Building a Renaissance Lute using original methods - a project report by Andrew Atkinson
Originally a series of lectures given at meetings of the Lute Society, and published in Lute News issues 70, 80, 82. and 87.
I attended the London International Early Music Exhibition a few years ago and having seen lutes there, decided I would like to build one. I bought some plans, and was told I would need tools including masking tape, aluminum for a rib template and PVA glue to build a lute. So I started wondering how they would have built a lute in the past, and having long been interested in "bygone" crafts, began to research this question. I had already completed a course in modern fretted instrument making at the London Guildhall University, and applied to do a part-time postgraduate course, with the aim (initially) of reconstructing a stringed-instrument maker"s workshop of around 1590, when Dowland was playing and composing. I hesitate to use the word "authentic" of such a workshop because the more deeply I go into this question the more I realise all one can talk about is possibilities and probabilities as to how things were done. I am coming to this from the point of view of a collector of and enthusiast for historic woodworking tools, and would welcome comment and criticism.
So where do we begin? There are some written sources, and some iconography, though not too much from before the time of the 18th century French encyclopaedists.
Written and pictorial sources
This first picture many readers will have seen before, that is the picture of a lutemaker from Jost Amman"s Book of Trades, dating from the 1560s (now available in a Dover reprint) and showing a lute maker in his workshop, with some of his tools.
This is in fact the only surviving picture of a lutemaker"s workshop as such. The verse below describes bending wood over a form. We can see two planes, two chisels, a mallet, a gluepot, a bench and a rather nice ax and block. A lute back sits on the bench, and a mould (perhaps?) for a bowed instrument hangs on the wall next to a completed lute.
The other early guide is Henri Arnault of Zwolle"s lute design, from his manuscript of c.1450.
David van Edwards has recently discussed this plan, including an explanation of why the shape seems slightly different from those we actually see in late-mediaeval pictures of lutes, in Lute News 69. The manuscript gives some useful information, including the use of a mould, the use of a hot iron, and the use of glue and paper strips to reinforce the back of the lute.
A third useful source is Marin Mersenne"s Harmonie Universelle (1636) which gives information on proportions, describes the use of "wooden nails" to hold the ribs onto a form, and the use of a soundboard-shaped piece of wood to keep the completed back in shape.
A fourth source is Thomas Mace, Musick"s Monument (1676), which gives some information of basic repair tools and methods. Interestingly, more than two centuries after Arnault of Zwolle, he too mentions the use of a hot gluing iron and glued paper strips, suggesting a very good deal of continuity in methods of lute making.
Finally Diderot"s Encyclopedie (1765) has good illustrations of lutherie, but this is rather a late source for our purposes. The written sources then are helpful, but are few and scattered, and in any case I think one can be sure that, as today, different lutemakers and different workshops must have used different methods to achieve the same end, making it harder still to talk with confidence about "authenticity", let alone one definitively authentic type of lutemaker"s workshop. Moreover, each craftsman would have made many of his own woodworking tools, giving rise to further diversity.
Besides these sources specifically pertaining to lutherie, there are many more general sources. Fortunately for our purposes both Joseph, father of Jesus, and Noah (when he had to build the ark) practiced carpentry. I think it is perfectly reasonable to assume that lutemakers would have used the general woodworking tools of their day—as of course lutemakers do today —so we can look to paintings of general carpentry for information. I think also that paintings of carpenters and carpentry tools must be accurate, because painters must have worked closely with woodworkers, for painters did not only paint fine art paintings on canvas or panels in the renaissance. For instance, the artists Francesco and Giacopo Bassano, painted not only paintings, but bedsteads, murals, the town clock, and so on (so I have read) working alongside the woodworkers who made these items, and many painters painted panels on carved wooden altarpieces, so I think we can assume real familiarity with woodworking tools. (Incidentally there is painting of Noah building the ark, by Leonardo Bassano, but too dark to reproduce here).
A very useful source is Hieronymus Wierix"s The Childhood of Christ. Wierix lived from 1550 to 1617, mostly in Antwerp. On the title page, notice the bench, without a vice but with a bench stop, some sort of marking gauge lying on the bench, frame saw, planes, chisels, ruler, and (behind the two-handed saw on the right) a long saw looking rather like a sword.
Title page from Hieromymus Wierix "The Childhood of Jesus" (c. 1600)
In this picture Jesus, guarded by guardian angels, is blowing bubbles, a childlike activity, but also evoking symbolism of the transience of human life, while Mary works on a piece of cloth. Note the rack of chisels on the wall, and the frame, compasses, and a plane on the shelf.
Tools shown in a 16th century intarsia guild chest
Illustration of joiner's tools from Joseph Moxon Mechanics Exercises (1683)
This source includes a small curved item on the bench, a "holdfast" (Top right of the above image) which, fitting into a hole in the bench, was used to hold a piece of wood in place, the British-style coffin-shaped smoothing plane, and the Continental-style planes with shepherd"s-crook handles (this print seems to have copied from a French source). The screw clamp on the side of the bench, and the handsaw, something like a modern one, were just beginning to come in at this period—the late 17th century.
Jost Amman"s Book of Trades shows a joiner"s workshop as well as a lutemaker"s, with a frame saw, some sort of marking gauge, wooden squares, and planes—the planing being done against a bench stop. He also shows carpenters at work outdoors hewing beams and making a timber frame for a building. Amman also has a picture of coopers, and I think this is of interest because a barrel is a little like the back of a lute, made of ribs, curved and beveled.
Amman"s picture of a bookbinder shows the use of a draw knife; also the bookbinders frame, just the same as hand-bookbinders use today, and a tool called the bookbinder"s plough used to trim off the edge of the paper. Throughout, the degree of continuity between crafts, and over time, is striking. His picture of a crossbowmaker (not such a common trade today!) shows a three-legged glue pot, and gouges.
A great many of the tools are immediately recognisable today, but there are interesting differences. For instance the saws are mostly frame saws, or saws slightly curved and shaped like a sword. They did not have the technology 400 years ago to roll wide flat strips of steel needed for the blade of a modern handsaw, so in the frame saw, they used a narrow blade, held in tension, to keep it stiff. (These saws were still popular until quite recently on the Continent, though the wide-bladed handsaw popular in Britain and America has really now taken over.)
Another big difference is in the techniques for holding wood. Until the 18th century woodworkers did not generally use vices for holding wood, instead using bench stops (when a piece of wood is being planed the forward movement of the plane simply holds the wood in place against the bench stop) or a bench dog, or the holdfast, noted above. (Craftsmen often give their tools animal names: shaving horse, saw horse, bench dogs—holding tools especially, which is why I regard myself as in search of the lutemaker"s donkey . . .). However, I have come across an illustration from 1505 showing a very modern looking workbench, with a tail vice and a side vice. Why was the vice not more widely used by woodworkers, for even primitive vices are not thought to have been widely used by carpenters until the mid or late 17th century? I think there are several reasons. Other ways of wood may have been regarded as perfectly adequate; also the productivity of the workshop was important and it is much quicker to simply put a piece of wood against a bench stop if you wish to plane it, than to spend time winding and unwinding a vice. Significantly, Jost Amman does show a vice in his Book of Trades, but in a cutler"s (Messerschmidt) workshop. Metalworkers had the skills and materials to make a metal vice, and more to the point, needed vices, because there are some metalworking operations which cannot really be performed without one. They would have been prepared to go the trouble and expense of making or buying in a vice (and spending time winding and unwinding it). Perhaps the vice was regarded as a smith"s tool; it would be interesting to know if there were any guild regulations concerning the ownership and use of vices.
