Posted 03 September 2003

The Wicki System—an 1896 Precursor
of the Hayden System

Robert Gaskins

Introduction

The concertina keyboard system known today as the "Hayden" system, which was independently discovered by Brian Hayden and patented by him in 1986, had also been discovered and patented 90 years earlier by a Swiss inventor named Kaspar Wicki. (This fact was uncovered by Göran Rahm.) 1

Wicki's patent is extremely clear, and his system has been described and illustrated by Maria Dunkel in both editions of her well-known book Bandonion und Konzertina, 2  but for some reason the Wicki system does not seem to have been generally noticed.

The patent is Swiss patent Nr. 13329, 30 October 1896, for "Tastatur für Musikinstrumente" (Keyboard for Musical Instruments), issued to Kaspar Wicki, "in Neumatt (Münster, Luzern, Schweiz)", one of twelve patents issued to Wicki between 1895 and 1917. 3  The entire text of the patent consists of three short paragraphs, with a single illustration which is completely unambiguous since it is labeled in standard musical notation:

The Wicki patent reads:

Diese Tastatur eignet sich hauptsächlich für Handzuginstrumente, wie Harmonikas, Konzertinas, Blasakkordeons und alle klavierartigen Instrumente.

Die Tasten, resp. Knöpfe in zwei Farben zur Unterscheidung der Normaltonleiter (c) sind in einer Weise geordnet, daß man alle Tonarten mit gleichem Fingersatze spielen kann. Die Töne stehen in den einzelnen Tastenreihen von links nach rechts immer um einen Ganzton von einander ab. Die zweite Reihe bildet zur ersten die zugehörigen Quinten, resp. Quarten und die dritte Reihe zur ersten die Oktaven, die vierte zur zweiten wieder Oktaven u. s. w. Die auf der Zeichnung durch stark ausgezogene Kreise dargestellten Tasten bezeichnen die C-dur Tonleiter, von hier nach links folgen die Noten, bezw. Töne mit b-Vorzeichen, nach rechts die Noten, bezw. Töne mit #-Vorzeichen, so daß man mit Zuzug der bezüglichen Töne aus C-dur auf der linken Seite die Tonleitern mit b und auf der rechten Seite die Tonleitern mit # hat.

Patent-Anspruch:
Eine Tastatur für Musikinstrumente, im wesentlichen, und abgesehen von der Anzahl der Tastenreihen, wie beschrieben. 4

The diagram was redrawn by Maria Dunkel in Bandonion und Konzertina, accurately, although Dunkel extended the pattern with three more keys on the right side (shown with dashed outlines) and re-labeled all the keys with letters:

(To read this diagram, one must know that her German notation uses "H" for the English note "B", and uses "B" for the English note "Bb". The suffix "es" means "flat", and the suffix "is" means "sharp". The apostrophes following the note names indicate the octave to which the note belongs.)

Dunkel provides some background on Wicki, in a section she entitled “The Block Concept”:

