William Castle - Violin Maker

© William Castle 2014



A better solution is the scrub plane. These planes have a flat sole but a rounded iron and were used in pre power tool days for the initial planing of timber after it had been sawn. For violin making, however, a sole that is rounded in both length and width is more useful, and absolutely necessary for hollowing. The primary advantage of using a plane is that the amount of wood removed is fixed by the set of the iron, so you can work much more quickly, whilst matching the amount of force needed to the resistance of the plane. This saving of energy, combined with an upright stance and free flowing movements means the procedure is much less tiring than using a gouge. As to its influence on the instrument, because the arching develops as a whole, rather than in a series of gouged hollows, it is easier to see the shapes, and because the scrub plane is used along the grain it tends to encourage arching that flows along the instrument. Equally, when hollowing, as long as the longitudinal curve of the sole is similar to the inside shape of the instrument, developing the hollow and keeping the thickness fairly constant comes much more naturally than when using a gouge.


The use of the mould

The last piece of equipment I want to consider is the mould and how it is used. In regard to the speed of construction, the mould has numerous advantages over making ribs on a board. When preparing the blocks there is less need to check the direction of the grain because you can mark the curves on both ends of the blocks and cut from either end. Bending the ribs is much quicker because as long as there is a greater amount of bend on the tighter curves than the shallower curves, the ribs do not need to be bent to the exact shape of the mould because the ribs are pulled tight against the mould as the ribs are glued to the blocks. By contrast when working on a board there is nothing to keep the ribs in shape so they need to be precisely bent, which might require repeated adjustments on the bending iron. When fitting linings, with the ribs on the mould both sets can be fitted before removing the rib structure which allows the linings to be cut back whilst still on the mould, a method confirmed by the knife marks on Stradivari’s moulds. Because the ribs are held securely against the mould, thicker shavings can be taken off the linings, often in a complete stroke from one end of the lining to other, without fear of twisting and breaking the ribs or cutting oneself with the knife.

Whether the classical Italian violin makers made their designs using the mould or the final outline as their starting point, that they generated the outline once the ribs were finished seems irrefutable, as the overhang between plates and ribs is usually fairly consistent. Looking at it the other way, if they had first cut out the back and belly and then made the ribs, some instruments, particularly cellos, would have a negative overhang! Although we know that it is quite possible to produce ribs which accurately follow the mould, the ribs and outlines on classical Italian instruments tend to be asymmetrical and deviate from the mould. In all probability this is partly due to the procedure the Italian violinmakers adopted to ensure that the neck of the instrument was in line with the centre line of the body, as put forward by Roger Hargrave. [3] This method utilises the pins which go through the back and into the top or bottom block as pivots around which the neck and ribs are rotated until the neck is aligned with the centre line of the instrument’s body, and in that process the ribs are distorted. Although this does explain some of the eccentricity, particularly when one c bout is high than the other, it does not satisfactorily explain the different curves sometimes seen on opposite c bouts, or the difference between the outlines of belly and back.


To return to index page of Videos and Articles - click here

The spirit of classical Italian violin making  -  Part 2

first published by the British Violin Makers Association



To return to index page of Videos and Articles - click here


scrub plane resting on the inside of a viola back

In the first part of this article I explored how the instruments made in Italy during the two hundred years from 1550 were informed by the cultural ideas which were part of the Renaissance, but how their realisation grew out of the practicalities of working in wood in a traditional craft-based society. As part of this practical working ethos, the need to work effectively and efficiently influenced both the techniques used by the classical Italian violin makers and the appearance of the finished instruments. This contrasts with our modern interpretation of quality, and the early training many of us received which often leads to rather effete working habits where we cut away from the line and then tickle away with small tools until the line is met. The evidence of classical Italian instruments, however, shows the makers were forthright in their use of tools. Whether they worked very precisely or with greater freedom, their instruments are always workmanlike, a term which encompasses the idea that the work was of good or acceptable quality carried out in a straightforward and no nonsense manner.


A workmanlike approach

A simple example of these differing approaches is the creation of the thickness of the edgework, which is usually a little thicker at the corners than elsewhere. The method in common usage when I was training was to set a marking gauge at 4mm for the majority of the violin’s outline but ½ mm more at the corners, and blend the two as the line moved away from the corners. The fluting was then brought up to a line which would define the high point of the edge, a millimetre or so away from the edge of the plate, and then the edgework was rounded back to the same line. Although this is not particularly difficult, neither is it very workmanlike, in contrast with making the edge thickness 4.5mm thick throughout, and then using the position of gouge when doing the fluting to define the edge thickness. With this method the marking gauge only needs to be set once, and the purfling platform does not need to be made as accurately or as smoothly because with the exception of the corners, none of this surface will remain on the finished instrument. When doing the fluting the gouge simply cuts to the very edge of the plate, except at the corners where it is gradually brought in to where you otherwise would have marked the high point of the edgework. Then as the edgework is rounded off, the thickness of the edge is reduced along the majority of the outline, but naturally swells towards the corners. The desired result is thus produced not just more quickly but also with more grace, because you are not working to an arbitrary line, but intersecting the existing defined curve of the fluting as the edge is rounded.


