Unsigned, undated bell harp with a watch key tuning mechanism
This instrument, M332 in the Horniman Collection, is the most highly crafted of all the bell harps/English harps listed here. It was. perhaps, a top of the range model, designed to be an elegant object in a wealthy person’s drawing room. It is certainly not a street musician’s instrument nor something contrived by an amateur.
It has no lid, no lugs and no septum defining a left side and a right side. It has brass strings and three roses. The sides are c.53cms in length and across the playing end, c.22cms and the bottom end, c.32cms. Unlike Simcock’s English harps, M332 has a soundboard in which the roses are placed (see English harp). The soundboard height is c.3.5cms.
Unusually, M322 has two ditals on the right side for altering the pitch of only two of the notes but nevertheless extending the basic, diatonic tuning.
The watchkey mechanism and the two ditals
Apart from looking elegant and expensive the most noticeable feature is the watch key tuning mechanism. All the other bell harps/English harps use wrest pins for tuning, as used on keyboard instruments. The watch key mechanism itself must have been much more expensive to produce compared to wrest pins (simply made from small nails). In all there are 52 strings arranged as triple courses except for a central, single course. On the left-hand side there are eight triple courses and on the right-hand side there are nine triple courses and in the middle, a single course.
The design of the roses and the watch key tuning mechanism obviously suggest a connection to the wire-strung English guitar that was very popular indeed in the second half of 18th century Britain. Nowadays this instrument is usually referred to as the ‘guittar’ (with two ‘t’s, although in the 18th century it was spelt either way and had other names too). Many guittars had wooden star roses just like the roses on M332 and many guittars had a watch key mechanism for tuning the strings rather than wooden pegs.
Examples of ‘star’ roses made of wood and bone or ivory.
Watch key mechanism on guittars.
The guittar was a very popular instrument for amateurs to play and a lot of music was published for it. It was especially popular with women and there are images of wealthy women holding a guittar as a sort of fashion accoutrement.
Georgian women and their guittars
The popularity of the guittar was even described as a ‘craze’ and it began in the mid 1750s. Simcock’s invention of his English harp (presumably a development of the bell harp in some way) in 1760 might have been his attempt to create a similar craze for his instrument and aimed at the same market of well-off amateurs. M332 is unsigned and unlikely to have been made by Simcock, but is probably aimed at a similar clientele.
According to Andrew Hartig, https://www.amhstrings.com/restorations-repairs/18th-c-guittar-rose-reconstruction guittars had either no rose but just a sound hole (very rare) or three different kinds of rose: carved into the soundboard (again rare), constructed from wood and bone (or ivory), and inserted into the sound hole or, finally, a cast metal rose inserted into the sound hole.
Cast metal roses were used for the most expensive instruments. Wood and bone/ivory roses are ‘commonly found on mid-grade instruments of modest or moderate figure and ornamentation’. (see Hartig, who also explains how these roses were made). Although M332 looks more expensive and elegant than other surviving bell harps/English harps, perhaps there were even more expensive models! On the other hand, M332 could well be, and probably is, unique.
The very earliest known watch key mechanism on guittars are from 1756 and 1758. See Poulopoulos 2011 https://era.ed.ac.uk/handle/1842/5776. The watch key mechanism for tuning guittars is usually thought to be later than this but these early examples do seem to be genuinely original, not later adaptations. Generally though, guittars with watch key tuning mechanisms appear from the 1760s onwards. Poulopoulos even suggests that watch key mechanisms began to appear on other instruments than guittars only later in the 18th century. He cites the examples of cither viols (‘sultanas’) from the 1780s. It is possible that M332 was made around that time too.
This instrument is probably a one-off and probably dates from the 1760s-1780s. As far as is known today, interest in the bell harp was waning by the end of the century so it was probably not made later than the 1780s.
As watchkey guittars are much more well-known than watchkey bell harps - M332 may be the only one - it’s probably fair to suppose that the maker of this bell harp was influenced by guittars rather than the other way around.