Das Blockkonzept

In der zweiten Hälfte des 19.Jahrhunderts werden unter den Pianisten einige Versuche bekannt, die Piano-Klaviatur zu modifizieren entsprechend der musikalischen Forderung nach Gleichbewertung aller 12 Töne einer Octav. Ein Modell, das das Privileg der C-Dur-Tonleiter bricht, stammt von H. J. Vincent. Anläßlich der Diskussion von Vincents Neuklaviatur bildet sich 1876 der „Chroma-Verein des gleichstufigen Tonsystems”. Durch diesbezügliche Artikel in der von A. Hahn ab 1876 herausgegebenen Zeitschrift Die Tonkunst angeregt, entwickelt Paul von Jankó 1882 die später nach ihm benannte Klaviatur, deren Charakteristikum ihr Aufbau in Ganztönen ist. Für die 12 Töne der Octav sind zwei Reihen mit je sechs Gantönen im Abstand einer kleinen Sekund nötig, also die Reihe c d e fis gis ais und die dahinterliegende, auf Lücke versetzte Reihe cis dis f g a h. 1896 erhält Kaspar Wicki, Münster/Schweiz, das Patent No. 99325 für seine „Neuartige Tastatur” (siehe Tabelle 20), die eine indirekte Anwendung des Jankó-Systems auf das Bandonion darstellt—Erwähnung findet letztere im Text des Patents allerdings nicht. Auch stellt Wicki die beiden Ganztonreihen nicht wie Jankó im Abstand der kleinen Sekund übereinander, und seiner Tastatur liegt interessanterweise nicht die temperierte Skala zugrunde. Für den Tonraum einer Octav veranschlagt er insgesamt 18 Töne (12 + 6), die er symmetrisch angelegten Tasten in der Gruppierung 3 + 6 + 6 + 3 auf vier Ebenen zuordnet. Es ergibt sich eine dreiteilige Tastatur: Die zentrale Region der natürlichen Töne ist umschlossen von den Akzidentienblöcken des Kreuz-Tonbereichs und des Be-Tonbereichs. Die Akzidentien könnten auch als doppelte Töne angelegt werden, „wenn nämlich die 12stufige chromatische Tonleiter der gleichschwebenden Temperatur zu Grunde gelegt wird. Es ist natürlich leicht, sich die Vortheile des Vorhandenseins dieser überzähligen Tasten und Töne zu Nutzen zu machen und diese letzteren als Bestandtheile der enharmonischen Tonleiter … einzuschalten.” In Wickis Disposition ist „jede Tonleiter jedes einzelnen Tongeschlechtes genau mit dem gleichen Fingersatz spielbar, ebenso leicht können gleichartige Akkorde in jeder Tonart gegriffen werden”. Ob die Idee der Ganztonanlage im Bandonionbau realisiert ist, muß zukünftige Forschung ermitteln. Wichtig bleibt hier festzuhalten, daß Wicki von der Tonschrittanlage ausgeht und für die Verteilung der 6 Ganztöne das Blocksystem vorsieht. Natürlich ist ihm dabei nicht entgangen, welche Intervalle sich in den Tastenketten ergeben, nämlich in der einen Diagonale Quinten, in der anderen Diagonale Quarten. So besitzt jeder einzelne Ton in räumlicher Entsprechung nicht nur seine Unterquint und seine Unterquart, sondern auch diese Intervalle in steigender Richtung. 5

The idea of repeating "blocks" of keys comes up repeatedly in dealing with an octave divided into twelve semi-tones. One familiar example is the Early Wheatstone Double duet system, based on arranging the buttons in three rows of four buttons; each button was one semi-tone higher than the button to its left, successive rows were not offset, and the nearest buttons in adjacent rows produced notes differing by two tones (four semi-tones) because the pattern was four columns wide.

         Ab     A      Bb      B
         E      F      F#      G      
         C      C#     D       Eb       

This pattern, on the Wheatstone Double duet, was repeated vertically, each octave occupying precisely three rows. The Wheatstone tutor for this instrument stresses how the same fingering pattern could be repeated from different starting positions to transpose keys (an advantage limited by the fact that the pattern was not also extended to left and right with duplicated keys).

Another system based on repeating blocks of keys was the Late Wheatstone Chidley duet system which was based on a block of 12 notes arranged in two rows of six notes; each button had a variable interval from the note on the button to its left, successive rows were not offset, and the nearest buttons in adjacent rows produced notes with no fixed relationship.

         G#     G      A       Eb     B     Bb
         C#     C      E       D      F     F#      

As with the Wheatstone Double system, the Wheatstone Chidley system was repeated vertically, each octave occupying precisely two rows of six buttons each. It did not offer the advantage of transposing with constant fingering.

(Of course, just because a keyboard is geometrically neat doesn't make it a great musical success; Wheatstone's Double system was spectacularly unsuccessful, despite excellent instruments and a lavish tutor, and the Wheatstone Chidley system was similarly unsuccessful in offering an option which would be more attractive than the Maccann system.)

The Wheatstone Double system had adjacent notes along each row with an interval of one semitone, and the Chidley system has no fixed interval between adjacent notes, but the Wicki system has adjacent notes in each row with an interval of a whole note, two semitones. This pattern of whole-note intervals along the rows is what connects the idea to the earlier piano keyboard ideas of Paul von Jankó.

