Choosing the model


When the news came that the Garbrieli Consort and Players were able to provide the financial support we had to move fast since a recording date was already fixed. I had been discussing, for some time with violinist Oliver Webber, the possibility of making reconstructions of instruments explicitly suitable for performing Monteverde for the new "Monteverde String Band". The first step was choosing the right model and finding an original instrument as the basis of the reconstruction. Monteverde himself had written that he preferred Cremonese violins which at that time could mean older instruments by Andrea Amati or more likely his sons, the brothers Antonio and Girolamo. A further consideration was that the instruments made by Andrea were probably originally set up in a very different way. The earlier ones were probably viole da braccio and had only 3 strings. If any of these instruments were used as violins by musicians working for Monteverde they would have been modified to have a set-up like the brothers were doing at that time. Therefore it seemed a much better choice to look for a brothers Amati violin from around 1610 or a bit earlier.


By a bit of extraordinary luck there happened to be a 1595 example a mere 15mins from my workshop at the Royal Northern College of Music. Another strong contender was the 1629 at the Royal Academy of Music. There are some significant differences in the models and this one is thought to be possible made or part made by the young Nicolo Amati. It also may be the first or a very early example of a model that they used from then on. The National Music Museum, The University of South Dakota has a smaller Brothers Amati violin of 1604, which has a great many features in common with the 1595. We felt it a reasonable assumption that Brothers Amati violins were more like these two than the 1629 one. Despite the beauty of the 1629 we felt that the 1595 was the best choice for the intended usage.


The 1595 has a similar corpus length but is a little narrower than the 1629 and has different archings and graduations. The style of the f holes is a good illustration of the differences.


Brothers Amati 1595 Brothers Amati 1604
Brothers Amati 1629 Hieronymus II 1695

Taking measurements


The Royal Northern College of Music kindly arranged for me to have access to their 1595 instrument. I had two sessions when I photographed this violin and took measurements. The photos were not taken with a view to a publication in the Strad but were to give me useful info and keep the appearance fresh in my mind. The measurements included outlines of back and front, wood thicknesses, arching curves, rib heights, scroll dimensions and f holes. I also examined the internal work for clues as to which blocks and linings might be original. I was also able to take similar measurements of a Hiernonymus II violin belonging to the Halle Orchestra. Though out of the date range for this project it was very helpful in focusing my mind on the ways in which the earlier violins were different. The Royal Academy of Music also generously provided access to their precious 1629 instrument and no doubt the time will come to make a copy of this one.


Deriving working drawings


When making a copy of an old instrument it is not enough to simply attempt to duplicate the dimensions and shape as it is now. Account must be taken of distortion, particularly of the archings of the back and front, due to wood shrinkage and the forces exerted by the string tension over the centuries. The measurements have to be interpreted in a way that points to how the instrument might have been originally. There is no scientific method for doing this that is any more reliable than informed guesswork based on observations of distortion in instruments of different ages.


It is generally believed that the Amatis used an inside mold to form the rib garland. Though it seems likely that they had a geometric system of constructing the outline or the mold it appears that they did not adhere to it with rigidity. There are therefore asymmetries and variations from one instrument to another that were made using the same mold. The process involved regenerating that mold and building the instrument with similar deviations. The intention is not an exact replica of the dimensions of the original but while keeping very close to the original some natural deviation commensurate with the working methods of the time is allowed. If this is not done then the copies will look too geometrically perfect and therefore not in the style of the Brothers Amati or any other early Italian maker.


The original is now set up as a modern violin and we turned to various sources to decide what sort of bass bar, neck, fingerboard, tailpiece, bridge and stringing would be appropriate. One of the pleasures of this project was working with Oliver Webber on resolving issues about set-up. He has a very serious and scholarly approach to the history of music, instruments and stringing. We were able to sift through all the information we could lay hands on and discuss its implications. We don’t think that we have made any serious mistakes, but this area of research is dynamic – new pieces of evidence are continually emerging.


Fortunately there is at least one surviving Amati bass bar which is very small by modern standards. There is a lot of information about Stradivari violins and their fittings which, though much later, can be used to corroborate other information. A major source is old paintings which are sometimes very detailed and almost photographic. Again fortunately there are a great many to look at and it is possible to form a reasonably precise idea about the necks, fingerboards, tailpieces and bridges.


For example this is Bartolomeo Cavarozzi. He worked in Rome but since the violin making industry had not developed there at that time this may well have been a Cremonese instrument.

Doing the woodwork


The rib are bent from flat strips of maple. The end and corner blocks are lightly glued to the mold. The ribs are bent to shape on a hot iron and glued to the blocks starting with the waist. When the ribs are all in place the linings, made of willow like the blocks, are fitted. The original had an unusual way of fitting the linings against the blocks. They are notched into the block from the bout side rather than set into a channel from the waist side.

The back and front were carved from chunks of maple and spruce respectively. They were roughed out first and allowed to settle before flattening the underside and completing the shaping. The bass bar is fitted after the f holes have been cut and the graduation complete. A little extra wood is left round the outline which is only finalised after the belly has been glued to the ribs.

The neck and scroll are carved in stages. First the profile is sawn and smoothed. Next saw kerfs are made to facilitate removing wood from the peg-box cheeks and the volutes. When this is complete the flutes are sunk round the back of the peg-box and round the head and the peg-box can be hollowed out.

When the neck and scroll are finished the ribs are removed from the mold and the neck glued and nailed in place after which the back can be glued to the ribs also with an allowance to trimmed later.

The neck was not shaped yet to assist in holding a vice. The belly was then glued on and the purfling (made of pear wood dyed black and poplar for the white strip) was inlaid. The fingerboard was made and fitted and then it was strung up to see if there were any problems.

The neck was not shaped yet to assist in holding a vice. The belly was then glued on and the purfling (made of pear wood dyed black and poplar for the white strip) was inlaid. The fingerboard was made and fitted and then it was strung up to see if there were any problems.

All was OK … the edges were shaped and the neck smoothed. Varnishing was the next step. A ground layer made up of very fine mineral particles in a pine resin and linseed oil varnish was applied followed by a few very thin coats of the same varnish but with added resin dyed with madder. The temporary pegs were replaced with the final ones and adjustments to bridge and soundpost made the violins ready to show to Oliver and Catherine. Needless to say: the strings were “Real Guts”. Equal tension was the only option which put the g string at a diameter just over 2mm. Before I delivered them I went back to the RNCM to see how the copies looked and sounded compared to the original.

The legendry yellowish colour of the Amatis can be very difficult to emulate and I had been concerned that I had not managed to get the colour anything like the original. When laid side by side the colour was much nearer than I had thought. Also, in spite of the radically different set-ups my copies clearly had some tonal features in common with the original which, in spite of being over 400 years old, has a brilliant immediate sound. There is much myth about Amati tone.. they are often said to be gentle in tone and only suitable for chamber music. This is absolutely not true. I believe that they were originally very powerful with depth on the g string and a piercing but clear and beautiful e string, the middle range having interesting colours.


I took the new violins to London for a road test session with Oliver and Catherine. After some minor adjustments we all felt that they were ready for the first concert. I was able to go to this concert : the Monteverde Vespers with the Gabrieli Consort and Players at Christ Church, Spitalfields. Of course, all the efforts towards historical accuracy in the instruments would count for nothing without musicians of a high calibre and a serious commitment to understanding the performance practices of the time. I was not disappointed; Paul McCreesh has devoted his career to this end and has sought to work with musicians who share his passion. I find no work more satisfying than this type of collaboration.

Back to instruments page