DESIGN CONCEPTS AND INNOVATION IN LONDON INSTRUMENT MAKING

 
Musical instrument makers in the seventeenth century were very aware that their purpose was not to create anything mystical or magical, and were instead concerned with more mundane details - how to make the instruments that players wanted, and how to make them as economically as possible. Instrument makers are providers of a service to musicians, and as such the surviving examples tell us mostly about their customers. It is only by a deeper and more thorough examination of the instruments that many of the details about the makers and their concepts, approach and methods emerge.
Despite all of the turbulence in seventeenth century England, the products of instrument makers have been both standardised in the sense of belonging to characteristic design types, and developing as musical and artistic tastes changed. The late seventeenth century saw the change from polychrome, highly decorated exuberance in visual design to the more restrained and smooth flowing decoration which reached perfection with Queen Anne style furniture in the first decades of the eighteenth century. In domestic keyboard instruments, for example, the mid-seventeenth century saw the production of many English virginals, all seemingly identically and overpoweringly decorated - the rectangular cases and coffered lids having natural oak cases, varnished to a dark brown colour and decorated on the outside only by iron strap work hinges and lock plates. But when opened, the instruments show a blaze of colour - polychrome paintings on the lid and fallboards; the faceboard, keywell and soundwell all decorated with expanses of mouldings using natural cedar, occasionally even painted green or maroon, contrasting with gilt embossed papers and painted panels. The soundboards are decorated with flowers and birds, with decorative border patterns and scrollwork. The painted decoration makes use of unmixed pigments and, on the lid and fallboards, are painted over a brilliant white gesso ground. Although no one will accuse the decorative work of being very fine quality, the brilliance and vigour is still very evident today and must have been even more striking when new. These instruments and their style of decoration, common as it was, gave way to the bentside spinet whose graceful shape was accentuated by casework of natural solid wood, usually walnut, though occasionally cedar or oak. The interior is usually of cedar veneer, the soundboards are undecorated, usually not even having a rose. The keyboards have ebony naturals with solid bone or ivory accidentals. The only decoration comes from the brass hinges on the outside of the cases and occasionally from marquetry inlay on the nameboard above the keys.
These changes must of course be considered as having been dictated by the customers taste, but there are musical consequences which also accompanied the change of design. The graceful size and shape of the bentside spinet comes in a great part from its small dimensions which has a direct bearing on the tone of the instrument. Although makers could use brass strings in the treble to compensate for the shorter string lengths and retain the same pitch standards, there is no equal alteration that can be made in the bass as the same stringing material is used in both virginals and spinets. By shortening the string length of the bass strings, as happened in spinets, the sound quality deteriorates. The shorter strings, coupled with a shallower case and smaller vibrating soundboard area makes a very noticeable difference in sound quality, something which must have been as obvious to seventeenth century players as it is to us today. The small size of spinets seems to be connected to a reduction in the number of notes as well. Virginals of the 1660's and 1670's typically had 55 notes, a compass going from GG/BB by way of a short octave in the bass, to f3 in the treble. Occasionally there were even more notes. The earliest spinets often had only 50 notes, going from the same bass but only going as high as c3 in the treble, and not until the 1690's did the treble usually go up to d3. Although the purpose of the virginal compass going to f3 is uncertain, in that the earliest English piece I know of that goes as high as d3 is Mortlack's Ground, by John Blow, dated 1689, the compass reduction must have made an impression on musicians. It is for musical reasons that spinet compasses in Purcell's time increased in a downwards direction, though it is achieved by splitting the lowest two accidentals so that the dimensions of the spinet did not need to increase. It was only in the second decade of the eighteenth century that instruments began to have a chromatic lowest octave down to GG, though even then the low GG# was often omitted.