The Jankó piano referred to by Dunkel was once thought of by many as the harbinger of a revolution in piano-like keyboard instruments, which would displace the conventional piano arrangement. It was based on the observation that an octave of 12 semi-tones could be arranged in two rows of six buttons; each button was two semi-tones (one whole tone) higher than the button to its left, successive rows were offset by a full button-width, and the nearest buttons in adjacent rows produced notes differing by a semi-tone.

         C#     Eb     F       G       A       B
      C      D      E      F#      Ab     Bb

Both rows together, of course, are no different from the usual order of the keys on a piano. Paul von Jankó made new keys for a standard piano. Each of the 88 keys--black or white, as usual--was cut in stair-steps so that there were three flat sections on which the key could be struck, and alternate keys were offset up and down. This produced Jankó piano keyboards which actually had one row of 88 keys cut in complex shapes, but which appeared to have six rows with 44 keys (and 44 spaces) in each row—which appeared to have 264 keys total—in the same horizontal space of a conventional piano keyboard. That was the 1883 patent, and it resulted in a keyboard which could be played with equal facility and with the same fingering in every key signature. Apparently experience with this immediately suggested that, since each row of keys contained 44 "striking points" and 44 spaces, the keys could be cut narrower with only the striking points left full width and projecting over the adjacent spaces. This improvement, in the 1885 patent, further made the keyboard more compact than a normal piano keyboard, so that longer intervals could be reached. The combination of advantages seemed decisive.

"If I were to begin my career anew it would be on this keyboard."
— Artur Rubinstein

Wicki's bandoneon keyboard, soon after Jankó, was based upon exactly the same two rows of six buttons as the Jankó piano, but with the rows realigned with respect to each other; each button was two semi-tones (one whole tone) higher than the button to its left, and successive rows were offset by half a button, but the nearest buttons in adjacent rows produced notes differing by 5 or 7 semi-tones (a fourth or a fifth).

         C#     Eb     F       G       A       B
                           C       D       E       F#      Ab      Bb   

or (choosing another section of the extended rows)

                       F       G       A       B       C#      Eb       
                           C       D       E       F#      Ab      Bb

This pattern was extended by repetition both right and left, and the pairs of rows were repeated vertically. But since Wicki was not constrained by the piano structure, he could make his vertical repetitions octaves (where Jankó was constrained to have unison notes).

The general pattern is very simple. Two full rows of 8 or 9 keys are filled out

         C#      Eb      F      G       A       B      C#      Eb       F  
    F#       Ab      Bb     C       D       E      F#      Ab      Bb 

(here supplying an extra "F" at the right end for symmetry, as Dunkel does), and then these two rows are repeated vertically (see the full diagrams).

In Wicki's keyboard, six of the notes in each octave were repeated, to left and right, in adjacent rows. So where the Chidley system put each octave into 12 notes on two rows of six, Wicki for each octave used 18 notes arranged in two rows of six plus two more rows of three. As Dunkel notes, for a 12-tone equal-tempered scale these duplicated notes were simply different locations for the same notes. But they were important, because it is these repeated notes which make it possible to play in multiple key-signatures using the same fingering.

    F#       Ab      Bb     C       D       E      F#      Ab      Bb 
         C#      Eb      F      G       A       B      C#      Eb       F  
    F#       Ab      Bb     C       D       E      F#      Ab      Bb 
         C#      Eb      F      G       A       B      C#      Eb       F  

Here, F#, Ab, and Bb are duplicated, and C#, Eb, and F are duplicated. These 18 notes, on four adjacent levels (3 + 6 + 6 + 3) are all in the same octave, the octave from the "C" shown up to the next "C" above it (two rows higher). In the Wicki keyboard, it is this full set of 18 notes which are repeated for each octave.