Lists of tools
The late mediaeval poem "The debate of the Carpenter's tools" provides us with a comprehensive list of tools available to the late fifteenth century woodworker when the tools debate among themselves why their master brings home so little money for his wife, concluding that he spends too much in the alehouse. More prosaic, and more immediately useful for this project is an inventory of tools left by an Essex joiner, Cornelius Eversen, at his death in 1592, reprinted in Newsletter of the Tool and Trades History Society 47, pp. 41–43). This includes:
Imprimis, Joiners plane of divers woods 15
Item jointers 2
Smoothing planes 1
Mitre squares 1
Frame Saws 1
One brace and five bits for the same
Two broad pairing chisels
Three mortise chisels
Three small Flemish chisels
Three ripping chisels
One line rowle with the line upon it
Two staples or bank hooks
Two rules of two foot apiece
Two spare planing irons
If a joiner had these tools, then a lutemaker could certainly have had them. A 1575 Spnaish inventory of a violero"s workshop tools is reproduced in Lute News 71, p. 12.
A few tools survive from my chosen period and indeed from much earlier; some Roman planes survive, as do Viking ones, and they seem remarkably familiar to the modern carpenter in their design. Some of the Viking tools have been shown to be steel-tipped, for instance iron plane blades with steel edges forged on; plane blades were still being made like this in the 19th century, which helps to justify my use of Victorian blades for my reconstructed tools.
Nautical disasters have preserved useful examples of preserved early tools, for instance planes and a carpenter"s rule survive from the Mary Rose, sunk in 1545 and partly recovered from the bed of the Solent in 1982.
Jointer plane from Mary Rose
Rule from Mary Rose
It is interesting to note that that the Mary Rose rule has an eighth of an inch as its smallest division. This gives an insight into the attitude to measurement in an age when every artefact was custom made: parts would be made to fit each other, not made to precise, predetermined sizes as is essential for modern batch or mass production.
A slightly less well-known disaster than the loss of the Mary Rose is the failure of William Barents" expedition to find the North East Passage to the Indies, by trying to sail round the north coast of Russia. The expedition got caught in the ice in 1596, and the explorers were forced to build a shelter and spend a winter in Novaya Zemlya. In the spring, they left their camp, only to find their ship had been crushed by the pack ice, and so they had to build a new boat for their return journey. They could not carry very much in this boat, and so had to abandon many of their tools. Their cabin, still standing, and containing tools, clothing and other items, was rediscovered in the 19th century, and these historic artefacts are now in the Netherlands, in the Rijksmuseum.
Tools recovered from Barents" camp in Novaya Zemlya, now in the Rijksmuseum, include a brace and bit, plane, plane blade or iron.
The date of 1596 is of course just the right date for my planned workshop (though I changed my plans later, as explained below), and I have based one of my planes on the Novaya Zemlya discovery. Some chisels found at Novaya Zemlya seem rather modern in design.
I have also seen photographs, in Gerrit van der Sterre's Four Centuries of Dutch Planes and Planemakers (2001), of a Dutch carpenter"s toolbox from a shipwreck, I think now conserved in Holland.
The European tradition of making lutes (as such) is a broken one, but some near Eastern traditions of making lute-family instruments are not. L. Pickens, Folk Musical Instruments of Turkey (1975) has useful photographs of a Turkish saz maker at work, bending ribs with a bending iron shaped like a trowel, which he presses down on a rib lying on his bench, chiselling bent ribs to size, then planing them like a cooper (that is, on an "upside down" plane which held on the bench, with the blade projecting slightly from it; the maker holds the wood to be planed and moves it over the plane, rather than moving the plane over the wood), and fitting ribs over a ribbed body mould.
Also, much can be learned from own rural woodcraft traditions. For example, the well known maker of windsor chairs Jack Goodchild had his workshop at Naphill, near High Wycombe, Bedfordshire, and photographs of his workshop in the 1950s have been published in books on handcrafts.
I believe it is documented that the great lute making workshops of the 16th century bought in large stocks of parts, such as ribs and wood for soundboards, that had already been roughly cut out near where the trees were felled. This was a matter of economics; transport was expensive and it made no sense to transport a lot of wood that would end up as waste over long distances. Cutting out and roughly shaping these parts was probably a job for rural people who had little employment in the winter. A lot of work was done in the woods, an energy efficient practicethat continued until recent times and is now being revived as fuel and transport costs rise.
(Pictures from the unbroken traditions section can be found in Lute News, backissue 70, 2004 July.)
Reconstructing lutemaking tools
In all this, I am fascinated by the idea of using seemingly simple tools to gain very precise results, producing objects of the greatest beauty and refinement. Woodworkers today tend to have a large number of very specialised tools in their workshops; I suspect that lute makers in those days would have had fewer tools than modern makers; and that skilled workers could use, say, an axe to get very precise results. See for instance this sculptor using a large axe for what one would imagine to be fine work.
The unfinished man, woodcut of 1533 by Erhard Schoen
(I think this humorous woodcut shows a mother presenting her wayward, drunken son to the sculptor, asking him to make him into a complete man, just as he is completing the wooden statue. Note the collection of tools at the sculptors feet, depicted with some care; the woodcut maker was himself a woodworker, of course). From modern times, I have seen a photograph of around 1950 showing an English craftsman making a malt shovel out of a large piece of beech, using a special axe to hollow the thin delicate shovel blade.
Finally, I have spoken with lutemakers practising today, and have come up with a preliminary list of tools for the lute maker"s workshop of the 1590s, as follows: smoothing plane; jack plane/trying plane; mallet; knives of various types and sizes; chisels; chopping block; glue pot and brush; axe/hatchet; straight edges; rule; dividers or compasses, large and small; squares for marking, setting out and testing; hammer; stamp/punch for decorating the lute bridge; bench; name-marking branding iron; gluing irons; saws of various types; scrapers; sharpening/whet stone; various holding and clamping devices; and setting-out board/table.
While some of these tools—chisels, hammers, mallets—were surely used then exactly as they are now, others were not, and require practical experiments to help to recreate lost methods. For instance, how were glue pots heated? It seems likely that wood shavings were often used, as these seem to have been a common cause of accidental fires. How were very small holes made? A bow drill might be a likely possibility and is relatively easy to make, in its simplest form. A gimlet could be used for larger holes? How was the neck of the lute fastened to the neck block? Modern lute makers invariably use screws but I am fascinated by the single big nail which we know all the historical makers used, clearly seen on X-rays of old lutes. You can buy hand-forged nails, with square shanks and large heads, these look the part but are made from mild steel not wrought iron, so it will be a matter of commissioning a smith to make some from this more historically appropriate material. A hole has to be drilled in the neck first, and I have read that there was a practice of heating the nail before driving it in. Of course there is the question of how to hold the neck in place during these operations. The sheer number of lutes and unassembled or partly assembled instruments attested in the inventories of the great 16th century lute makers workshops (or factories, perhaps we should call them) suggests that luthiers may have employed substantial workforces, and division of labour, so "authentic" lutemaking practices might reflect this.
