My Erat single action, however, had the shutter system working on small brass hinges, and a vacant slot where the pedal had been removed, so I decided to try to reinstate the pedal on this instrument.

The results of experimentation, both on my own harp and those with shutters that I was able to play was, to say the least, disheartening. A slight ‘bulging’ of the tone was the best that I could manage, and most of the time, no noticeable effect at all. So, was Bochsa correct when he called it ‘a pretended effect’?

This gave rise to my second question. What did I expect it to do? To be honest, I expected a dramatic Forte and Piano, with a vibrato, created by rapid pumping of the pedal, to rival a vibraphone. Well, I can tell you it does not do that! Failure to get any noticeable effect bought me back to my first question, and why the pedal survived in production for over 100 years if it does not work.

Consequently, I set about practical experimentation to find out what I could make it do, and very quickly discovered what I have been missing. Partially, this came from a meeting with harpist and single-action devotée Masumi Nagasawa, in Amsterdam. Amongst many subjects, we discussed the swell pedal and its application. It was the first time that I had had the chance to hear the pedal played in an otherwise quiet room, and to have someone listen to me play and give me ‘feedback’ on the results. Surprise number one is that the effect is much less audible at the instrument, and much more apparent in the developed sound.

We also discussed the application of Krumpholtz’s instructions for using the pedal. The principle of a swell pedal was very familiar in the 18th C, as it was used on harpsichords, and even some pianos as a means of creating dynamics in the search of ever more emotional expression. It was J B Krumpholtz, however, who suggested the application of the system to the harp. His suggestion was that the back stave of the instrument should slide, but the harp maker Naderman, in 1785 produced two harps with oval soundholes in a broader back stave, which can be opened by pressing the pedal. As if eight pedals were not enough, a ninth was added, which operated a cloth mute, or Sordune, which touched the strings at the soundboard and made the instrument play automatic étouffés.

Krumpholtz and Naderman both use the same system of symbols  to indicate the pedal usage, and we are fortunate that in the Krumpholtz/Plane méthode of about 1800, we are given a key to these symbols.

This system was ‘lifted’ directly by London harp teacher Mademoiselle Merelle for her Complete instructions for the pedal harp, which also contains a number of compositions in which these symbols appear. Unfortunately, whilst both Krumpholtz and Merelle tell us what the action of the symbol is, neither describes HOW to do it, so practical experimentation is the only way forward, and it is essential to listen to the results, not to try to force the sound into a preconceived response.

The first realisation I had, and which I am convinced is responsible for the statement that the swell does not work, is that, when it was introduced, the natural state of the harp was a closed back, with small soundholes in the soundboard. Compare the result of application of the pedal that suddenly opens the back of the instrument with the reaction of a player used to the sound of an open-backed instrument hearing the slightly damp squib of closing the shutters. I say ‘damp squib’, because the result is subtle. It is not a major drop or rise in volume, although I have to say that this is extremely variable, and some harps are more responsive than others. The major effect is a change in tone colour. The best analogy I can give is a singer changing from an open-mouthed vowel like an Ahhhhh to a more closed mouthed vowel like Ooooo. Given that the Oooo sound is the ‘natural state of the harp’, suddenly applying the pedal having struck a chord, or on the raised-stemmed notes of a chant de pousse, adds a warmth and intensity that the fingers alone cannot impart.

Then there is the issue of interpreting the symbols. Most are fairly self explanatory, but the V symbol to open and close the shutters immediately, and the zig-zag line for the undulation of the sound, are not so simple.

If one opens the shutters, plays the chord and then closes the shutter, it brings a more rapid decay, but if one plays the chord and then depresses the pedal and releases it immediately, very little happens. If, however, one rather more squeezes the pedal open and closed, the effect is far more noticeable. Of, course, it has to be done whilst the energy from the chord is driving theboard, but not so quickly that the tone does not have time to change.

Similarly, the zig-zag line, which Krumpholtz describes as ‘to make the sound undulate’ must be done slowly enough to allow the sound to develop. It should be pointed out here that the effect is not an actual vibrato. The pitch does not change at all, but again, the vowel sound changes subtly, and perhaps Merelle’s description ‘to make echos’ is more apt.

Bochsa, in his Nouvelle Méthode, says that the effect is horrible, and that if any other instrument were to make such a sound, it would be thought unpleasant. Bochsa, however was writing in 1814, twenty years after the introduction of the swell, and one has to consider how composition had change in that period, not to mention the difference in harp tone between a small French harp of the 1780’s and a harp post-introduction of the Grecian, the first commercial double action harp. Certainly, Bochsa is still describing a French harp in his Méthode, but he also says that harps are too lightly strung, and one should use the thickest strings possible without damaging the harp, and shows a crochet-action harp of the Empire type in his illustrations.

Tension is, however, one of the critical factors that affects the efficiency of the pedal. The effect can only be heard whilst the sound is at its most powerful, and the more rapid the decay, the less time there is for the sound to be modified. The stringing must be at a high enough tension to drive the board but not so high that it prevents the board moving and consequently acts as a damper. As many old harps were later strung with gauges closer to modern stringing, the boards are stressed to the point where the volume and tone is reduced, and the decay so rapid that the swell pedal has no time to make a tonal difference.

Only since research and experimentation have given us realistic tensions to apply to these harps have we heard a more truthful sound from these instruments, and heard what the shutters are capable of. It is my belief that the swell was introduced in an attempt to provide another tool to help the harpist create effects that expressed emotion, and to provide the harp with a means of affecting the tone once the note had been struck. It is therefore not surprising that a composer like Krumpholtz, who was pushing the envelope in terms of tonality and emotional expression, should espouse the system, whilst a composer like Bochsa, whose works are rather more ‘bravura’ in style, considered it redundant. Genlis, in her Méthode of 1802 says that the pedal is of great effect, but that petticoats somewhat reduce the effectiveness. I am happy to defer to her experience on his point, but certainly placing a blanket over my knees whilst playing does reduce the intensity of the effect.

It is interesting that, with the introduction of the Grecian in 1812, one of the most obvious structural changes visible is the reduction in width in the soundholes. My Erat single-action has a maximum hole width of 4 5/8” (11cm) compared with 3 1/2” (9cm) on my Erard Grecian. I propose that harpists had become so accustomed to the tone of the instrument with the shutters open that, post 1800 or so, that became the natural state of the harp, and so the soundhole size was reduced so that the open state was a more appropriate size for the Helmholtz resonance of the body, in order to balance the boost to the upper frequencies that larger holes would provide. For instruments built without shutters between about 1790 and 1820, the holes tend to be smaller, even being fitted with ‘baffles’ to reduce the opening further in relation to the body size.

                                                       Challiot harp 1820

The absence of much dedicated repertoire for the harp with swell is, I think, indicative of both the complexity of use and changing sensibilities, but for the harpist prepared to live with the vagaries of another buzz variable, another activity to co-ordinate whilst playing and the need for much further research, it offers a depth and tone colour that can only add to the beauty and understanding of the repertoire.

© Mike Parker 2008-02-21