Eli Colman's Box of Bells

Almost all the information in this section comes from Rod Howell. His article, "Eli Colman and his 'Box of Bells'." Folk Music Journal, vol. 12, no. 2, has more details on these instruments. (References). Many thanks to him for allowing me to use his photographs. 

Eli Colman (1842-1919) lived and worked as a carpenter in Pitminster, Somerset. He was a life-long bell-ringer, a hand bell ringer and he also played the violin.  He constructed and played instruments of his own invention too, and in 1869 in the Taunton Courier had a report on the Taunton Working Men’s Association Industrial Exhibition which included:

    ‘E. Coleman [sic], Corfe, musical box of bells made by himself’.

Colman is mentioned, over the years, in the local parish magazine for Pitminster. In 1878:

  ‘the playing of Mr Colman, quite a village genius, on two instruments of his own invention’

And in the same parish magazine, this time in 1879:

‘Mr Colman played on The Bells, a musical instrument of his own invention’.

There are other references in the parish magazine to Colman playing ‘the Bells’ and, in one instance:

  ‘Mr Butler and Mr Colman [.  .  .] played on “The Bells”, a musical instrument made by the latter’. 

The Taunton Courier, March 13 1878 also has an account of an entertainment in Pitminster with a Mr Eli Coleman [sic] playing 'bells'

'Another feature of the evening was the playing of Mr. Coleman on two instruments

of his own invention-'

As well as these there are also a couple of references to Colman playing the fairy bells and it might seem that Colman’s ‘Bells’ instrument was actually the widely known fairy bells (widely known at that time) and that Colman was claiming to be the inventor. But some of Colman’s instruments survive and they are distinctively different from the fairy bells. Rod Howell is surely right to suggest that in these particular references, Colman was playing an instrument of his own invention, different from, although in some ways similar to, the more familiar fairy bells.

In 1988 Rod Howell spoke to Eli Colman’s grandson, Hugh. Hugh Colman said that his grandfather made many instruments over forty years in his workshop, alongside his main job as village carpenter and maker of coffins. Rod also spoke to Clarrie Lee (who died in 2002 at the age of 110). She had been postmistress in Colman’s village and still remembered him. Both Hugh Colman and Clarrie Lee confirmed that Eli made two-course and three-course instruments. The smaller, two-course instruments, were swung when played but the larger, three-course instruments were played on a table.

Rod Howell has a Colman instrument with 15 triple courses and on the underside, in pencil, is written: ‘This Box of Bells was made by Eli Colman Blagdon Taunton in the Queen’s Jubilee year 1887’. Colman called his instrument the ‘Box of Bells’ although his local parish magazine simply called it ‘the Bells’, but both his grandson, Hugh, Clarrie Lee and other local players referred to these instruments as ‘boxes of bells’ too.

As can be seen, this instrument is larger than a typical fairy bells (length c.77 cms) and it is very robustly built. It has a soundboard with decorations rather than a typical fairy bells which is a simple box built on a baseboard with planks for the hitch pins and the wrest pins, sides, and a lid with slots allowing access to the wrest pins. In fact this Colman ‘box of bells’ isn’t designed to have a lid or lids or any kind of top and so it is quite different from both the 18th century bell harps/English harps and the later fairy bells of Colman’s own time. A distinctive feature of bell harps/fairy bells, unlike dulcimers and psalteries, is that the strings are enclosed but Colman has not followed this kind of design and has deliberately left the soundboard visible and indeed has ornamented it. In overall outline the ‘box of bells’  seems to look back to the 18th century instruments and it is clearly different from the plain rectangular shape of the fairy bells. 

Like the fairy bells, and unlike the 18th century instruments, the wrest pins are on angled planks away from the playing area. On bell harps and English harps the hitch pins are on angled planks and the wrest pins are near the hands behind the playing area.

The crucial element that links this ‘box of bells’ to the fairy bells and the 18th century bell harps and English harps is the stringing arrangement. The word ‘JUBILEE’ is incised onto across the soundboard in the playing area, showing that the player must be playing at the narrower end with the strings running away from the player, just like playing a bell harp or a fairy bells instrument.

