The crochet harp showed a remarkable consideration in its engineering for adjustment.  Intonation of extant harps by Hochbrucker could likely only be adjusted in 2 ways: retuning the open string, or adjusting the length of the pedal rods (and thus the degree to which the crutches displaced the string.)  This was likely to have been sufficient for a short time, but typically, as the harp ages and the wood shifts, it's often necessary to adjust the mechanism to compensate and maintain proper intonation.


Some crochet harps (though not all) introduced a third way of adjusting intonation: the moveable string nut.  What on previous harps was a simple metal stud driven into place by a hammer, evolved into a more complex piece of hardware with the "moveable string nut" (pictured in figure 9).  This device was attached to the neck by a screw, allowing it to be moved up in down, thus increasing or decreasing the pitch of the semitone produced upon engaging the pedal.

The crochet mechanism had several benefits and several drawbacks.  The most prominent benefit to this system (in the case of crochet harps with adjustable string nuts) was the easy and simple regulation of the instrument.  Performers using crochet harps could easily adjust the intonation to their own taste by adjusting the position of the string nut, the position of the crochet on the spindle, and the length of the pedal rod.  However, the consistent action of the crochet against the strings made string breakage very common, and the strings generally had a slightly duller tone when in the “fretted” position than when open.  The lateral displacement of the string also pulled the fretted strings out of line with the open strings.  Pierre-Orphée Érard describes this issue in his 1821 publication The Harp in its Present and Improved State Compared with the Original Pedal Harp:


"A great defect in the construction of these harps is, that the action of the pedal, to give the string a second sound, draws it out of the vertical perpendicular, to make it rest upon the neck of the instrument … which lateral motion greatly increases the difficulty of the execution, by destroying the uniformity of the distances between the strings, and tends to put the string out of tune..."1


Unlike modern harps, the early crochet harp’s sound holes were cut into the soundboard, rather than in the back of the body, which was solid and staved.  In 1786, harp player Jean-Baptiste Krumpholtz presented his invention of a system of portes de renforcement (“swell doors” in English).  These were hinged doors  (see figure 10) cut into the body of the instrument which could be opened and closed by an 8th foot pedal, slightly altering the sound.  Several prominent harpists and composers advocated strongly for the effect produced by these doors, and their importance is reflected in their retention well into the mid and even the late 19th century in the case of some makers, however, they fell out of use in the early 20th century, for a number of reasons.


Mme. De Genlis praises the effect in her Nouvelle Méthode, but emphasizes that it was hardly noticeable when employed by female harpists, as their voluminous skirts muffled the sound, rendering the difference in volume negligeable (Genlis 1811, p. 12).  The mechanism operating these swell doors was also reliant on a leather hinge connecting the door to the body.   This hinge quickly wore out with use and age, rendering the doors completely disconnected from the body, and thus unable to operate normally.  The difficulty of repairing this mechanism, necessitating the complete removal of the doors, fabrication and application of a new leather hinge, and replacement of the doors, was such that many surviving harps originally appointed with swell doors have subsequently had the doors and the corresponding mechanism removed entirely.


While swell doors have not survived in modern harps, their presence left its mark in the form of today’s sound holes, now still carved in the back of the harp, facing the player (see figure 10).


The strings would have been secured inside the soundboard by small carved slotted pins (“boutons” in French), generally made from precious wood or ivory, which could be removed from holes in the center strip, allowing the string to be inserted, then replaced to secure the string.  In his 1772 Essai sur la vraie manière de jouer de la harpe, Jacques-Philippe Meyer confirms that the coloring system for the strings as still used today was already established by his time, writing that

“Par precaution on emploie des cordes rouges pour les ut, & des cordes bleues pour les fa."2


The wooden parts of the crochet harp, with the exception of the soundboard, were generally made from maple, although other luxury woods were occasionally used, such as satinwood, mahogany, and rosewood.  Soundboards were consistently made from spruce, and were composed of several successive planks of quarter sawn spruce, arranged with the grain running horizontally, glued together and tapered in both thickness and width into the typical elongated trapezoidal shape of the soundboard.  The back of the body was usually staved, although later harps featured rounded backs, and the neck was carved from a single, solid piece of wood.