Of course, we cannot be certain that the surviving musical instruments are representative of those in common use at the time, but perhaps we can make some judgements about them. There are 25 keyboard instruments of the earlier virginall type, 22 of them rectangular instruments and 3 harpsichords. Two of the harpsichords are dated 1579 and 1622 and both pre-date the earliest dated rectangular virginal. There are about 30 to 35 bentside spinets of late seventeenth century style - though many are undated and may actually be from the first decade of the eighteenth century. Again there are three harpsichords - the Charles Haward single manual instrument dated 1683 at Hovingham Hall in Yorkshire, the Joseph Tisseran double manual at the Bate Collection in Oxford, and a Thomas Barton single manual in the collection of Dr Rodger Mirrey in London. In addition James Talbot gives measurements of a single manual harpsichord, probably an instrument built by John Player around 1685.
There are other references to the manufacture of harpsichords as well. For example, Thomas Salmon, writing in his "Vindication of an Essay to the advancement of Music" in 1672 refers to a "curious pair of phanatical harpsichords made by that Arch Heretick Charles Haward". Stephen Keene's trade card printed in Playford's "Introduction to the Skill of Music" in 1672 mentions him as a maker of "Harpsycons and Virginals"; an inventory on the death of Thomas Britton refers to a harpsichord by Philip Jones, who has left a surviving virginal dated 1671. A letter preserved with the Tisseran harpsichord refers to a similar instrument by John Player, and John Hingston had a position in royal service which was later taken over by Henry Purcell as the "keeper and repayrer of his majestie's organs, pedalls, harpsichords and other instruments". It is surprising that there aren't more large harpsichords still in existence, but the evidence suggests that the smaller instruments, rectangular virginals and then bentside spinets, were the common instruments of the seventeenth century keyboard player. It could be pointed out that perhaps all of the best instruments were played until they were worn out, and the surviving instruments we have now were no good to begin with.
It is likely that it was these small instruments that were the common tools of professional musicians such as Purcell. There is nothing in the music, nor do I know of any contemporary documents to suggest that professional players used the biggest and best and that the smaller instruments were the seventeenth century equivalent of the upright piano, even if that is how we in the twentieth century look on them.
Elaborate decoration is different, and is essentially an example of the owner's wealth, rather than in any way implying a finer instrument. In some cases the more costly materials can lead to inferior instruments. Silver trumpets are an example of this and have a tone which is considered dull when compared to cheaper brass trumpets. William Bull was the most famous trumpet maker of the late seventeenth century and he has left three trumpets, two of which are here in the Museum of London. One of these instruments is in silver, the other of brass. His third trumpet, now in the Ashmolean Museum in Oxford is silver. The silver instruments are most elaborately decorated, particularly the ball in the centre of the bellpipe. However, it is clear that Bull himself preferred to make instruments of brass. James Talbot, when compiling his list of musical instruments and their details, recorded comments from Bull himself and noted that Bull considered "Bastard brass mixed with solid brass" to be the best for trumpets and horns. Bull had a position as a household trumpeter to the King, and was responsible for the repair, and presumably the manufacture, of many of the instruments used in the royal service, and it may be that the ceremonial duty had an important part to play in the use of silver as an instrument making material.
Towards the end of the seventeenth century there appears to have been a change in the approach of instrument makers and the products they offered, moving from what appears to be more-or-less instruments made strictly as custom orders, even if they may have been built using previous instruments as a design basis, to a situation where customers had various models to choose from. There certainly wasn't mass-production in any modern sense, though it is likely makers were producing what is essentially the same instrument as effectively as they could. An interesting piece of evidence which would support this comes early on in the manufacture of spinets at a time when the rectangular virginal was the preferred choice. On April the 4th, 1668, Samuel Pepys visited the workshop of Charles Haward, writing in his diary "...Up betimes, and by coach towards White Hall, and took Aldgate Street in my way, and there called upon one Hayward, that makes virginalls, and did there like of a little espinette, and will have him finish it for me : for I had a mind to a small harpsichon, but this takes up less room, and will do my business as to finding out chords, and I am well pleased that I have found it." That Pepys saw the instrument and was going to get it finished for him, rather than place an order for one suggests Haward was making these instruments speculatively rather than to commission. For some reason Pepys didn't order the instrument at that time, and returned to Haward's workshop after dinner on the 10th of July in 1668. He again balked at buying one, but returned three days later and bought a spinet. The instrument was delivered two days later. 1668 is a very early date for a bentside spinet in England, I can't think of any earlier references and the earliest dated spinet is 1680. Perhaps Pepys was influenced a little by the novelty value, but in any case it seems clear that Haward was regularly producing these instruments on a speculative basis and finding customers for them. Certainly customers could have either a deluxe version or the standard model, depending on if they wanted to pay for marquetry, but it may even be that the maker decided to do a marquetry model instrument speculatively at a higher price, rather than it being commissioned specifically by a customer.