I have built copies of tools from the various historical sources discussed above. Inevitably there are compromises; I am using modern metal for the metal parts (including 19th century blades in some cases), whereas metal may have been of variable quality in renaissance times. My drill bit, of Victorian date, has a square shank; a more flattened shank, to wedge into the brace and bit, would be more authentic and would fit more firmly in the brace /bitstock I have made, described below.
Tools I made at an early stage in the project include the following:.
Copy of Dutch smoothing plane based on the 1596 Barents" expedition find.
Copy of Dutch "Schaaf-Type" plane of 1618, illustrated in W. L. Goodman, The History of Woodworking Tools (London, 1964), p. 81, fig. 83. It was hard to make, and tends to "chatter" in use, leaving ridges on the surface of the wood. (I have not used this plane because of its poor performance, preferring the barents plane which worked well straight away, which indicates that the best preserved historical tools and lutes may have been used little and survived in unworn condition because they were no good?)
Wierix/Mary Rose jack plane and trying plane I based the lengths, iron/blade widths and blade angles on the planes found on the Mary Rose. These are a little too early for the purposes of my project, so I used the Wierix engravings to arrive at other details such as the handles, and the body shape was based on pictures of Dutch planes of the 17th century. Both planes were made deeper than the Mary Rose designs as I presume that these planes were not new when the ship went down and planes lost depth over time as their soles are "trued up" by themselves being planed. These planes both have single irons, as the "double iron" did not appear until the early 18th century. The irons used are old, probably 19th century examples, which I believe will function in an identical manner to those of the 1590s.
Frame saws I have based mine on the several examples in the Wierix engravings. I believe my use of modern blades is acceptable, because even though a blade of 1590–1600 would have been hand made, the smiths of those times would have been capable of producing a very accurate and serviceable item. That this is so is shown by the many examples of metalworking of an unbelievably high standard preserved in museums. For the curved parts of frames I have selected wood which has a natural curve in the grain, as this is stronger than cutting curves out of straight-grained material.
Squares, straightedges and rules Squares and straightedges are seen in more than one of Wierix engravings, and I have based my square on these, and on illustration in Goodman (1964) and S. Landis The Workbench Book (London: Collins, 1989). For my rule/straightedge I have taken the designs of the carpenter"s rule from the Mary Rose, described in Richard Knight in Tools and Trades (vol. 6, pp. 43–55).
Rule and dividers
Brace/bitstock I have copied this from the most complete example of the three left on Novaya Zemlya in 1596. The brace works reasonably well and I have made it so that I can change the bit sizes by using interchangeable wooden pads.
Mallet I have made a mallet like the ones illustrated in several of the scenes in the Wierix engravings. I made it as I like the pleasing cylindrical head, curved along its length, and its archaic appearance.
I experimented with forging different shapes of bending iron, and metal holdfasts, amongst other things.
Some of the old tool designs seem to have hidden advantages; such discoveries are one of the things that makes experimental archaeology so worthwhile. For instance, while a wooden square may go out of true if you dropped it on the floor, it is light in weight, and would be unlikely to accidentally mark the wood you are working on. I read that these were made from old casks, so I visited Young"s Brewery(!) where a kind employee gave me some old barrel staves. I think the reason they used this wood was that it was of good quality, and cut in a certain direction (quarter sawn, or rather split) for greater dimensional stability. Planes without handles, clearly shown in pictures, while a little harder to hold than those with handles, are easier to lay upside down on the bench when planing lute ribs (moving the wood over the plane rather than the plane over the wood). Old chisels tend to be thinner than modern ones, perhaps because steel was so expensive, but this makes them easier to sharpen, as there is less metal to take away. One more thing I have found is that making you own tools teaches self-reliance. Perhaps all wood and metalworking students should be required to make some of their own tools!
Before launching in on my blow-by-blow description of my lute building project, perhaps I should reiterate a philosophical point, common to all experimental archaeology. We have a mixture of certainties, probabilities and possibilities. Where we are not absolutely sure what the lutemakers of old did, all we can establish with certainty is that such and such a method, using the technologies and materials that we know they possessed, could have been used, or may have been used, to produce the historical instruments that we see.
The earliest surviving lutes date from the early 16th century; a little older is the clavicytherium in the Royal College of Music. Of south German origin, from c.1480 it is the earliest known surviving stringed keyboard instrument.
Clavicytherium (c. 1480)
There are often clues to making practices in the form of scribed setting out lines and tool marks in interiors or parts of instruments usually hidden from view. Another very useful source on historical instruments are paintings and representations in sculpture and intarsia, especially because they can show us the designs of instruments at a specific date and location. One painting further to those mentioned above is of the building of Noah"s Ark, of 1423, in the Bedford Book of Hours, which shows the use of wedges as a holding device, rather as modern joiners use wedges and a notched piece of wood in the process of planing a door. From my own workbench, this picture shows how simple dowel pegs fitting into holes in the bench, and a wedge, can be used to hold wood in place for planing.
Other interesting pictures are in Agricola"s De Re Metallica (1556) showing the use of dividers in dividing angles and workmen following templates; and this drawing of a workbench, from Nuremberg in 1505 including a representation of an early vice, not generally in use for woodwork until much later.
Initially I aimed to make an Elizabethan lute from the age of Dowland, but later moved the project backwards to the 15th century. This is because of the existence of a uniquely detailed, illustrated manuscript from this time, describing the design and making of a lute, the manuscript of Henri Anaut of Zwolle of c.1450 (above). I decided to follow Ian Harwood"s translation (Lute Society Journal, 1960) as best I can, trying hard to resist the temptation to introduce too many "improvements" of my own. I am aware that David van Edwards has suggested that Arnaut misunderstood some of the practices of his own time (see Lute News 70) but decided on balance to follow Arnaut as closely as I could.
I have mentioned that woodworkers seem not to have used screw vices so I have based my bench top on this illustration from the Nuremburg Hausbuch, showing a joiner working at bench with a three-peg system and other pegs to use as stops to hold wood for planing.
This bench with three pegs and wedges (as used by chair makers and modern green-wood workers or bodgers) is very easy to use especially for uneven, tapered instrument parts. The photographs above show how this system can be used to hold a piece of wood vertically; here the pegs are used as shooting board end stops—exactly as in the picture of the 15th century joiner above, with the layout of the pegs in the same pattern.
Clamping with pegs and wedges
The following photographs show the successive stages in my attempt to reconstruct the methods of a late-medieval instrument maker as described by Arnault over 550 years ago. Note the simplicity of the technology, and the absence of working drawings.
I followed all of Arnault"s instructions the best I could and added things of my own where necessary ( if not mentioned in Arnault's manuscript) and using materials and technology readily available and appropriate to the time period.
