The Puzzle

The curious disposition of the original double-manual harpsichord has been the subject of considerable speculation since 1739, but especially so over the past fifty years. The problem, simply stated, runs as follows.

During the first half of the 17th century, the Ruckers family made two basic kinds of harpsichord. They made a standard single, 1x8’, 1x4’, C/E-c3, and they made a double-manual instrument, 1x8’, 1x4’, upper manual like the single’s, lower manual GG/BB-c3, but with the lower manual shifted to the left by a fourth, so that its apparent range was C/E-f3.

No contemporary explanation is known as to why the Ruckers made the lower manual differently, but there exists a letter showing that the family, in 1637, refused to make any other kind of double. The things that make the lower manual different from the upper (and hence from the single manual harpsichord) are:

a. The lower manual sounds a fourth lower than the upper.

b. The lower manual’s being at a different pitch requires the addition of extra strings at Eb-G#.

c. The lower manual plucks much further toward the middle of the string.

d. The lower manual has its buff stop divided exactly as the single does, so that the buff stop on the upper manual is shifted a fourth down.

So the problem can be expressed as “Given the differences between the two manuals on the transposing harpsichord, what was the lower manual intended to be used for?”

Short octave arrangement of upper keyboard is cranked
to left to accomodate transposed lower keyboard.

Four Proposed Answers

A number of answers to this puzzle have been proposed. It has been variously supposed to have been intended for:

(a) routine transpositions down a fourth or fifth to accompany other instruments

(b) the transposition from Chorton to Kammerton

(c) to provide less experienced or assistant choir masters a convenient way to accompany singers reading from chiavette

(d) to provide contrasting sounds between the two keyboards

While it is true, as Franz Felix suggested to me, that these instruments were “...intended to provide multiple practical solutions to working musicians”, I don’t believe any of these explanations can have been the primary reason for the production of these instruments.

For example, many writers (including Raymond Russell) have suggested that the lower manual was added to help players in the routine transpositions down a fourth and a fifth, which they commonly had to do. This would be a most satisfactory solution if the lower manual could be imagined as accompanying, say, an alto recorder playing music written for the descant. The problem is that the difference in pitch is wrong; the lower manual is a fourth below the upper. An alto recorder is a fifth below a descant.

A similar problem precludes the Chorton/Kammerton difference. The Kammerton (chamber pitch) is used to describe the pitch used for domestic music-making. The Chorton (choir pitch) was that of church organs, and hence church choral music. This distinction is attractive as well, but again the difference is different. Chorton is between a major second and a major third higher than Kammerton , and hence could not account for the difference of a fourth.

The interval of a fourth between the reference clef and chiavette , however, does appear to match the interval between the keyboards. Vocal music in Italy was for a time written in a “little key”, transposed up so as to allow the tessitura to fit better within the lines of the staff. While any competent organist could have performed the required transpositions, Grant O’Brien argues in his great book Ruckers that the lower keyboard was made a fourth lower because for “rehearsals and training, it was necessary to use a harpsichord to avoid the expense of engaging blowers to operate the bellows of the organ. A student or assistant taking these rehearsals would have found the Ruckers double-manual harpsichord...indispensable.”

To see the problems here, we must first imagine the assistant choir director rehearsing the choir on Wednesday night in Amsterdam. The young director is seated at the instrument, and playing the various choral parts from music written in chiavette , a fourth up from reference.

The piece appears in F. If he were to transpose it down a fourth, and play it in C (on the upper manual), it would sound in the original pitch, and there would be no problem. But not being competent to perform the transposition, he plays the transposer’s lower manual, so that the sound that results is at the same pitch that would be played by the upper, if the piece were written in regular clef. This explanation might make sense had the Ruckers not put in extra strings for Eb-G#.Though the pitch is in general correct, when the assistant tries to play the G# key, expecting a G#, he instead gets Ab.The chiavette explanation renders inexplicable the fact that the Ruckers spent so much effort providing the extra Eb-G# strings, which are only of use if the music being played on each keyboard sounds in a difference key. If the “little key” transposition on the lower keyboard makes the two keyboards sound in the same key, the extra strings don’t make sense. They are a hindrance, not a requirement.

Secondly, we must confess the general implausibility of the scene described above; If your assistant director couldn’t transpose down a fourth, you’d be more likely to beat him with a stick than to buy him a $20,000 harpsichord. These instruments were the top of the Ruckers’ line - they cost between twenty and thirty pounds (with valuable decoration). It is inconceivable that these instruments were made for church rehearsal halls instead of the halls of royalty, the court, and the better homes of the bourgeoisie. Indeed, in van Dijck’s exhaustive catalog of harpsichords in old Netherlands paintings in Het klavecimbel in de Nederlandse kunst tot 1800 , while there are about a dozen two-manual harpsichords pictured, half of them transposers, most are in domestic scenes, and none are placed in any ecclesiastical situation - certainly not in choir rehearsal rooms.