This is in contrast to English virginals where there is very little evidence of any multiple production of models. The surviving virginals are all very individual. Sometimes a common design basis can be determined - Thomas White seems to have used the same soundboard layout on several of his instruments, but in one of them he had a larger compass, extending it at each end. The decoration also seems very much an individual choice. One particular example of this can be seen in two virginals - one by Stephen Keene, the other by Philip Jones. Both of these instruments are in the homes of their original owners and are less than ten miles apart. The Philip Jones instrument was made first, in 1671, and is distinctively decorated - the keyboard having snakewood natural keys and solid ivory accidentals, and the mouldings are green. The compass is 55 notes from GG/BB to f3. The Keene instrument was made in 1675 and shares all of the decorative features - the soundboard and lids were painted by the same artist, the keyboard again has snakewood and ivory, and the mouldings are also green. It seems likely that the ladies of one house saw their neighbours virginal and wanted one just like it. There are some differences between the instruments, most noticeably that the two instruments appear to be designed to play at slightly different pitch standards, which perhaps suggests that the pitch of the instruments was chosen to match other instruments that would be played with it.
This should not imply that when the spinet came along that customers were unable to make any specific requests. All makers have their preferences to do with many aspects of the instruments they make, and one concerns the octave span. The modern piano is entirely standardised so that musicians can go from one piano to the next without any discomfort, but in seventeenth century England that was not at all the situation. I have collated details from as many instruments as I have been able to examine, and it appears makers generally worked to three different octave spans, 6 1/4 inches, 6 1/3 inches or 6 1/2 inches. In virginals, most surviving instruments have been built with a 6 1/3 inch octave span, the others seem to be fairly evenly divided between larger or smaller widths. With bentside spinets it seems makers again had their preferences. John Player mostly used a 6 1/4 inch span, Keene a 6 1/3 inch span. Haward even used a 6 1/2 inch span. But there are examples of individual choice. Two otherwise virtually identical spinets by Stephen Keene, both now in the ownership of an American collector differ in their octave spans, one of them Keene's usual 6 1/3 inches, the other being 6 1/4 inches. The instrument with the smaller octave span was probably made as a special order.
It should be pointed out that there weren't commercial makers for all types of instruments. There were many keyboard makers, both those making the small domestic instruments such as Keene, Player and Haward, and others making organs of both the chamber and church varieties. Father Smith and Renatus Harris were the best known organ makers in London. A number of organ makers also had connections, and probably built, string keyboard instruments as well. The Exeter maker John Loosemore has surviving organs and virginals, an inventory taken at the death of Charles Rewallin shows he made organs as well as virginals, and the virginal maker James White, one of whose instruments is in this museum, was a business partner of Ralph Dallam and completed several organs after Dallam's death in 1673. Purcell and his predecessor John Hingston were also expected to work on organs, as their job description states.