Here is the wood for the lute mould base being rub jointed.
Here the lute mould base is being set out following Anault"s instructions, showing all the tools needed: compasses, scriber and straightedge for a design based on a fixed set of proportions, the measurements derived from a chosen string length. Note the stick at the rear, not mentioned by Arnaut, but on this can be "stored" all of the main measurements or proportions I derived for setting out a lute of my chosen string length. Something like this would be easy for a jouneyman luthier to carry away when his apprenticeship had ended, perhaps. It is interesting to note that Arnaut"s design for a clavichord, likewise, derives all its dimensions from a single initial measurement, and proportions based on it.
Lute mould base
Cutting the channels in the mould base board for the vertical "bulkheads" to sit in.
Again this demonstrates the multi-usefulness of the pegs to hold work against. I used a bow saw for this, a knife and a chisel.
Planing the "bulkhead" wood to the correct thickness to fit in the channels in the base (to gauged lines).
Marking the semicircles on the "bulkheads", simply sticking the wood for the bulkhead in the channel in the bass, and using the dividers to mark the bulkhead, with the mould base dictating the diameters of the bulkhead circles.
Sawing the bulkheads to shape.
Fitting the "bulkheads" to the base board and chiselling their tapered profile. I marked the angles on them but cut the facets on later after they were glued in place.
Gluing the bulkheads in their channels—please forgive the use of a modern aluminum pan! Note the small "square" used to ensure they are perpendicular to the mould base.
Preliminary shaping of lime end blocks with a side axe; Arnaut"s design has a block at each end. This axe is resembled the one shown in the Jost Amman "Lautenmacher" woodcut.
This axe was purchased on "eBay" and is from Transylvania.
Planing the end blocks against bench pegs so that they will have a good 90 degree angle with the mould base board. The block can be held in place between pegs and wedge in spite of its irregular shape with the aid of scrap wood.
Peg clamp for planing
Using an awl to make a nail hole in the completed mould base. This is to hold the end blocks onto the mould base temporarily, while the body is being assembled.
Ready to temporarily nail the rough-shaped end block to the mould; note the angle of the nail to pull the block against the bulkhead.
Nailing the end blocks to the mould.
Chiselling the blocks to profile and with facets so that the ends of the ribs contact properly when glued.
I made simple wedge clamps (using appropriate technology and materials) to clamp the thin rib wood to the bench top so it can be planed to thickness. Arnaut does not give any clues as to how this would have been done, but the Bedford Book of Hours shows wedges in use, so inspired by this I made a simple clamp to hold the ribs in place for planing, and these worked very well.
Here is the clamp in use for rib thicknessing.
Here is another (slightly more complicated) possibility: a planing board for holding thin wood while planing it.
There is little point in thicknessing strips of wood much wider than you will require for the ribs, so here I am gauging rib wood to width before thicknessing, using a marking gauge.
The surplus width can be broken off along the line scribed by the point of the guage. The ribs are then planed to the correct thickness, being held in place with a simple clamp, as shown above.
How does one gauge the thickness of the ribs to make them consistent? I used a piece of scrap wood of the correct thickness (I cheated a bit here: I had measured the wood, it was 1.3 mm in thickness) and I held it against the ribs to feel when the correct thickness was achieved. This method seems to me to work well and is quick as the wood doesn"t need to be removed from the clamp on the bench as is necessary with modern dial calipers. (At this point in my talk, I was gratified to be informed by Stephen Gottleib that he had been using this same method for the last thirty or so years and it had served him well—thus demonstrating that if a method is simple and it works then other people will have thought of it, and it is very likely to have been used in the past.)
A toothed iron in the plane is needed to avoid tearing up the grain of the more figured wood. There is a Roman toothed plane in existence, so this is an ancient technology. The 'furrowed' surface left by the toothed blade is then scraped smooth. One of the 16th century inventories mentions sword tips in use as scrapers.
Here I am heating the bending iron to bend the ribs, obviously compromising on the use of 16th century technology(!); instrument makers would have had some sort of small brazier in their workshop, also used for heating other things such as the glue pot.
Heating the iron
Arnaut of Zwolle mentions the use of an iron; this iron is based upon the Turkish insrument maker"s bending iron illustrated in Laurence Picken"s Folk Musical instruments of Turkey; reproduced in Lute News 70, p. 11.
Here is the iron in use (in my kitchen, as you can see!)
Using compasses to transfer the rib widths from the mould to the bent rib.
A rib template could have been used here, along with a scriber to mark the outline on the rib. (I wonder if charcoal, chalk or a lead stylus could have been used for marking on wood before the days of the carpenter"s pencil.) Using this method, consistency somewhat depends on the accuracy and consistency of the mould.
Outlining the rib
Rough shaping the rib with a knife, a chisel
Then more precisely, with a plane. The rib is compared often with the mould. (Many of these pictures show me working one-handed simply because I have the camera in the other hand . . .)
Waxing the mould to avoid the ribs sticking to it when gluing the ribs together.
The centre rib is ready for gluing, with the small nails, hammer etc. ready to hold it on the end blocks while the glue sets. I decided to use nails because I have read that some old lutes have small holes under the end clasp. The end of the rib is pre-bored with an awl to avoid splitting. A couple of nails were put in two of the bulkheads along the centre rib"s edge to help make sure it wouldn"t slip out of its proper position.
Tools ready to place centre rib
The centre rib glued in place.
Fitting the next rib.
Removing small nails when glue has set. For all of the other ribs only one nail in each end block was needed.
Using a small knife to make sure the ribs are not glued to the mould; this is done after gluing each rib and allows easy removal of the completed body from the mould.
For Anault"s "rather warm iron" used to glue each rib to its neighbour I based my design of gluing iron on Thomas Mace"s description in his Musick"s Monument. Here I am using the iron and glued paper strips in the way a modern maker uses masking or sellotape, to hold the glued ribs together. The hot iron sets the glue so the paper strips can be used to "clamp" the springy ribs together while the glue sets.
Setting paper strips with hot iron
For some reason at first I used short glued strips of paper crossways as described by Thomas Mace. This worked, but probably created more unnecessary mess and glue to be cleaned up and was quite "fiddly" to do, picking up and gluing the small pieces. Next I used Arnault"s long strips of paper and found that these worked better, tending to help keep the rib edges in line along the whole length and allowing me to work my way along the rib pushing and pulling the joints together (even where the fitting was not as precise as it should have been!). I found that it was easier to brush plenty of glue (I think I was probably over-generous) on the rib for a few inches, the glue going naturally between the ribs, and applying the paper strip to the glue on the rib rather than onto the paper creating a sticky (and weaker, because a bit soggy) ungainly paper strip, doing three or four inches at a time and holding the joints together while using the iron to set the joint in place. I warmed the ribs and end blocks to help avoid chilling the glue and in one instance scorched the edge of the last rib fixed—so take care! I was also a bit over-zealous with the rather-too-warm iron and scorched my ribs in many places as I "set" the glued paper strips. I expect more practice will help me here, but on the whole I was very pleased with how well it worked, and I didn"t scorch my fingers, luckily.