By far the most widely used type of mechanism among 18th century French harp makers was the crochet mechanism.  The precise origins of the crochet mechanism are unknown.  Several theories exist, among them that Hochbrucker himself developed it later in his life, that it was the work of another German luthier, or indeed that it was invented by an early French maker.  

In her 1811 Nouvelle méthode pour apprendre à jouer de la harpe, Caroline-Stéphanie-Félicité comtesse de Genlis credits the invention of the pedal-operated harp itself, making no mention of Hochbrucker or of the crochet specifically, to a Mr. Gaiffres (also known as Goeppfert, Geoppfer, Köpfer, or Keipfer, among a myriad of other spellings).  Mme. De Genlis claims that Gaiffres was her first harp professor, though she also claims that his talents lay not in playing, but rather in building harps, lauding him as a gifted mechanic, but as a relatively unskilled harp player, having, along with his brother, invented the instrument approximately 60 years prior to her writing of the method (approx. 1751.)3  This date would, however, put said invention well after the definitively dated harps by Hochbrucker, making this unlikely. Furthermore, no instrument made by Gaiffres is known to have survived to the present day, further obscuring the early history of the instrument and that of the crochet mechanism.

In the same text, Genlis claims that early Parisian harp makers Salomon and Louvet were the earliest French harp makers, first copying, then enlarging and expanding upon the design of Gaiffres (Genlis 3).  Interistengly, the earliest material evidence of the crochet harp comes not from any of the three makers mentioned by Genlis, but rather from a harp preserved in the Musée de la Musique in Paris by Saunier, dated ca. 1760.

By the mid-to-late 18th century, the crochet harp had become the standard in France, and makers known to use this type of mechanism include Saunier, Louvet, Renault & Chatelain, Naderman, Holtzmann, Zimermann, Wolters, Clermont, Cousineau, and Camus, among others.  The crochet mechanism operated on the same basic design used on later Hochbrucker harps, but made several key changes.  The pedals worked much the same way, and the primary differences are seen in the part of the mechanism housed inside the neck.  The most important of these is of course the crochet from which this type of harp takes its name.

Much like on Hochbrucker's harps, the top of the pedal rod was flattened, with a small nodule on each side which fit into a slot in the boomerang-shaped pedal rod/chain coupling levers at the low end of the neck’s mechanism.  The arrangement of these levers was changed, however.  In Hochbrucker's design, these coupling levers were stacked one atop the other in a neat row.  Crochet instruments rearranged these levers in one of two ways.  A “linear” type, saw each lever positioned separately in a line.  A “stacked” type sees the levers placed one atop the other in two sets (one of 3 levers, the other of 4).  Figure 8 shows the lever arrangement of two crochet harps, a linear type (left) and a stacked type (right).

Individual innovation was quite rare among French harp makers of the 18th century.  One notable exception was George Cousineau.  Starting in the late 18th century, Cousineau began making harps with a new type of mechanism, now called the “béquille” (confusingly translating as “crutch” in English, it will be exclusively referred to as béquille in this paper to distinguish it from the Hochbrucker crutch mechanism).  On béquille harps, rather than crochets, each string was equipped with two béquilles placed on either side, which, upon depressing the pedal, would turn to engage the string at two separate points (see figure 11).

The above figure shows a crochet in both the neutral (left) and engaged (right) positions.  Rather than the L-shaped crutches of Hochbrucker, the crochet harp featured threaded “crochet spindles,” exiting the neck horizontally through drilled holes adjacent to each string.  Upon each of these spindles was a threaded brass crochet (“hook” in English).  Upon depression of the pedal, the crochet would retract into the neck, pulling the string towards the neck and against the "sharpening nut" and raising the pitch by a semitone.

Crochet harps featured an updated version of the pedal spring.  As opposed to the relatively rudimentary springs seen on Hochbrucker's instruments, crochet harps included finely constructed flat coil springs, housed still in the knee-block as shown below in figure 7.