With the player in playing position, for the left hand there is a left-hand side set of strings with the strings ascending left to right, and for the right hand, a right-hand side set of strings with the strings ascending right to left. Just like the bell harp and the fairy bells, the scale ascends at first from left to right towards the middle and then from right to left towards the middle. This can be seen in the photograph by looking at the two angled wrest planks with the two hitch pins for each triple course. If the wrest planks with the single wrest pins were removed this instrument would look like a normal bell harp/fairy bells-type instrument.

But this Colman instrument is different from other bell harps/fairy bells. Both the left-hand side and the right-hand side of this instrument have two wrest planks. Each triple course of strings has two strings fastening at wrest pins on one wrest plank and a single string fastening at a wrest pin on a different plank. On the left hand side the single string member of each triple course is shorter and so, presumably, at a higher pitch. Rod Howell suggests that this string could have been tuned an octave higher. On the right hand side of the instrument, however, the single string is on the wrest plank further away from the playing area and so longer and presumably lower in pitch than the other two strings making up the triple course. It’s a puzzling set up and when Rod Howell acquired the instrument it seems that only one gauge of string, 0.64mm, was used throughout. Rod wonders whether this instrument was an experiment.

Despite these peculiarities this is a 15 triple-course instrument with the distinctive left-right then right-left ascending scale, unique to bell harp-type instruments. It is too heavily built to be swung and so it would have to be played on a table or perhaps on the lap. Its 15 notes, tuned diatonically, give two octaves. A 15-course swung instrument, unlike this one, would have to be primarily a melodic instrument given the need to hold the instrument and play with the thumbs. But this ‘box of bells’, played on a table, would leave each hand to move around more freely and perhaps use fingers as well as thumbs to pluck the strings. 

Rod Howell also has a later 15-course Colman instrument. On the underside, in pencil, is written: ‘No 200 Made by E Colman Blagdon Taunton 1895’.

On the soundboard the letters E C B are inscribed, probably indicating ‘Eli Colman Blagdon’. This is a much more straightforward instrument. There are eight courses on the left-hand side and seven on the right. This sort of arrangement is similar to 18th century instruments but this  instrument is very robustly built indeed compared to bell harps/English harps and weighs nearly 5kg with a soundboard thickness of about 2 cms!

Like the 1887 instrument this one is too heavy to be swung and again there is no lid or top.  It also appears to have been strung throughout with a single gauge of piano wire, 0.64mm. 

Rod Howell has restored this instrument and, played with fingers and thumbs, psaltery-style, it has a very attractive sound, despite its heavyweight construction.

Although the two have some differences, as a pair, they are distinctively different from any of the 18th century bell harps or English harps. And they are very different from the fairy bells that were much more popular at the time when Colman was making his ‘boxes of bells’. 

Rod Howell has a photograph of a third box of bells, this time a 15 double-course instrument, made by Eli Colman which, in 1988, was in the possession of Clarrie Lee and had been for most of her life.

As this photo from 1988 shows, the instrument was not in good condition, and it is not even known if it still exists. It was made a lot later than the previous two instruments, c.1914, and it is different again from either of the two previously discussed instruments. Clarrie Lee confirmed that Eli Colman made the instrument. The raised, curved sides, the robust wrest planks and the ornamental piece of wood covering the hitch pin block on the soundboard, do indeed look like Colman’s work.

It is a smaller and lighter, two-course instrument, and again without a lid. Unusually, the two angled wrest planks are equidistant from the playing area. On bell harps, English harps and fairy bells the right-hand side wrest plank is closer to the playing area to allow the string lengths to get shorter throughout the range of the instrument. This arrangement on this c.1914 instrument is very rare but not unknown on the fairy bells, for example this:

This very unusual fairy bells instrument also has both wrest planks equidistant from the playing area. It is in a German museum but its origin is given as ‘England?’. 

The c.1914 ‘box of bells’ has an additional, surprising feature: a diagonal strip of wood, like a bridge, with ‘alignment’ pins, going across the soundboard in the area between the wrest pins and the hitch pins. 