Here I am marking the end of the next bent rib ready for fitting.
Here I am applying glue to long paper strips, then using the gluing iron. To complete fixing the rib the two rib ends were secured with the small nails in the end blocks, though the glueing iron does such a good job that this is barely necessary, as it fixes the end of the rib to the end block; but it carries the risk of scorching the wood.
Gluing the paper
Using the gluing iron
Removing the nails from the completed body.
Freeing off the end blocks with a thin bladed knife.
At last taking the body off the mould. You can see once again the nail which has held the end block in place.
Body free from mould
Success! The body seems satisfactory apart from a couple of gaps between the end blocks and the last ribs, possibly due to imperfect end-block shaping or excessive forcing of the ribs into place. The next stage is to put the paper strips over the rib joints on the inside.
Cutting the interior paper strips with a knife and straight edge.
Cleaning glue from the inside of the body with a small chisel.
Gluing on the paper strips.
I carefully used a damp cloth to soften the paper and glue before scraping. I discovered here that I had been a bit over-generous with the glue and over-zealous with the hot iron, causing scorch marks; I am worried about scraping the body too much if I am to remove them.
Cleaning off the paper from the outside of the body.
Here I am starting to make the end clasp.
I used a trowel type of bending iron heated on the stove to bend the end clasp strip to shape.
Here is am using a knife to roughly trim the end clasp to a line previously scribed upon it by "pegging" it on the end of the lute body (with pegs that you can see in one of the pictures below) and drawing around it. As the strip is quite wide it is hard to get it pressed flush against the body; there is a certain amount of judgement required in marking the line with the scriber. I should perhaps have smoothed end of the body down first.
Trimming the end clasp
Here I am using compasses to find and mark the midpoint of the end clasp so it can be aligned with a centre mark on the body during gluing; I want to make it symmetrical!
Using wooden peg type clamps (these are like modern violin bass bar clamps), of which I made quite a few, following the suggestion of a lute maker after one of my talks at a Lute society meeting, when glueing the end clasp.
Wedges are used to apply pressure, and they work surprisingly well. The pegs and wedges are easily made from scrap wood. beech and sycamore—I like the idea of making something useful from scrap wood. A range of sizes allows most clamping jobs to be done.
The end clasp after gluing was rather "lumpy" and uneven (I had applied too much pressure in an effort to force it onto the body, as I had mistakenly not smoothed this area out first; the thin wood is quite flexible especially when coated with warm glue) so I decided to smooth out the lumps in the glued end clasp by using my bending iron as a smoothing iron to soften the glue and iron out the wrinkles. This worked quite well.
Smoothing the end clasp
Now it was time to make the neck. I split a piece of sycamore with an axe and a club. I used the same sycamore as for other parts of the instrument; I could have used beech.
Start of the neck
Then I rough planed the neck after splitting and trimming with an axe. I am not using a vice, but simply planing up against a bench stop.
Planing the neck
I have squared off the end of the neck; now I am using a marking gauge to mark the depth of the neck on the body end. I derived the neck proportions as much as possible from Arnaut"s drawing and then from measurements of the George Gerle lute c.1580 given in Robert Lundberg"s book on lute making. I have used this book frequently in this way to fill in the gaps in Arnault"s description and as my main source, suitably modified or perhaps I should I say "retrofied" to suit my time period, on details of lute construction
Marking the neck
Using the same setting on the marking gauge to mark the depth of neck on the body.
Here I am using compasses to mark the ends on the tapered neck, perpendicular to a previously marked centre line. (A square could not be used as I had already planed a rough taper on the neck. An adjustable bevel could be used but I had not yet made a suitably "retrostyled" example).
Marking the taper on neck
Here I am using a chisel to trim the neck joint surface after sawing (I had not yet made the "notch and wedge" vice on my bench—it appears later, in my pegbox-making pictures—so I had to hold the neck in my hand whilst working on it, meaning I was unable to use a plane on this joint surface).
Trimming neck joint
Using compasses to transfer neck joint width measurements onto the body.
Starting to chisel the joint surface on the body.
Here is the body after chiselling the neck joint. A tiny hole on the right is not a woodworm hole, but the result of a small slightly misplaced nail, used to hold the ribs in place while the glue was drying.
Body after chiselling
Now it was time to try the fit of the neck joint. Here is a side view of the neck being offered up to the body. Because of the proportions I decided upon for my neck—relatively deep and narrow—my joint surface has ended up sloping slightly "backwards" the opposite way to that usually seen.
Here I have refined the joint to an almost finished fit. I have chiselled the end roughly to its finished cross section as the corners of the squared up neck made the fit of the joint edges difficult to see.
Refined neck joint
Next I began work on the pegbox. Here I marking out the peg box on piece of sycamore previously split from a small log and then planed ready for use.
Marking out peg box
But how to hold firmly pieces of wood with a taper? Here is the aforementioned bench notch with its accompanying wedge (made in some willow I had spare) and the peg box "blank". I have made an attachment for my bench (screwed in place and held in the relatively modern vice) as an experimental bench top. This bench notch is shown in a number of historical illustrations, and a Roman bench top exists (See William Goodman"s History of Woodworking Tools) with this feature. If I had made this before I made the neck I could have used it to hold the neck, and planed it, instead of smoothing and squaring it off with a chisel, as shown above.
Here is the notch in use for the first time.
In fact the wood tended to move easily when wedged into the notch so I chiselled a slight undercut on the end faces and this solved the problem (another example of the value of practical historical reconstruction!) I was proposing to saw the back off the peg box using my large-frame ripsaw. Here I am rip-sawing the back off the peg box. Note the use of wedges to tension the rip saw; the idea for the corner wedges was taken from an old veneer saw in Mercer"s Ancient Carpenters" Tools. (Normally these ripping saws are bigger, two-man saws. With an operator guiding the saw each side any deviation from the marked line can is noticed and immediately corrected. As I was working on my own to keep to the line I would do 20 strokes from one side, and then move the saw around to the opposite sideand do 20 from the other and so on.)
I had decided on a construction method that would combine the minimum of gluing and clamping with ease of construction; I would saw the back off and saw off one side only, to allow sawing and chiseling out of the slot; then I could plane the back and side and glue them back on. Here I am planing the back of the peg box after sawing it from the peg box blank.
Planing peg box
Sawing the side off the peg box.
Transferring peg box slot measurements from a drawing I have made, pricking through the paper with a scriber, onto the remaining part of the peg box.
Sawing down into the peg box slot to allow easy chiselling out of unwanted wood.
Here I am chiselling out the peg box slot.
Here are the three parts of the peg box ready for gluing back together.
The next picture shows gluing on the side of the peg box, making use of a rather handsome stone as a clamping weight; a hearth stone, I believe, that I salvaged from a heap of rubble produced when Gateshead Council were digging in Pipewellgate to make a platform for a sculpture.
Next, I am gluing the back on the peg box after it had been planed flat. Pegs and wedges were found to be more effective than the stone this time! As ever, I use no screw clamps, but use devices which would have been very easy for the lutemaker himself to make.