For convenience in exposition, we have been using a single set of names for the twelve notes of the equal-tempered scale. For better comparison with the Hayden patent, though, observe that other—equivalent—names can be used for the same notes:

         Db      Eb      F      G       A       B      C#      D#       E#  
    Gb       Ab      Bb     C       D       E      F#      G#      A# 

This re-spelling of the note names doesn't change the notes at all. "C#" and "Db" are the same, "F#" and "Gb", "Eb" and "D#", "Ab" and "G#", "Bb" and "A#", and even the unfamiliar use of "E#" for the same note as "F". When the notes are labelled in this way, we can see the three vertical sections of the keyboard that Wicki indicated on his patent drawing by making the circles of the central section bolder—all the flats are on the left, all the natural notes of the C major scale in the middle, and all the sharps on the right. Each row of nine keys still contains only six different notes plus three repeated notes (but repeated from a different octave).

Dunkel sums up the advantage of Wicki's keyboard, in words which sound familiar from their later application to the Hayden keyboard: "In Wicki's arrangement the scale in every key signature can be played with exactly the same fingering, as can the equivalent chords in every key."

Turning now to the Hayden system, it is necessary to consult only the Hayden patent. The title of the patent is "An Arrangement of Notes on Musical Instruments". The abstract says "The present invention places notes on musical instruments along several adjacent paths [rows]; so that along the paths, notes at intervals of whole tones are close together, and between the paths, notes at intervals of fourths and fifths are close together … Keyboards for Organs, Accordions, and in particular Concertinas are described in greater detail. Several different forms of multi-stringed musical instruments are described; and a few details of other musical instruments that may use the invention are given." So the invention is purely the arrangement of notes, independent of any particular instrument such as a concertina (as was true of Wicki also).

Hayden's patent does mention the Jankó piano keyboard patent ("Jankó 1885 Pat No. 536"), as an example of a "neutral chromatic keyboard". It observes that "The disadvantages of these keyboards are that they are difficult to begin to learn to play, having complex fingering patterns, and are also difficult to learn to read music with. Only the Jankó Keyboard has these disadvantages to a very slight extent, but on the other hand it has most of the large linear width difficulties associated with the Standard [piano] Keyboard."

The Hayden patent mentions that in the Hayden system "…notes sounded at consecutive Touch Points [buttons] on the same Musical Path [row of buttons] have a musical interval of a whole tone between them. (No claim is made that this in itself is new. This is used on the Jankó 1885 Keyboard …)." It goes on immediately to clarify then what is novel about the Hayden invention: "The Musical Paths [rows of buttons] are arranged relatively to each other in the present invention in such a way that the Note sounded at one Touch Point [button] on one Musical Path [row] and the Note sounded at the Nearest Touch Point on an adjacent Musical Path are a Musical Interval of a Fourth or a Fifth apart; or if two Touch Points on the same Musical Path are equidistant from a Touch Point on an adjacent Musical Path, then one of these pairs across the paths will sound a Musical Fourth apart and the other pair across the paths will sound a Musical Fifth apart …." Of course, this particular relationship between two adjacent rows of notes at whole-tone intervals, where each note is diagonally adjacent to its lower fourth and fifth and upper fourth and fifth, is also precisely the claim of Wicki's patent.

Hayden's patent does not mention Wicki's patent (as Wicki's patent does not mention Jankó's patent). By a curious coincidence, the research undertaken for Hayden's patent missed finding two patents from the same year, 1896: the patent for the Crane Duet was not found and Brian Hayden was for a long time under the impression that it had never been patented, and apparently this patent for Wicki's bandoneon keyboard was missed as well.

A close comparison of the Wicki keyboard 1896 (taken from the patent) and of the Hayden keyboard 1986 (taken from its patent), will demonstrate that the arrangements are essentially identical. Since the Wicki keyboard was drawn for an instrument with 51 buttons on one side (or 54 buttons in Dunkel's redrawing), the corresponding Hayden keyboard example is Fig. 12 from his patent, described there as "an arrangement of notes suitable for the treble keyboard of an Accordion" with 60 buttons, nearly the same size. The Hayden patent goes on to say "Several of the notes appear on one side of the Keyboard as "sharps", and on the other side of the Keyboard as "flats"; these can be linked together to play the same sets of reeds …." It will be noted that the repeated keys are the same as on Wicki's keyboard, with the same duplicated keys within an octave as were discussed above. (The two arrays break the edges of the pattern at different places on some rows, but the pattern is the same.)