Rod Howell made a diagram of how the strings go over the intervening diagonal bridge-like strip of wood:

The pins effectively stop the strings and shorten the string lengths. The strings run only as far as the diagonal ‘brIdge’ and not to the wrest plank. It would be completely baffling if this central diagonal ‘bridge’ with its alignment pins ran across the whole soundboard. If that were the case, the actual positioning of the wrest planks would be irrelevant and the strings would descend in pitch from left to right across the instrument. This, of course, is the opposite of all keyboard instruments, xylophone-type instruments etc

But, in fact,  the five courses on the left side run the whole length from hitch pin to wrest pin and so, as usual, the strings ascend in pitch from left-right. The other strings, ten pairs, ascend in pitch, as usual for these instruments, from right-left. In fact this diagonal ‘bridge’ is exactly where the right-hand wrest plank should be for this sort of instrument. So, as usual, the string lengths get progressively shorter at first from left-right and then from right to left. 

Why, on this instrument, did Colman place the right-hand wrest plank at the same distance from the hitch pins as the left-hand side wrest plank? It ruins the gradual shortening of the string lengths as the scale ascends. Perhaps he liked the symmetry of the equally placed wrest planks and intended to use strings of different gauges. Perhaps that was unsatisfactory. Whatever the case, the diagonal ‘bridge’ between the hitch pins and wrest plank looks like an ad hoc addition to make a satisfactory instrument. 

The instrument has only five courses on the left and ten on the right. That seems a bit surprising but the next instrument to be discussed has only four courses on the left side and ten on the right. 

All three of these Colman instruments have differences but they all look, as it were, similarly different from other bell harps or fairy bells. Also the 1887 and c.1914 instruments have some very odd features, Colman must have been interested in experimenting with different ideas and designs.

Recently another instrument has surfaced which is probably another of Colman’s. I got it from an elderly woman in Northampton who was convinced it was some type of harpsichord. It now belongs to Rod Howell and he has restored it.

This 14 double-course instrument is also heavily built. Like the other Colman instruments It has no lid and it has raised, stepped and shaped sides and robust wrest planks with large wrest pins. Although it has only two strings per course and according to Hugh Colman and Carrie Lee, Colman’s ‘two-wire’ instruments were swung, this one is far too heavy to swing. When I got it, It was strung with different gauges of modern acoustic guitar strings which were lifting the wrest planks. I wrote about it in FoMRHI (see References).

This instrument does look much more like a Colman ‘box of bells’ than any bell harp or English harp or fairy bells, even though it is somewhat different from the other Colman instruments above.

This instrument has a groove in the hitch pin block for a metal saddle but it is missing. There is a small floating bridge with a metal saddle, seemingly arbitrarily placed. It may be a later addition but it is difficult to see what possible use it could have. The brass knob is simply a knob for undoing a piece of wood that covers the hitch pins (the hitch pins are, in fact, screws).

The wrest  planks are placed in the usual way for bell harps/fairy bells so that the string lengths get progressively shorter, at first from left to right and then from right to left. But there are only four double courses on the left-hand side and ten on the right-hand side, giving an uneven distribution of effort between the two hands. The left hand has just four notes on the left side and the right hand has ten notes to play on the right. Fourteen notes tuned diatonically is somewhat unsatisfactory. In the key of C, starting on C, the top note would be a B, the leading note, with no possibility of leading to the tonic. Perhaps the instrument was not tuned diatonically throughout. The left hand could be playing the tonic, dominant, subdominant and one other note, acting as harmonic basses to accompany simple melodies with up to ten notes played by the right hand. But with the instrument on a table the two arms and hands are free to move around all the fourteen strings and some other way of tuning and playing might have been used.

Here is the instrument restored by Rod Howell:

The floating bridge and the brass knob have been removed. A broken-off part of the right side has been replaced and the whole thing has been cleaned up and oiled. Rod has chosen replacement strings based on his experience with the other two Colman boxes of bells. Currently he has it tuned C D G A on the left side and B C D E F G A B C D.  Played with the flesh of fingers and thumbs, it produces a very attractive sound, rather like a musical box.