Gluing to the peg box
After I planed the sides, and here I am finishing the peg box with a chisel. It has been left over-length to be cut later when joining to the neck.
Finishing peg box
Now I decided it was time to tidy up the end clasp. Here I am using compasses to make marks so the end clasp can be cut to an even length to both sides of a centre mark on the end block before cutting the familiar decorative ends to the end clasp. (Lundberg mentions that historical instruments often have marks on their ribs indicating that this feature was cut after the end clasp was fitted.)
Marking end clasp
I chiselled neat angled ends to the end clasps, and then made a small template from a piece of scrap rib; this was scribed around a few times to create a deeply cut line to guide the chisel
Clasp end template
Chiselling the end clasp decoration.
Now it was time to start work on the soundboard. Most frustratingly, I do not currently have access to my best stocks of soundboard wood, and the only pieces of spruce I had were narrow and of poor and uneven quality so I had to join several pieces. Here I am rubbing jointing wood for the soundboard. I like a challenge!
Feeling wood quality
The jointed soundboard or belly
Scribing around the body on the soundboard after planing and scraping the outside face, leaving a small allowance of half an inch all round.
Using a knife to cut the outline on both sides then carefully breaking off the waste wood; this is done by simply deepening the line cut by the scriber. (There is no point in spending time and energy in thicknessing the waste wood. Also my thicknessing method will be more accurate if I am planing and comparing as near as possible to the finished outline)
Breaking free spare wood
As when thicknessing the ribs, I used a piece of wood of the correct thickness and compare often with the soundboard while planing, simply by feeling with the fingers. One can plane the whole soundboard to fairly near the desired thickness, though you have to be careful as you get towards the middle, as this method offers no way of feeling the thickness of the wood close towards its centre. Frequent testing for flatness of the surface with a straight edge can help here. However, as the final thickness is nearly achieved, if a relatively long plane is used and its blade is set to take very fine shavings it will in theory only allow the middle to be made one shavings thickness thinner than the edges.
Planing the soundboard
This is followed by scraping
I had wondered how I could gauge and control the thickness of the rose area when thinning it down, to around a millimetre. I thought I could put some veneer, paper or card of the correct thickness below this area (enough to raise this area by the thickness that I needed to remove). I hoped that this selective thinning could then be achieved by planing this raised area flat. I then thought that I could use a piece of cloth as this would not mark the already smoothed and soft outer face of the belly as veneer or cardboard might do. Here I am measuring the thickness of two layers of cloth taken from and old worn tee shirt (about 0.75 mm; and I wanted to reduce the centre of the soundboard from about 1.7 mm to about 1 mm). I used the bits of wood on either side so as to avoid compressing the cloth with the calipers. I know the calipers are not appropriate but I wanted to save time and avoid experimenting with cloth and its effects. I now know that two layers of worn tee shirt seems to do the trick. Such a trick could have been found out empirically.
Cloth and calipers
The two pieces of worn cotton cloth, torn to size, are carefully placed, avoiding wrinkles, under the soundboard in the rose area.
Now I planed the raised area flat. When planing a definite area the shape of the cloth emerged. I had expected the edges to be blurred as the wood bent up to accommodate the thickness but the edges and the effect of the cloth were quite "sharp".
Thinning rose area
I used a candle (inside a jam jar to help avoid setting the workshop on fire!) to assess the evenness of the thinning. Candling the belly, the square shape of the cloth can be plainly seen! The belly was then scraped to even out the gradations in thickness this was checked by repeated "candling".
Candle behind soundboard
Now it was time to prepare to start cutting the rose. I derived the design from a painting by Hans Memling in Robert Lundberg"s book, much simpified. I drew a pattern on handmade paper, which is stronger than modern paper (I practised cutting this on a very coarse grained piece of spruce and tried various knives and experimented unfruitfully with drilling out some waste) Unfortunately I used slightly water soluble ink for the drawing, which caused some problems later.
I glued the rose pattern to the inside of the belly. I used the scriber point to help find the location precisely by pricking it through the paper and engaging the point into a mark made previously with the scriber in the middle of the soundboard. I rubbed it over with some kitchen towel to flatten it down.
Preparing rose template
I did not want to use a scalpel for cutting the rose, but an old (sharpenable) knife instead. A most important feature of the knife is how thin its blade can be made (like a modern disposable scalpel). If the end of the blade is shaped like a blunt wedge, there is a danger that the physical width of the blade will cause the thin, short-grained parts of the rose to break as you try to cut them. I thinned the blade on a whetstone but probably not enough. The steel has to be good, to allow it to be made very thin whilst retaining sufficient strength.
Rough cut of rose
I sized the cut rose with thin glue to reinforce two or three cracked pieces. This was a mistake: unfortunately the moisture in the glue caused the wood to swell and buckle, opening the cracks. I used a large old tailor"s iron which I luckily had nearby, to dry it out and flatten it at the same time! This worked well, drying out, flattening and consolidating it, but unfortunately the water in the glue size caused the ink from my drawn pattern to bleed through around the edges on the front of the rose . . .
Using a small chisel to carve bevels on the rose design. I had tried cutting V-grooves on the surface of my practice roses and decided to save this difficult process for a future occasion.
Now it was time to prepare to fasten the neck to the neck block. Here are wrought iron (not mild steel) nails. These nails were kindly supplied by Chris Topp of the Real Wrought Iron company, who has done work for Westminster Abbey, and other prestigious clients. They are a little large for my small lute (the sketch I sent him cannot have been accurate enough) but I wanted to use wrought iron nails as it is historically appropriate and it apparently does not rust as readily as modern mild steel.
Here I am boring the pilot hole for the nail in the neck block, using a brace and bit (a Victorian shell bit in fact)—rather daunting as the hole has to be made very accurately. It is difficult to keep the bit horizontal so I used my pegbox as a visual guide placing it close under the bit. Using the pegbox for this purpose made me realise that a wedge-shaped piece of wood was very suitable for this purpose as the height can be varied easily, to suit the hole and the diameter of the bit.
Boring pilot neck hole
Having bored the pilot hole in the neck block I poked the shell bit through from the inside and placed the neck and body together but offset so I could check the angle of the hole was correct and possibly modify it if needed. It was OK. I had aimed to place the nail near the back of the neck as a a luthier told me it would form a stronger joint when glued if the nail lies towards the back of the neck.
Checking angle of hole
I then used the hole in the neck block as a guide for boring the nail hole in the neck. My wooden bitstock or brace would not fit inside the small lute body so I made a rough wooden handle to use with the bit inside the body like a gimlet. I held the body firmly on the neck (which was wedged in my notch vice) with my left hand, keeping two marks lined up whilst boring the hole with the right hand. A little moisture on the joint surfaces helps prevent them from sliding about during this work.
Boring the hole for the iron nail
Once the hole is established in the neck, it can be continued without the body as a guide, frequently using the body to check the hole has not wandered.
As the nails are tapered in section, once I hade bored a small hole to accommodate the length of the nail, a larger bit was used to widen the hole to fit the wider upper part of the nail. To achieve a more accurately shaped hole I held the nail in a pair of pincers and as it is forged to a square taper, with fairly sharp corners, it could be used to ream its own hole. It worked quite well but I plan to make a reamer in future to make this process easier.