While certainly an ingenious system, Hochbrucker’s mechanism had little capacity for adjustment, an issue which would soon be addressed by later harp makers.  By the mid-18th century, harpists playing harps like Hochbrucker’s soon began to make appearances at Paris’ Concert Spirituel, sparking a fanatical interest in the instrument in France.4 This new popularity would give rise to the great French harp makers of the 18th century, many among them of German origin, such as makers Naderman, Holtzman, Zimmermann, and Wolters.  Harps of this period were generally made with one of two types of mechanism: the crochet or the béquille.  All of these are known today as “single action” harps, irrespective of their specific type, as their mechanisms allowed only a single alteration in pitch, as opposed to the two alterations possible on modern, or “double action” harps.5

French harps of the late 18th century share two defining characteristics: a staved body, and a highly decorated "scroll" at the point where the column meets the neck.  A typical French harp is shown below in figure 5.

The béquille system solved the problem of dull tone in the fretted position, as well as that of the string alignment.  The béquille system is, however, by no means the only innovation of this curious inventor.  In 1782, Cousineau is credited with building the first double action harp.  This instrument, today in a private collection, and something of a novelty, featured two sets of béquille action and two rows of 7 pedals each at the base, for a total of 14 pedals.  It is easy for the modern harpist to understand how, from a practical standpoint, this harp could be incredibly complicated to play.  Indeed, at the time, when music written for the harp was generally designed around the instrument, why make such a complex instrument for which there was little use in existing repertoire?  The answer may lie with temperament.


While today’s harpist might be quick to imagine that the double action system emerged simply so as to allow for further enharmonic substitutions, this is far from the only reason.  The limited chromatic ability of the single action harp meant that, while by the modern standard of equal temperament it would be capable of playing each tone of the chromatic scale, the 18th century advocate for the meantone temperament would find this system lacking.  In the meantone system, there are two separate types of semitone, a diatonic and a chromatic.  However, the harp, like the keyboard, by necessity of its design, was unable to distinguish between the two.  In his 1782 Memoire sur la nouvelle harpe de M. Cousineau, luthier de la reine, Pierre-Joseph Roussier praises Cousineau’s 14-pedal harp, while issuing a scathing indictment of the single-action system—




“La méchanique qu’on y a ajoutée depuis quelques années, a réellement pour objet de fournir à la harpe toutes les modulations en usage.  Mais, pour n’avoir pas consulté les principes de la Musique, l’Auteur de cette invention n’a pu remplir que très-imparfaitement cet objet, en divisant chaque ton, au moyen de pédales, par un seul son intermédiaire qu’il a cru pouvoir server indistinctement & de dièse à la note inférieure & de bemol à la supérieure.  C’est-là exactement le systême absurd sur lequel sont établis nos divers instruments à touches, restes des siècles de barbarie.  Mais comme selon les principes de la Musique, les intervalles formés par le dièse & par le bemol loin d’être le même, se distinguent au contraire en deux sortes de demi-tons, l’un appellé majeur, l’autre mineur.  Le Sieur Cousineau s’est propose d’introduire sur la Harpe ces deux sortes de demi-tons, & de corriger ainsi le vice des demi-tons factices que l’ignorance des principes y avoit fait transporter, à l’imitation des Instruments à touches.”6




While the mid-to-late 18th century would see the general triumph of 12 tone equal temperament, we may well have the meantone system to thank for our modern double action harp.

While Cousineau’s double action harp was never sold on a large scale, and quickly abandoned by the maker himself, likely due to the practical complication of playing such an instrument, it provides fascinating insight into the motivations of late 18th century harp makers.


Cousineau, ever creative, is credited with a third mechanical system for harps later in his life, in the early 19th century, this one a single-action mechanism.  On this type of harp, called the harpe à chévilles tournantes (harp with rotating tuning pins), the string was tied directly to the tuning pin, which, upon depressing the pedal, would rotate to tighten the string.  Any harpist can understand why this system might not be ideal, as regular tightening and loosening of a string in this manner, rather than by more traditional “fretting” techniques can cause unwanted and rapid stretching or contraction of the string itself, making it very difficult for the instrument to remain in tune.