But what is the point of having a watch key tuning mechanism on a bell harp? Guittars, with wooden pegs and thin wire strings, can go out of tune easily and can be difficult to get back in tune. Although this is, to some extent, true, many guittars were made with wooden pegs. Citterns and other wire-strung plucked instruments (bandoras, orpharions etc) used wooden pegs with wire strings for over two centuries before the introduction of the watch key mechanism. One can cope with wooden pegs and tuning mechanisms on guittars were, to some extent, a marketing strategy of the time. But it is also true that since the 19th century, plucked instruments with metal strings have used some kind of mechanical tuner rather than wooden pegs and a variant of the watch key tuner is still used on the Portuguese guitarra.
Apart from M332, bell harps and English harps use wrest pins for tuning the strings (as used on keyboard instruments) and they are not unstable in the way that wooden pegs on a plucked instrument can be. There can't be a great difference between tuning wrest pins with a tuning key and tuning a mechanism with a watch key. There is no great need for a watch key mechanism on a bell harp but it would certainly make accurate tuning easier, especially on the higher notes on the right-hand side.
Wrest pins on a Simcock English harp
On bell harps/English harps, tuned with the usual wrest pins, the wire strings are attached the same way as on a keyboard instrument. The wrest pins have no hole for the wire. The wire is attached using a special technique and then wound around the pin many times. But with a watch key mechanism each string, for each string length, has to have a loop for both the hitch pin and the wrest pin. Each string has to be just the right length. With wrest pins it is possible to wind as many times as needed, but with a watch-key mechanism string lengths and loops have to be much more precise. Having a supply of ready-made looped strings for each course would make it very easy to fit broken strings. So a possible, very slight, advantage of a watch key tuning mechanism on a bell harp would be the ease of changing a broken string for an amateur player. However it is more likely that the watch key mechanism on this instrument was used for its striking appearance and, probably, with its association with the popular and fashionable guittar.
Could M332 have been made by a guittar maker? There were many: Hintz, Preston, Rauche, Liessem, Claus and others. This is a possibility but the specialised skills and knowledge that plucked instrument makers need to have are mostly irrelevant to the construction of a bell harp. At a guess, a maker of keyboard instruments seems more likely.
One last thought on the watch key mechanism on M332. On guittars, the watch key mechanism is sometimes thought, negatively, to add to the weight of the instrument and possibly to affect the sound too. But M332 really is surprisingly light. I didn’t notice any appreciable difference in weight when I examined it alongside a Simcock and an unsigned instrument in the Horniman Museum.
It is true that there are no lugs for holding and swinging M332, but there are no lugs on other bell harps and all the 18th century references confirm that bell harps were swung. There would be no problem about holding and swinging this instrument.
The instrument has no lid and there is no record of it having one. Without a lid its three roses and varnished soundboard can be seen and fully appreciated. On Simcock’s instruments, for example, the area of the baseboard that is covered by the lids is unfinished. The Simcock instruments are not meant to be seen without lids.
Simcock English harp with unfinished area on baseboard. The lids cover this area.
Unlike M332, other bell harps and English harps have a septum which marks off the right-hand side and the left-hand side and also supports the lid or lids in some way. M332 has no septum and a single string course is in the middle of the instrument. Perhaps M332 never had a lid. It is possible that a seller in the past thought that the instrument looked more interesting, collectable and valuable without a lid. And so the lid was discarded.
But every other bell harp, English harp and fairy bells has a lid or top so that the strings are partly enclosed during play. And there are slots on the sides where a lid could be slotted in.
Grooves on the insides of the sides
There are slots or groves on the interior of the sides and they have no obvious function and they are not decorative. They are in the same positions as the grooves for the lids in Simcock's English harps.
Simcock English harp with grooves in the sides
If the slot on each side of M332 is followed from the bottom of the instrument to near the playing area a curious, semi-cylindrical piece of wood can be seen. There is one on each side of the instrument.
Semi-cylindrical projections on each side.
The outside string of the outside course touches the wood.
They don’t seem to have any function and, worse still, the outermost string on each side of the instrument actually touches them.. Playing the outermost courses on each side would create an unwanted noise as one of the triple courses hits the wood. This surely cannot be right on an otherwise professionally designed and constructed instrument. If these semi-cylindrical pieces were removed there would be no problem and the strings would sound clearly.