Continuing boring the hole
Reaming the hole with the nail
Reaming the neck
I checked the depth of the hole by eye, boring it to a depth which I hope will provide a strong joint without risking splitting the neck. I lined up the joint and tapped the nail nearly "home" and all seemed well. To ensure I insert the nail in the same position when I glue the neck I filed a small mark on the nail head and made a corresponding mark nearby on the neck block.
Finished making the nail hole
Pulling the nail out
Now it was time to start work on the barring. Arnaut"s design shows only three bars. I split the bars for the belly from a piece of spruce as I did not have any usable offcuts left over from the making the belly
Splitting spruce for the barring
Here I am planing the bars straight, to height and slightly oversize, to markings made with the marking gauge. (I realised later after gluing that I had made the bars very high. This was because the instruments I had derived these dimensions from were much bigger than mine. So I reduced the heights of the bars to be more in proportion to my small lute).
Planing the bars
Here I am gluing the first bar, once again using my wooden clamps to hold them when the glue dried. I used a chisel to remove a little paper off the inside of the rose, where the bars were to go.
Clamping the first bar
Here I am making then trimming the small rose reinforcing bars, using the chisel to trim them. I cut these from a piece of scrap from the soundboard, easily cut with a knife.
Cutting rose bars
I tried using a small hot iron to set the glue on the first couple of rose bars and it didn"t work too well; I burnt my fingers, and found that one end of the bars would rise up as you glued the other down so I just I clamped the rest using my old iron (not heated) as a clamping weight and it worked well. ( I was later informed by a luthier that if I had thined down the ends of these bars before gluing, then the hot iron would have worked better as the heat would have penetrated the thinnner wood)
Using a small iron to fix rose bars
Weight to hold the bars down
I then glued on another main bar. I am placed the belly in the airing cupboard. I had stored it there for a couple of days before gluing the bars it has low humidity.
Here the third and final brace bar is being added glued to the inside of the soundboard, again using clamps rather like traditional violin bass bar clamps. Once again avoided the use of non essential metal tools, such as screw clamps, since these seem to have been little used, apart from the blades, drills, hammers etc, by woodworkers in the early modern period.
The third final bar is glued and clamped
To make the bridge; here is the piece of piece of apple wood (I had no pear unfortunately) I split the bridge from.
Splitting out a piece of apple wood using an axe and a simple club.
Sawing off potentially cracked ends of the apple wood, to cut it to length.
Here I am trimming the bridge further with a side axe. I believe that the craftsmen of old used wood economically, and I try to follow this spirit, seeing if I can get enough wood for two bridges out of this block.
Trimming bridge blank
To position the bridge correctly I transferred measurements from my mould onto the soundboard, using compasses and a straight edge. Throughout the project I avoided the use of modern rulers for precise measurements. Arnaut talks of the bridge position in terms of the proportions of the body.
Measuring position for bridge
Lightly scribing a line on the soundboard to mark the position of the bridge.
I was anxious to avoid the bridge slipping out of position when I was gluing it, alas, this did happen in the event, so that the string length is slightly longer on treble side. (This is a feature of many original lutes; see David van Edwards"s paper in The Lute, 1985 part 1, and more recently, Chris Coakley, "Tapered lute strings, angled necks and bridges" in FoMRHI Quarterly 109, August 2008, pp. 35-43, where he suggests that this was done on purpose to counter the tuning problems of lute strings which were fatter at one end than the other though this is highly debatable) A lute bridge is quite a complicated shape for a small component and is thus difficult to hold while it is being worked upon. I planed the angled surfaces and trimmed it against a bench stop.
Planing bridge blank
I also made a special jig to hold it; I carved and trimmed it with a knife, chisels and a gouge. I took the bridge design from a Gerle lute, discussed in Robert Lundberg"s book on lute construction. The recess at the back of the bridge was marked with a knife and finished with a chisel.
Carving bridge end detail
Marking where to remove wood
Chiseling area marked out
Here I am gluing on the bridge; I neglected on this occasion to use my straight edge to hold it in position, and it is now glued on very firmly, but slightly out of true. I hope the reader will appreciate my candour in reporting all my little mistakes, experience is cheap at any price . . .
Clamping bridge in place
I made a little thumb plane, based on a Dutch design, to trim the bars on the back of the soundboard, having first chiselled the bars down roughly. Chisels are used to cut down the angled ends of the bars.
Using thumb plane
Yet another mishap was that I left the belly in the airing cupboard for rather too long, it became too dry, and the joints started to open—it needed restoration even before it was finished! So I wet the soundboard, around but not on the openings, with hot water and a brush to make the wood expand again, closing the cracks, applying some hot glue and hand-made paper with a warm gluing iron to seal the cracks closed.
Now it was time to start thinking about assembling soundboard, body and neck. I had to finalise the height of the ribs and endblock, using a chisel, and used pieces of wood wedged in place to help prevent the delicate body from flexing too much while I was levelling off the rib edges and endblock.
Now it was time to further finish the neck. I marked the height I wanted with a scribing point, and then trimmed the neck with an axe, not too close to the scribed marks!
Gently does it!
Axes were much more widely used before the days of bandsaws and portable power tools, being used for instance by joiners to make the bevels of a hipped roof as late as the 1960s, as an old joiner told me. Then I used a coarsely-set roughing plane to get the rough shape, then a chisel, then a plane, to create the curves of the sides of the neck, against a bench stop. It saves a great deal of time planing against a bench stop, because one does not have to keep winding and unwinding the vice screw.
Tapping the nail in place, having glued on the neck.
Now I was starting to think about the fingerboard. According to Robert Lundgren lilac was sometimes used for making pegs, and I had a piece of lilac, though curiously I seem to be slightly allergic to the dust of lilac wood. It is an attractive wood, hard, with faint reddish purple stripes, reminiscent of tulip wood, though the colour seems to fade on oxidation. It may not have come to England until the 16th century, but it is a local wood found in many in gardens where I live. I sawed off a strip of wood to use as a finger board; discussed further below.
I planed the surface of neck to end block joint against a bench stop to get the it really flat prior to gluing on the soundboard. One more minor mishap was that the body seemed to distort slightly once I had taken it off the mould and was planing the edges of the ribs down (I think I planed a slight hollw in the two bottom ribs when I was attemting to plane them straight) , with the consequence that when I glued on the soundboard I found it slightly dished. Perhaps I could have pulled it back into shape, but it is easy to panic when using the hot glue, paper and iron method of jointing! I held the soundboard in position with a bradawl in the neck while gluing. I had trimmed the soundboard almost to its final shape many months earlier, so did not have much leeway to allow for disortion of the body, which I now regretted. I started gluing at the bottom block. This was the stressful part; you have to keep glue and iron hot, while fixing pieces of paper and checking the fit of the join. I singed myself a few times through using pieces of paper that were too small; it is best to cut them quite long and wide, so you have an end to pull on and can keep your fingers away from the hot gluing iron.