From the earliest single action harps made ca. 1720 until the last, made between 1830 and 1840, the number of strings steadily increased.  Hochbrucker’s 1728 harp housed in the Musée de la Musique has only 34, while the latest known crochet harps, made in the early 19th century, typically had between 39 and 42 strings.  Indeed, in span of only a few years method books tell us that the number of strings increased from only 32, according to Philippe-Jacques Meyer’s 1772 Éssai la vraie manière de jouer de la harpe7 to 42 according to Guillaume-Pierre-Antoine Gatayes’ method, ca. 1790.8

The question of stringing on single action harps with regard to gauges is a hotly debated one among scholars and historical performers alike.  Indeed, little verifiable evidence remains of 18th century stringing practices.  Few string gauges exist from the time and stringing is rarely discussed in method books.  Fortunately, however, we are generally aware of the materials used.  Unlike its sisters of the keyboard family, pedal harps were strung with gut strings in the higher range, while the lower few strings (the number of which is determined by the range of the individual instrument) were strung with silk-core strings overspun with a metal, often silver, much like those used on some modern acoustic guitars.  The gut strings were not treated with lacquer in the manner of modern strings, and as such had to be regularly oiled.

One significant barrier to the scholar hoping to answer the question of 18th century stringing is the complete lack of standardization.  Method books of the late 18th and early 19th centuries are arguably the most enlightening source of information on the subject.  These methods paint a picture of a world of heated disagreement, both between harp makers and harpists themselves, on the proper stringing of the instrument.


In his Méthode, thought to have been published ca. 1810, harp builder Jacques-Georges Cousineau advocates for a strict adherence to the ton de l’Opéra de Paris, praising this diapason as the lowest of all orchestras and thus the most appropriate for the harp, and warns against tuning to a higher pitch.9  Mme. de Genlis, similarly advocates for tuning to the ton de l’opéra in her 1811 method, while noting that she felt the “standard” gauge of strings of the time to be too light, stating that she prefers that the strings be thicker, stating the following:


“Je voudrois que les cordes fussent un peu plus grosses au même ton et par consequent plus tendues.  C’est ainsi que je fais jouer mes élèves.  Il faut beaucoup plus de force dans les doigts, l’exécution est plus difficile, mais le jeus plus net, le son plus beau.  Il en résulte que les cordes tirant advantage sur la Harpe il faut des tables d’une grande solidité, sans quoi ells ne résisteroient pas à cette forte tension.”10


It is telling that later in the same text, Mme. de Genlis also describes Cousineau’s harps thusly—


"Les Harpes de Mr. Cousineau ont en général un beau son, mais ses tables sont faibles et ses mécaniques d'une extrême fragilité."11 (Genlis 1811, p. 11)


 Guillaume-Pierre-Antoine Gatayes echoes de Genlis’ sentiments in his method book, claiming that “all harps are strung with strings that are too thin; it is true that this fault is that of the instrument, as the soundboard is too weak to resist the strong tension of thicker string; one must therefore, when one can, choose only harps that are very solid…” (Gatayes 1790).  Robert-Nicholas-Charles Bochsa similarly encourages the use of heavier-than-standard strings in his 1820 Petite Méthode pour la harpe—

“Il est indispensable pour obtenir un beau son de la Harpe de la monter généralement en cordes un peu fortes.  Par là on évite le son criard et frisé qu’ont toutes les Harpes montées en cordes trop fines.” (Genlis 1811, p. 12)12

Given the clear lack of true standardization of stringing during this period, it is hardly surprising that modern performers using period instruments (or indeed reconstructions thereof) rely on a combination of information gained from period string gauges, advice from these method books, and personal experimentation, experience, and tailoring to the specific needs of each instrument.  This being said, today's performers, harp restorers, and scholars have generally agreed on a standard of stringing for most 18th and early 19th century harps, with crochet harps usually strung about two full octaves lighter than modern standard strings, such that, for example, a modern C5 string would be used on a French harp as a C3.

2. Cousineau and his innovations: the béquille harp and the harpe à chévilles tournantes

Figure 11-A late 18th century béquille harp by Cousineau, preserved in the Musée de la Musique in Paris, photograph by Jean-Marc Anglès

Figure 8-Linear (left) and stacked (right) levers on two French crochet harps, both ca. 1790, collection of the author

Figure 7-Springs on a ca. 1790 harp by Victor Clermont, collection of the author.

Figure 9-A moveable string nut on a French harp by Victor Clermont, ca. 1790, collection of the author.

 Figure 6-Diagram of a crochet in the neutral (left) and engaged (right) positions.

Figure 5-A French crochet harp by Victor Clermont, ca. 1790, collection of the author.


Figure 10-Swell doors on a French harp by Challiot, ca. 1805, collection of the author.

1. The Crochet Harp

Stringing Practices

The "French Harp"