The instrument (photo taken before restoration)  has a number of metal ‘ornaments’ inset on the soundboard.

Rod Howell has recently found a book from the 1920s illustrating different designs of coffin nails: 

As Colman made coffins for many years and this instrument uses coffin nails as ornaments it would seem to be conclusive evidence that Colman was its maker. However Rod Howell has examined the instrument in detail and there are some significant differences in construction between this and the 1887 and the 1895 instruments above.

The Northampton instrument does look rather different from the other two. It is possible, then, that instrument was not made by Colman but it was surely made by someone with a knowledge of Colman’s instruments, perhaps a friend or family member. However, as Colman made instruments over a long period of time and the first three instruments described above are all different from each other, it is not really surprising that other instruments that he made would have significant differences too. There is no evidence of anyone else, anywhere else, making instruments like these.

Every 18th century reference to the bell harp states that it was played with plectrums attached to the thumbs and that a plectrum was needed to properly sound the strings. The bell harp typically has three strings per course but sometimes four. In contrast the single-string fairy bells instrument was played with the thumbs alone. It is not known how Colman played his instruments but, following the manner of play of the 18th century instruments, plectrums would be expected. However the piano wire used by Colman would have been different from the brass and steel wire used in the 18th century. On the restored Northampton instrument Rod Howell is using modern strings from the 21st century.

Eli Colman’s ‘box of bells’ is mentioned as early as 1869, two years before advertisements began to appear for the fairy bells. Of course it is possible that the fairy bells instrument existed before the adverts from 1871 but it does seem to be the case that, whether Colman’s ‘box of bells’ did or didn’t actually pre-date the fairy bells, it had a completely independent existence. And the ‘box of bells’ itself, on the basis of these four instruments, describes a series of similar instruments rather than a basic, standard ‘box of bells’.

Three of these four ‘boxes of bells’ are far too heavy to swing so they cannot be swung zither instruments. There are some reasons to suspect that the two bell harps in Edinburgh were not swung but both could have been; they are small enough and light enough. Colman made smaller instruments to be swung and larger ones that could not be swung. Without the possibility of being swung (and without lids to enclose the sounding strings) can the larger Colman ‘box of bells’ be considered a type of bell harp? Colman’s larger  ‘box of bells’ is a zither instrument but with some fanciful pretensions to being bells and it has the, as yet unexplained, left to right and then right-left ascending scale design. It is, perhaps, a borderline case.

It is interesting that in the few nuggets of information that Rod Howell was able to unearth there is no mention of the bell harp itself. It’s also an interesting but unanswerable question: what inspired Eli to create these instruments? His instruments are not like any kind of fairy bells and seem to be entirely independent of them. But he named his creations, ‘boxes of bells’. He must have thought that the instruments were somehow bell-like. They are, in a way, more like bell harps/English harps than they are like fairy bells but they are not much like the 18th century instruments either. Crucially they have no lids and there is no possibility of swinging the larger instruments and so no possibility of creating the unique acoustic effects that result from swinging.

Colman’s instruments, like 19th century dulcimers, were a product of the availability of piano wrest pins and heavy duty piano wire resulting in heavily constructed instruments. They are far heavier than the lighter 18th  instruments but are designed for thicker, high tension piano wire.

 I think it is unlikely that Eli Colman ever saw a bell harp or English harp but perhaps he had read about them or heard about them and was inspired by his own idea of them. Rod Howell has pointed out that Colman lived not far from Bath and Bath is where John Simcock announced his invention of the English harp in 1760. Simcock is only known to have lived in Bath for a few years and by 1764 he had moved to Leek in Staffordshire but it is possible there is some link between Simcock's instruments and Colman's.

It seems most likely that Colman played popular, familiar tunes suitable for a diatonic instrument. The report from the Taunton Courier (March 13 1878), mentioned earlier, has this about his performance on the two instruments of his own invention:

'one reproduces very sweetly a full peal of bells, the other a merry jig, or dare was say

something not quite so lively, if required, only its powers were not tested in that 

direction.'

Probably it was the smaller instrument, swung about, that created the sound of a peal and the larger sort of 'box of bells' that he used to play the

jig.