While rehearsing a choir from chiavette may have been an occasional use of such an expensive instrument, the idea that this was its primary or ordinary function is simply not credible.

The fourth proposal, as set forth by R. T. Shann, is that the differences in the lower manual are not for transposing at all, but intended to provide additional contrasting sounds. He asserts that if the Ruckers had only wanted to make transposing harpsichords, then all they had to do was couple the keyboards together, as they did in the keyboards of the Mother and Child virginals. While I find this assertion wholly convincing, it does not really explain the real use for which the lower keyboard is intended. For an argument to be persuasive, it should account for all the differences listed above - pitch, extra strings, pluck position, and the unusual buffing arrangement.

A New Solution

David Ledbetter’s fascinating book “Harpsichord and Lute Music in 17th-Century France” helps explain an important aspect of the musical environment in which the transposing harpsichords were built. Lute music was at its greatest ascendancy during the first half of the seventeenth century. The gradual rise of the harpsichord as the favorite secular instrument took place during this period, but the established standard was the lute. It was in this period that the Ruckers made the transposing doubles (and refused to make aligned doubles, even when asked by Charles I).

The tessitura of the lute is about a fourth lower than that of the harpsichord. This is illustrated in the examples Ledbetter provides of musicians rewriting lute music for the harpsichord. Typically they transpose up a fourth , often from a minor to d minor. The fact that such rewrites were done clearly indicates a desire to play lute music on the harpsichord; I believe it also implies a desire to “play like a lute” on the harpsichord, to make a harpsichord sound like a lute. (One obvious provision for playing like a lute during this period was the lautenwerk, a gut-strung, shorter version of the harpsichord, which first came into existence during this same period. It however couldn’t be made to sound like a harpsichord , and nothing much came of it.)

So the decision to buy a harpsichord in Europe, in the early 1600’s, was somewhat complex. The buyer wanted to play the literature for the harpsichord, but everyone also loved the sound of the lute (which was frequently said then as now to be one of the most ravishing in all music).

Making a Harpsichord Sound Like a Lute

Now if I wanted to make a harpsichord sound like a lute , I’d play it in the lower tessitura, I’d buff it, and I’d pluck it nearer the center of the string. If we examine the differences between the Ruckers upper keyboard (the standard single) and the lower keyboard, we find that four accommodations have been made:

It plays a fourth lower. This would have been to play harpsichord music in the lower tessitura, so as to make it sound more like a lute playing lute music. (Philip Tyre has pointed out that the lower must have in fact been the more important manual, since the tuning pin labels on the Transposer are given in reference to the lower manual, not the upper.)

It has the extra Eb-G# strings. These were required, in meantone tuning, to keep the player from having to retune when moving to the lower keyboard and playing music that sounded in a different pitch.

The plucking point is much further toward the center of the string. While the lute, like the guitar, is occasionally plucked close to the nut, the normal playing position is proportionally much closer to the sound-hole. A quick survey of baroque paintings of lutenists, and the observance of modern performance practice will make this abundantly clear. The pluck point for the Transposer lower manual is dramatically further toward the center of the string, and it makes a very different, lutey, kind of sound.

The standard buff arrangement for a single manual harpsichord has been moved to the lower manual. A harpsichord can be made to sound more lutey by the simple expedient of pulling on the buff, which is often called the “lute stop” in spite of the other meaning of this expression. The buff hastens the decay of the sound and reduces some of the overtones. In the Transposer, the standard single arrangement, which is split at f-f#, has been moved so that the lower manual is split at this point, the upper necessarily now split at c-c#. This implies that the lower manual was the one more likely to be buffed. Significantly, the Ruckers never buffed a four-foot; in lutes, the unison strings are of gut, the octave of brass, so buffing a 4’ would have defeated the purpose of “sounding like a lute”.

All four differences between the two manuals are thus explained by this conceit: the Ruckers made the transposing doubles for secular use, for a market which valued the sound of the lute, and which they were able to supply by a modification of their basic single-manual harpsichord. This quickly became a popular form of harpsichord, and probably did “provide multiple practical solutions to working musicians”. The Ruckers family continued building them until the second half of the century, when the popularity of the lute began to decline.

Reid Byers

Princeton, NJ, 1997

Originally published in Continuo Magazine, April, 1997.