There was also a long tradition of bowed string makers in London which dates back to at least the 1580's. In addition to using old instruments by makers such as John Rose and Henry Jaye, viol players could buy new instruments from makers such as Richard Meares and Barak Norman. There were also violin makers in London - Jacob Rayman was a highly respected maker in the mid seventeenth century, particularly of tenor violins - what we would nowadays call violas, and later both Edward Pamphillon and Thomas Urquhart were producing fine instruments. Perhaps in the seventeenth century Barak Norman began to make violins and cellos, and was certainly doing so in the eighteenth century, as was Nathaniel Cross. Woodwind instruments were also in common use and were made in seventeenth century London. The earlier recorders were single piece instruments, what are now termed "Van Eyck flutes", for example those of the London born maker Richard Haka, and an anonymous instrument in a private collection in Scotland. The French style gained popularity, and there are many surviving examples of French style baroque recorders built by Peter Bressan, a Frenchman who moved to London in 1688. He has also left several baroque flutes. The French style oboe also made its appearance at this time, there are surviving seventeenth century examples in the Bate Collection in Oxford and at the Edinburgh University Collection of Historic Musical Instruments. These oboes replaced the renaissance style shawms, although both existed side-by-side into the 1690's. Brass wind instruments were also made, though probably not so much for the amateur market. Makers such as William Bull made trumpets, Flat trumpets, horns and sackbuts. Talbot, in his collected manuscript mentions other, non-orchestral instruments which appeared to be popular such as cornetts, serpents, harps, and wire strung instruments like bell-citterns, bandoras and orpharions which all appear to have been made in London in the late seventeenth century.
The earlier van Eyck type recorders were invariably single piece instruments, usually built in ivory. The instruments of Bressan followed the French style, and were made in smaller sections, usually of boxwood with ornamental ivory turnings to disguise the joints. This is a major innovation which is both simpler for the maker and more convenient for the player, since should the instrument be damaged he needed to have only the damaged section repaired rather than the whole instrument. Recorders were built in consort and Bressan used simple mathematical relationships in the various sizes of the instruments. Compared to the treble recorder in F, the total external length of the 4th flute in Bb is in a 3:4 ratio, the voice flute in D has a 6:5 ratio, and the tenor in C a 4:3 ratio. Only the bass doesn't follow the same relationship. Bressan's flutes were also built in sections, there are surviving 3 and 4 section flutes. The seventeenth century oboes are in three sections.
Trumpets were also made in sections which were easily dismantled, rather than soldered together into a single pipe. Both the silver Bull trumpet here at the Museum of London, and the Bull trumpet at the Ashmolean Museum clearly show they are made of individual sections which were probably held together with only beeswax to reduce leakage. In an instrument which was as common on the field of war as the concert room such an approach was probably essential.
Other instruments appear not to have been made in England despite their popularity. The lute and guitar are examples of instruments which had many players but didn't have any native makers, musicians relying instead on instruments made in continental Europe. At first inspection this seems strange, since there is nothing inherently difficult about the construction of either instrument, and there were makers of viols which are also gut strung and fretted, and makers of wire-strung fretted instruments. There must have been craftsmen who were intimately familiar with the instruments in a repairers capacity and knew all of the construction features which are normally hidden from view who could have at the least made copies of the instruments they had in front of them. Lutes in particular were highly respected in England throughout the seventeenth century. Early in the century the Elizabethan / Jacobean school of composers such as John Dowland, Anthony Holborne, Francis Cutting, and Robert and John Johnson produced much fine music comparable to the virginalist school of Byrd, Bull and Gibbons. Although the renaissance-style lute fell from fashion it was replaced by the French style lutes, so much so that the two-headed lute that Thomas Mace writes about in "Musick's Monument" is referred to by James Talbot as an "English Two-headed Lute", whereas the French Lute had a single head.
The baroque guitar was also popular, despite its position as a solo instrument and the relatively simple style of music that was played on it. It had players at the highest social level, Samual Pepys was sent to France to get an instrument for Charles the 2nd, and despite an initial dislike of the instrument, Pepys himself seems to have followed the King's lead and played and wrote music for the instrument. Probably Pepys also followed the King's example and purchased an instrument from France, as likely did most of the seventeenth century guitar players, although there were also very fine schools of guitar making in both Italy and Hamburg.