Reattaching soundboard with strips of paper and hot iron
Clamps for good measure
I left it overnight, and found in the morning that the glue lines were too wide, so I tried to rectify it with a hot iron, similarly to what is done when veneering using hide glue, but this softened the glue too much, opening joints, and causing a slight crack in the soundboard. I closed the soundboard crack with hot water and rubbed glue into it before it closed and reglued the joint using more paper strips and glue, causing a bit of scorching, fortunately not too visible. A real process of trial and error. The paper strips dry very quickly, but tear easily if too wet; one realises how good the old luthiers were at managing fire, water and iron.
Using hot iron
I cut off the belly extension over the neck in the proportions of the golden section for no other reason than it seemed to me to be in the spirit of things and looked right, the surplus wood coming off easily as I had avoided apllying glue to this area.
Removing excess wood
Marking out the fingerboard with a scribing point around the outline of the neck
I sawed out the fingerboard (holding it place with my foot, I suspect these cavalier methods would have been used in olden times for speed of working) leaving a good allowance all round.
I then planed it to the correct thickness, holding it once again with bench stops and wooden clamps. I scraped the fingerboard veneer and the neck to get them as flat as possible, and then glued the veneer on.
Gluing and clamping newly cut fingerboard
Now I returned to work on the pegbox. I took a spare scrap of rib wood to make a template for the peg holes to be made in the sides of the pegbox, spacing the nine holes evenly along a scribed line.
Marking the peg box
I tacked in on to the pegbox, and pricked the peg positions through onto the sides of the pegbox with an awl, taking particular care to position the first pegholes correctly. When drilling the pegholes with a brace and bit I used a wedge of wood up against the edge of the pegbox to make sure the holes were horizontal. This is quite hard because the pegbox is tapered, and the holes I ended up with are not perfectly true or in line, as you can see in one of the pictures below. Before putting on the pegbox I wanted to finish the neck. Using a straight edge against the neck casts a shadow which can give a quick visual aid to when judging if the profile is even.
Finishing the joint between neck and body; a chisel has to be used in the corners where the plane cannot reach.
Marking out the end of the fingerboard with a scriber
Chiseling the neck to length
Sawing the neck to length.
Using compasses to match the width of the neck to the width of the pegbox where the two parts join.
Having decided on the angle of the pegbox, I drill the hole for the nail which will fix the pegbox to the neck. I did this before cutting the pegbox to length, to reduce the danger of splitting the end of the pegbox. Sadly I had no wrought iron nail for this, only an old cut nail. I used a larger drill to counterbore the nail hole,I ended up using several drill bits to drill the hole to an approximate taper to match the nail. The end of the nail will finally be covered up; I believe an ebony dot was sometimes used for this, on some later 16th century lutes the nail was concealed by the bottom block of the pegbox, which was glued on afterwards.
Drilling peg box joint hole
I filed the nail square, and flattened the top, to make it resemble more closely a wrought iron nail.
Sawing the pegbox to length.
Sawing the neck part of the pegbox joint to shape, holding the lute body and neck assembly in place with wedges.
Here is the finished joint before assembly.
Now I made a reamer to make the peg holes tapered to fit the pegs that I would make later. I tried using a reamer triangular in section, but this did not work very well, so I filed hexagonal and found it worked much better I tried hardening the reamer but could not get sufficient heat on my mother's gas cooker; at least it ended up harder than mild steel. I then sharpened the edges on a sharpening stone. I do have a grindstone, but I find a long flat piece of sandstone is useful for rough sharpening of tools and is in keeping with my 'simple as possible' approach. It is very important to drill the hole correctly so that the wood does not split when the nail is driven home.
Using the bit of my bowdrill to finish of the nail hole.
Then I glued and nailed the pegbox to the neck.
Now for the pegs. I believe small bow lathe or pole lathe may have been used, (Rather interestingly given that period instrument strings are made from gut, I have read that twisted gut was used for pole lathe strings), but unfortunately I did not have time to make a simple lathe, so decided to make the pegs by hand. I starting using elderberry wood (I also used this for the nut); I really like this wood and would recommend it, it is close grained and can have an attractive yellow colour, but did not have enough of it, so used lilac thereafter, as I had for the fingerboard. I split out the pegs blanks with an axe and club then roughly shaped them with the same small axe. Then the peg blanks, held at times in a specially made jig, were finished with chisel, scraper, and knife.
Series of pictures of the process of hand making pegs:
Using a bowdrill to make a string hole in a peg, which I am holding once again in my pegmaking jig.
Running a piece of string from the pegbox to the bridge to test how the strings would lie, I marked the positions of the outer strings on the bridge with a bradawl, and then used dividers to make the positions of the other string holes, copying over the measurements from a rough template.
Marking string positions on bridge
I used a long bow drill, (it needed to be long to allow it keep the bow clear of the lute body when drilling), to make the string holes in the bridge
Drilling holes in bridge
Marking a piece of wood (roughed out as usual with axe, and then planed against a bench stop) which will serve as the nut.
My lute was just about complete. It only remained to think about the finish, fret it and string it.
I use the a file to round and smooth all sharp corners on the lute. After scraping, for final smoothing horse tails or "shave grass" were often used in earlier times (they were also used for scouring pewter). There are many varnish recipes surviving from the Middle Ages. There is a theory that a sort of egg tempera was used to give a "blond" finish to the soundboard. I believe that soap would have been used to make pegs turn more easily as it had been around for a long time in the fifteenth century.
By way of an epilogue, here is a very bad limerick relating to the fact that workmen often gave animal names to their work-holding tools (sawing horse, bench dog chairmakers donkey, etc) and worked without screw vices, causing me wonder how earlier luthiers had clamped and held their workpieces, which is why I entitled this project as being in search of the lutemaker"s donkey.
It may come to you as no surprises
That there was an old lutemaker who possessed no vices.
When things went wonky
He reached for his donkey
As it was one of his favourite devices.
What I should do now is to make another lute, which would surely turn out better than this first prototype, to consolidate and reinforce what I have learned about the old methods . . .
Andrew Atkinson concluded this series of lectures at the Lute Society by playing a piece on his newly strung lute, using a plectrum; it had a lively and strong sound.
Edited for the web by James Jackson and David van Edwards
In commemoration of the hard work of Andrew and the website team, this galliard was written by James Jackson for renaissance lute.
Besides the works cited in the text above, the following are of interest:
Edlin, H., Woodland Crafts in Britain (Newton Abbott: David and Charles, 1949)
Goodman, W.L., British Planemakers since 1700 (Needham Market: Roy Arnold, 3rd edition, 1993)
Harwood, I, "A fifteenth-century lute design", The Lute Society Journal ii (1960) pp. 3–8
Jenkins, J. G., Traditional Country Craftsmen (London, Routledge, Kegan and Paul, 1965)
Knight, R., "A Carpenter"s Rule from the Mary Rose" Tools and Trades, vi, pp. 43–55
Kylsberg, B., "A Seventeenth-Century Saw at Skolklosters Castle, Sweden" Newlsetter of the Tools and Trades History Society, 31, pp. 27–33.
Salaman, R.A., A Dictionary of Woodworking Tools, c.1700–1970 (Taunton Press, 1990)