Purfling cutters

With the cutting of the purfling channel, we have the purfling cutters used by Stradivari to indicate the method he used. In contrast to using a purfling marker to simultaneously mark both inside and outside lines and then using a knife to cut the channel, the single line cutters of Stradivari have many advantages. Instead of having to pay very close attention as when using a knife to ensure that it is running true, when using the purfling cutter the pressure against the edge of the back or belly ensures it cuts in the right place, and because it is held in front of you, you can see that it remains upright so the cut also remains vertical. Because the marker is held with your whole hand, the muscles of the whole hand are used in transferring the weight and power of your arm to the work, as opposed to using a knife when all the pressure is applied through one or two fingers. Defining the depth of the cut when using a knife is also somewhat hit and miss, usually requiring some more work with the knife after the wood has been removed from the channel, but when using a cutter the depth stop automatically regulates the depth of the channel.  The thickness of the cutter blade also means the wood is compressed after it has been cut, so the final channel is wider at the top than at the bottom. This makes it easier to put in the purfling, especially if it is inserted in separate strips, and when the glue is applied the wood swells back to its original position, tight against the purfling.


Hollowing the back and belly

One of the most inefficient techniques we indulge in is the use of gouges to arch and hollow the back and belly. The trouble with using a gouge is that the thickness of the wood removed with each cut is dependent on the angle of attack, so we have to develop a technique which prevents the gouge from running too shallowly and coming out of the wood, or going too deeply and bringing the gouge to a halt. A common method is to bend over the plate with a two handed gouge, and tensioning the shoulders and arms for control, slowly remove parallel strips of wood. The result is thus achieved painfully slowly, with great concentration, stiff shoulders, a hurting back and a general feeling that there must be a better way.


Thus the way a mould is used demonstrates the difference in approach between the classical Italian violin makers and the customary approach of our own time. Although it may seem an improvement to use a clamp rather than pieces of string, the changes imposed by the use of a clamp fundamentality alters the aesthetics of an instrument; the rib shape influencing the outline and the corner shapes, which then affects the purfling and the alignment of the f-holes.


Gaps in the evidence

In trying to reproduce old working techniques, there is bound to be some guesswork because of the gaps in the evidence. The weakest case is for the use of scrub planes, because if there are tool marks on the inside of instruments near the top or bottom blocks, these could equally well have been made by a large thumb plane or a gouge. But by the same token, the evidence for using a two handed gouge is just as weak. With any technique there may also have been differences between individual makers. The method of regulating the edgework, for instance, would need to be modified in order to produce the thinner corners of G B Guadagnini, and it won’t work at all on instruments such as those of Pietro Guarneri of Venice, Belosio or on some del Gesus, where the fluting outside the purfling is so steep that the curve would never reach the outside edge.

Nonetheless, no matter how individuals varied in their techniques, the utilisation of straightforward procedures was essential to working quickly and efficiently. With constant repetition this became intuitive, giving them the freedom to work with boldness and fluency, which in turn allowed them to effortlessly imprint their intention on the instrument.


1 -  Roger Hargrave, from ‘Guisseppe Guarneri del Gesu’ by Carlo Chiesa, Roger Hargrave et al. 1998


These variations can be explained by the way the mould was used, specifically in the method of clamping the ribs to the blocks. This is most easily seen by looking at how the c ribs were clamped. The way I was taught was to use a piece wood planed at an angle at both ends, with which to push the ends of the ribs towards the mould as the clamp was tightened. Although it is a peculiar set up to have a clamp working at right angles to the surface being glued, this method works reasonably well as long as the rib ends and the surface of the blocks are square, and that the curve of the block continues the curve of the mould. The advantage of this method, if you think it is important, is that the rib is pushed hard against the mould, so reproducing the intended outline. As long as the edge of the rib is pushed down onto the bench, the rib remains square so the outlines of back and belly are the same. Neither of these features, however, are typical of the old Italian instruments, whose blocks are often not square and the top edge of the rib takes a different route between blocks as the bottom edge.

However, using the method of using clamping blocks and string, indicated by Stradivari’s moulds and tools, solves all these problems. As far as the clamping function is concerned, it is no longer necessary to cut the blocks square because the string allows the clamping block to work at an angle, and if the curve of the block is not quite continuous with the mould, the rib is still held tight against the block. In comparison with the modern method, the rib is not pushed tightly into the mould; that depends on keeping pressure on the rib and clamping block with one hand whilst wrapping the string round the counter-block and peg with the other hand. When gluing the rib it is possible for the rib to slip so it is no longer parallel to the mould. If this occurs, when the other end of the rib is glued to the other corner block a twist can develop in the rib, so even when blocks are cut square and accurately to the corner shapes, there still can be a difference between the shapes of the top and bottom edges of the rib, as one side is pulled away from the mould.


outlines of c ribs when clamped to blocks using string

Top and back outlines of the c ribs showing the discrepancy which can be created when clamping using string. The  blocks on the bass side [left] were cut square but the rib was glued as described for the previous picture. On the treble side the blocks were cut ½ mm out of square.

A small scrub plane, 135 mm long

Glueing the rib to the corner blocks by wrapping string round the counter-block and peg.