The reason English makers didn't produce these instruments is surely nothing to do with ability or understanding, since makers were repairing these instruments and , in the case of lutes, converting them from the earlier renaissance-style to the two-headed type mentioned by both Mace and Talbot. It is probably simply that making these instruments wasn't economic for English makers. If the customer had a reliable supply and choice of instruments there is little incentive for a maker to try and struggle to get their own custom. It seems that lutes were built very much by some sort of mass-produced factory type methods, at least as far as is possible with a handmade instrument, since inventories taken at the deaths of Laux Maler and Moise Tieffenbrucker each refer to around 1000 finished lutes. To have 1000 unsold instruments suggests that these instruments were being produced and sold very quickly. There are also references of wholesale pricing of instruments, suggesting that makers had agents who sold instruments for them. It is likely that the majority of the English custom was supplied by importers of the continental product or by rebuilds of older instruments. There is nothing to suggest there was ever an increase in popularity that required a larger demand which could have been filled by native makers.
Despite the existence of makers such as Rayman, Pamphillon and Urquhart in London it is possible that the supply of violins was also partly satisfied by continental products. The English court were familiar with buying instruments abroad. Peter Holman, writing in his book "Four and Twenty Fiddlers" mentions the purchase of a "Cremonia vyolin to play to the organ". This purchase took place before the Civil War, but it would still appear, on the face of it, to be the ideal commission for a local maker since he would know the pitch of the organ and what size the instrument would need to be to reach the higher pitch.
There are comparatively few surviving instruments for modern researchers to form opinions as to exactly how a maker designed and built his instruments. However, it is certain that the methods must have been simple, since the master maker had to show the processes to his apprentices and journeymen; and as economic as possible, since the makers needed to sell the instruments so they could eat. There is little evidence to suggest that makers ever sat down at a drawing board and made full scale drawings of the instruments they were going to produce. In fact, the differences between instruments that can be observed suggest that deviations were expected and the methods of construction took this into account.
This is most apparent in bowed string instruments and even more so in keyboard instruments. Viols were probably built around a mould, the maker bending the sides of the instrument until it fitted the solid wood inner mould as closely as he felt was necessary. The surviving instruments suggest that this accuracy may not be as close as modern researchers would like to believe. For example, frequently when examining instruments it appears that the two sides are not symmetrical. A interesting attempt to analyse the shape of viol outlines was carried out by Ephraim Segerman who wrote up his findings in the FoMRHI Journal, number 64, published in July 1991. He makes a passing comment that for one particular instrument he was able to generate three equally convincing geometric design approaches that the maker could have used, all of which were different from another analysis by another researcher who also used geometric analysis. In the unlikely event that the maker himself used all four methods to produce one outline it is clear that at least three of the equally good interpretations are wrong. Another example of geometric analysis was published in an earlier FoMRHI Journal by a different researcher who concluded that a 1694 violin by Antonio Stradivarius made use of three different foot measurements, two of them Cremonese, and one from Brunswick in North Germany. I am not entirely sure why Stradivarius should need a Brunswick ruler as well as two from his native city. Perhaps most enlightenly, Stewart Pollens, the Conservator of Musical Instruments at the Metropolitan Museum in New York has recently published a book including photographs of the moulds which are reliably believed to be those used by Stradivarius himself. They have no sign of any geometric markings to suggest the instruments were designed in this way.
Another method of analysis makes use of common measure, which is using the local foot measurement as a basis for the instrument design. This can also have its drawbacks, for example a German researcher concluded on the basis of metrology analysis that a clavichord was made in Frankfurt, whereas it was later shown by two other researchers independently to have been made in Naples. Another German researcher provided very convincing evidence to suggest a harpsichord was made in Milan, when it has since been proved to have been made in Rome. Just last week an American humorously informed me that a Stephen Keene spinet he restored wasn't made in England, at least according to his computer.
