent for use in an enclosed department where their tone is less stifled by the swell box than an ordinary principal would be.
The path followed in the evolution of principal tone since those of Schnitger in the 17th century shows a continual desire for ever greater power. Gottfried Silbermann responded to this over the following 50 years or so by developing a stronger and brighter tone which nevertheless remained eminently musical. Although some subsequent builders used his work as a basis for their own, most principals developed in the later 19th and 20th centuries became more opaque in tone, good for making loud noises in large spaces but not for much else. These trends were seen at their most extreme in Britain in the work of builders such as Hope-Jones and Compton, whose principals were little different to extremely loud flutes in terms of their harmonic structure in which most of the acoustic power was concentrated into the fundamental frequency. Most musicians and builders at the end of the 19th century and into the 20th saw nothing wrong with such stops, which were apparently admired and widely copied elsewhere. It is remarkable that it took so long for better taste and judgement to return, and inexplicable that even today the work of Compton remains so highly regarded in some quarters.
1. "The Tonal Structure of Organ Flute Stops", C E Pykett 2003, currently on this website (read)
2. "The Modern British Organ", Rev N Bonavia-Hunt, London 1948.
3. "The Organ", W L Sumner, 3rd edition, Macdonald, London 1962.
4. "The Organ", Peter Williams & Barbara Owen, Macmillan, London 1988, p. 272.
5. ibid, p. 281.
6. "How the Flue Pipe Speaks", C E Pykett 2001, currently on this website (read)
7. Quoted in "The Organ - its tonal structure and registration", C Clutton & G Dixon, London 1950, p.41.
8. "De la Détermination des Dimensions des Tuyaux par rapport à leur Intonation", A Cavaillé-Coll, Académie des Sciences, Paris, 23 January 1860.
9. "A Protest against the modern development of unmusical tone", Thomas C Lewis, London 1897.
10. Jean-Louis Coignet, consultant for the new Casavant organ in the Brick Presbyterian Church, New York City, reported that:
"the entaille de timbre has to be opened one diameter from the top of the pipe. Its width should be either 1/4 of the pipe diameter for most principals, 1/3 of the pipe diameter for strings and some principals, or 1/5 of the pipe diameter for flutes. It should be noted that the harmonic part of Flûtes harmoniques has to be cut dead length and without slotting". (The Diapason, December 2007).
These data are contained in documents from Cavaillé-Coll's workshop now in M Coignet's possession, showing that he most certainly did not slot his pipes merely to assist tuning them as most writers believe. On the contrary, they support my opinion that Cavaillé-Coll had developed an understanding of how the slots work to a remarkable degree for its day. I am indebted to Robert Sproule for drawing this to my attention.
11. An article elsewhere on this website discusses the issues of "transparency" and "opaqueness" in more detail by referring to the regulation characteristics of principals made by Silbermann, Cavaillé-Coll and British builders around 1900 (see Gottfried Silbermann's Fluework).