There is no doubt that Brian Hayden independently re-discovered this keyboard layout in the late 1960s, as he has clearly described, and no doubt also that the idea was unfamiliar to (nearly) everyone who learned about the Hayden system thereafter. There is nothing about the unearthing of the lost Wicki patent now which should detract in any way from Hayden's originality nor from the interest which his system has for concertinists.6

But it is not true that the total human experience with the design is confined to the handful of people who have contrived to buy Hayden instruments since the 1970s. Instead, as we know now, this keyboard design was known to at least some players of large free-reed squeezeboxes (so-called "unisonoric" bandoneons—not unlike large duet concertinas) as early as 1896, more than one hundred years ago. Indeed, the “Wicki-Hayden” system (using this to mean a Hayden-system duet with a button field 9 columns wide) is only twelve years younger than the Maccann system.

It may well be that the “Wicki-Hayden” system has just never been tried adequately, that some great performers on it may yet come along, that it is the next great idea in concertinas, and so forth. But it is also true that the “Wicki-Hayden” system turns out to be in a position not unlike that of the Wheatstone Double system, a system which has been proposed but which failed to gain popularity—indeed, failed so completely that it was very nearly forgotten, even by those most interested in free-reed instruments.

The “Wicki-Hayden” system may, in fact, be better suited to the bandoneon, and to other instruments rather larger than a concertina. The unisonoric ("same note on push and pull") bandonion illustrated at the top of this page, for example, was made by Harry Geuns in 2003–4; it uses the "Kusserow" keyboard rather than the Wicki keyboard, but it is about the the same size, with more than 50 keys on each side. This is nearly the same size as the Wicki system, so it serves an an example of what a Wicki instrument might look like. (According to Geuns's layouts, on the right side the Kusserow arrangement of notes along the rows mostly increases by half-tones in chromatic order, like a wider version of the early Wheatstone "Double" duet system, whereas the Wicki always increases by whole tones in chromatic order.) It has about twice as many keys on each side as a Hayden system concertina, which typically has 25–30 buttons on each side. In order that a player may reach all the keys, it has a loose hand-strap running from edge to edge.

In the experience with recent Hayden-system duet concertinas, a very frequent complaint has been that the number of buttons on the concertinas was too limited, so that the regular pattern of notes was interrupted by "edge effects" as the player ran off the button field and had to "wrap around" to the other edge, or to alter fingering patterns for lack of enough rows. An instrument limited to 46 or 57 or even 67 keys has seemed too small to make good use of the button layout—some players call for larger instruments with more repeated notes and more rows, while others caution that large instruments are heavy and clumsy.

The conflict seems intrinsic to the “Wicki-Hayden” system. The full system (with duplicated keys to right and left) is what makes it possible to play with uniform fingering and in any key signature with the same fingering, but it requires 18 buttons per octave where the Maccann system, for example, uses 12 buttons per octave. So a 46-key Maccann would have the range of a 69-key “Wicki-Hayden”, and to have the range of a 67-key Maccann would require a “Wicki-Hayden” with over 100 keys--and that is very nearly what the 51 or 54 keys per side of the Wicki patent provides. But without the extra keys in each octave, the uniformity and the ability to transpose with constant fingering is lost. Indeed, it seems that the compromises required to fit the “Wicki-Hayden” system onto a smaller concertina limit or remove entirely many of its advantages. (See Which Duet Concertina—Hayden or Maccann?)

Wicki System:
Figure from the Wicki Patent (1896)
Dunkel redrawing from Wicki Patent

Hayden System:
Figure 12 from the Hayden Patent (1986)

Contents

• The Wicki System—an 1896 Precursor of the Hayden System
• Diagram of Wicki System (1896)
• Diagram of Hayden System (1986)
• Notes

© 2004 Robert Gaskins.

Links to related documents

• Other articles
  by Robert Gaskins