In fact, both common measure and geometric relationships were probably used in instrument design but it is fanciful to think a maker was over-concerned with such approaches. Modern instrument makers seem to work to far higher tolerances than their seventeenth century counterparts. A maker today may make perhaps four or five keyboard instruments in a year, taking perhaps three or so months on each one. The quality of seventeenth century workmanship suggests that those craftsman made the instruments as quickly as they could, working as artisans rather than the artists that today's makers tend to consider themselves. The old makers show many signs of having made small adjustments as they went along with their instrument. Once the sides of the viol were bent and placed against the form the back and soundboard were made to fit the ribs, so that any deviation from the form didn't matter.
Likewise, in keyboard instruments the makers worked in such a way that later parts added to the instrument compensated for any earlier inaccuracies. It is almost impossible to predict the final shape of a harpsichord or spinet bentside. To make it, the maker either immerses a piece of wood in water or uses steam to soften the fibres and then bends it around a form. When it has dried several days later, it is removed from the form and the wood springs back a little. I have yet to see two instruments by any old maker with identical bentside curves. Once the bentside has sprung back it is stable and the maker can then cut his baseboard to fit the shape of the bentside and make the rest of the case. The major concern of any maker is that his instrument played reliably. The string lengths of an instrument are important, since they determine the pitch it plays at, but makers weren't concerned with millimetre accuracy throughout the instrument. Instead, the maker chose two or three strings and made sure those lengths were accurate, and by using a bridge template was able to ensure the bridge shape meant all of the other strings were close enough. That is why the string lengths of different instruments are seldom the same. By examining the soundboard bridges of instruments it can be seen that makers marked out that bridge to ensure the string spacing was satisfactory, put in the bridge pins, and then put in the pins on the nut having made sure the instrument had room for all of the quills to pluck evenly. No owner is ever going to know that the strings on his spinet may be a few millimetres different from the spinet down the road.
It is important of course that the maker had a basis of design so that the string lengths would be close to what he wanted for the instrument to sound best at a particular pitch. Dominic Gwynn has shown that many organs following the Restoration were designed to be played at "quire pitch" which is about 1 to 2 semitones above modern pitch. I have found that the surviving English virginals have pitches which mirror what he has found in the organs, with over half of the instruments appearing to be designed for this nominal pitch standard. The makers achieved this in a practical sense by knowing that the top c string, two octaves above middle c, needed to be 6 inches long to play at that pitch. If a maker wanted an instrument at a lower pitch he would make that c string perhaps 6 3/4 inches long so it could play a tone lower. English virginal string lengths have very little consistency, and yet always that c string, or else the top string, works out as a single unit of common measure from which the maker can ensure the instrument was built to play at the desired pitch.
In spinets the same applied. Charles Haward used iron strings and had the top c 6 inches long. John Player used brass strings which have a different tensile strength. In his case he made the top c string 5 inches long. Stephen Keene's earlier spinets use iron strings and then he changed to brass. In his instruments of the 1690's he also appears to design his instruments so that the top c string is 5 inches long. It cannot be determined exactly what pitch the makers wanted, but there is enough evidence of string keyboard instruments in general to suggest that brass strings of this length were designed to play at just below a semitone above modern pitch.
Recorders and flutes, on the other hand, can have their pitches determined accurately, and Peter Bressan's instruments seem to play at around A408, which is a little more than a semitone below modern pitch. Neither Dominic Gwynn in his analysis of seventeenth century organs, or my own research on string keyboard instruments have found any instruments which played this low. Likewise, the so called Division Viols by Barak Norman have gut strings which suggest that they should also play best at high pitch standards. I have elsewhere suggested that there must have been some form of standard transposition between instruments, and that the keyboards and viols were designed to play a tone above the woodwind. Since the "quire" pitch of the organs got its name from human voiced choirs, perhaps it is that the standard pitch in Purcell's time was around A450Hz.
In summary, the design of seventeenth century English instruments in Purcell's time was increasingly concerned with a form of consensus and standardisation, with makers aiming to produce a product that was desired by customers and economic to make. The innovation was also there, not so much in form of technological advancement of constructional method, but by making instruments of a greater visual elegance, and by making instrument forms where the construction techniques were simpler and more reliable.
Edinburgh, April 1995.