Why are these semi-cylindrical pieces there? They didn't want to move with light pressure applied to them so it is possible that they are glued to the sides for some unknown reason. But it’s very tempting to suppose that they do move but have jammed in place over time. Could it be that they are set in the grooves ready to slide out to permit a lid to be put in? And if this instrument was swung when played and therefore held downwards, the lid would slide out unless something held it in place in the grooves. The Simcock English harp and the very similar unsigned instrument in the Horniman Museum both have little pins that pass through the lids and fit into a loose-fitting transverse bar and prevent the lids from falling out of the grooves when the instrument is held downwards for swinging.
Unsigned English harp with one of the pins that holds each lid in place.
In M332 perhaps these semi-cylindrical pieces are not only removable but can be put back into the grooves and somehow locked to keep the lid from falling out.
The showy upper large rose in the playing area is, and would always be, visible with or without a lid. The lid, if it had one, could have been designed to expose the two lower roses too, while still enclosing the strings. It’s only a conjecture but I think that this instrument did (like all other bell harps, English harps and fairy bells) have a lid.
The back of M332 is quite plain and unfinished. On the back of Simcock instruments, which are built on a baseboard, there are screws for the two angled hitch pin planks, with two screws for each.
Back of s Simcock English harp.
The same is true of the later fairy bells, except the hitch pins and wrest pins are reversed and the screws are for the angled wrest planks. M332 has a separate soundboard so the two hitch pin planks or bridges are either simply glued to the soundboard or there is some additional strengthening support underneath. (See also MIMed 3360 in the St Cecilia's Museum in Edinburgh Instruments 1). The soundboard on M332 looks quite thin, not as thin as a guittar soundboard, but too thin to withstand the tension of multiple wire strings, so there must be some strengthening support for the hitch pin planks underneath the soundboard.
This instrument must have a more complicated tuning arrangement than all the others except for the V&A Simcock English harp (with its two octave, almost fully chromatic, tuning).
See Tunings for more discussion of this instrument.
Simcock English harp with (almost) chromatic tuning in the V&A.
M332 has two ditals for changing the notes of two of the courses on the right side of the instrument, and two of the courses are offset from their expected positions on the left side. One offset course is shorter than expected and one is longer, presumably a semitone higher and a semitone lower.
The two offset bridges on the left-hand side.
The offset positions are puzzling: surely a brass string suitable for one pitch could tolerate being tuned either a semitone higher or lower without either snapping or being too loose to sound properly. The offset positions could be there to underline the fact that they are deviations from the expected diatonic range of notes.
There is also a central single-string course. Currently this is just another thin brass string. The near-chromatic V&A Simcock English harp also has a central single string (or nearly central because it has a septum). Carl Engel, writing in the 1870s, (see References) described it as a wound string and the lowest note of the instrument. But M332 seems different. The longest course, as on Simcock’s other instruments and on Edinburgh 3360 (see Instruments 1) is on the right side and thereafter, ignoring the central string, the strings ascend as expected as usual, at first from left to right as far as the centre of the instrument and then from right to left from the right side of the instrument again towards the middle. The central, single string is a puzzle. As a single string course in contrast to all the other triple courses it would sound noticeably different. Perhaps it had some special use.
A possible stringing/tuning set up could be this: lowest note on the right side is g. Then on the left side and ascending the hitch plank, the notes are a, b, c’, d’ e’ f.’ The offset note below the hitch pin blank is d#’ and above the hitch pin plank is f#’.
On the right side the notes are first, the low g, then, carrying on with the ascending scale: g’, a’ b’ c’’ d’’ e’’ f’’ g’’’. The ditals affect the notes c’’ and f’’ which can be altered to c# and f#.
Ignoring the offset notes and the ditals the plain notes are simply g-g’’. With the offset notes and the notes using the ditals:
(RH) g (LH) a b c’ d’ d#’’ e’ f’ f#’ (RH) g’ a’ b’ c’’ c#’’ d’’ e’’ f’’ f#’’ g’’
That is two octaves from g-g’’ with some chromatic notes available. This gives two octaves of C major, two octaves of G major and not quite two octaves of D major. Within those scales are the natural minors and the ‘folk’ modes, dorian and mixolydian, which are sometimes useful in country dance tunes and in some Scottish airs.
All these notes are sounding from the triple courses. That leaves the central single string unused. If it is really meant to be a thin wire, as a single string course, it would sound noticeably different from all the other triple courses. Perhaps it had a use as a kind of special effect? Otherwise, if originally it was a thicker, wound string, it could be a low c or d.